aiLL 8/10/23 - Surviving the AI Revolution While Getting Stuff Done

Description

Chapters

Transcript

0:01 foreign
0:06 [Music]
0:23 welcome to the lab the AI learning lab
0:26 what's going down with shaking What's
0:28 Happening
0:31 welcome welcome welcome
0:34 all right let's get this party started
0:37 we're gonna get this water started
0:43 all right
0:46 let people get rolling on in here
0:50 I think I got to do something about
0:52 these eyebrows starting to get all Santa
0:54 clausy
0:59 I can't have Santa 's eyebrows
1:01 it's not good it's not good for anyone
1:07 what do you do for Dallas I am a founder
1:12 and CEO of a company called storyvine
1:15 and uh we've been around for 11 years we
1:17 do a lot of work in healthcare and
1:19 Pharma it's an automated uh video
1:21 storytelling platform and uh I do this
1:25 here uh AI stuff for for Fun and Profit
1:28 you can make money with jbt
1:31 not a lot of profit right now but
1:33 spending a lot of time on it
1:35 um rolling a lot of AI into my current
1:38 company the current product Vision
1:39 things like that so that's that's the
1:41 day job and then at night time I'll do
1:44 do these things
1:45 do a little bit of sim racing things
1:47 like that
1:49 um yeah this is the the AI learning lab
1:50 and uh
1:52 I'll talk about these things here in a
1:54 in a second we'll let some people roll
1:56 in here
1:57 Rick Olson how are you sir Jeff Clark
2:00 welcome
2:02 welcome
2:12 breathing water not smart
2:16 how did he go I waterboarded himself
2:19 [Laughter]
2:23 do you believe in palantir well
2:28 I believe that
2:33 they're strapping guns to large language
2:35 models and that's [ __ ] terrifying
2:39 do I think they'll be effective Maybe
2:44 we did a piece on palantir in uh
2:46 everyday AI in the newsletter I did for
2:49 a while and uh yeah it's interesting
2:52 stuff it's interesting stuff what
2:54 they're doing like it's potentially
2:55 beneficial to Warfare but
2:58 I don't know these large language models
3:01 and weaponry right now feel like they
3:04 shouldn't be anywhere near each other
3:05 because if I can't get this thing to
3:06 write a [ __ ] email predictably I'm
3:09 not sure I want it targeting the enemy
3:12 just yet
3:13 why don't we give it a year or two and
3:15 let it let it settle out and figure out
3:18 if it actually works or not anyway
3:19 welcome everybody my name is Kyle
3:21 Shannon this AI learning lab a little
3:23 bit about me I'm an entrepreneur many
3:26 time entrepreneur started I don't know
3:28 12 15 companies something like that
3:29 number of which should have never been
3:31 started number of which have been really
3:33 successful and lots of stuff in between
3:37 um
3:38 I was an entrepreneur back in the early
3:39 days of the World Wide Web so I have
3:42 seen I've been in one of these things
3:44 before and one of these things meaning
3:47 uh one of these
3:50 technology changes that's gonna
3:52 transform things and uh it's wild
3:57 it's wild I can tell you that from from
3:59 the from the early days of the web kind
4:01 of writing writing the web rocket up and
4:05 being part of the.com bubble and
4:07 then.com burst
4:09 and all of that sort of [ __ ] it's pretty
4:12 intense so I didn't think I'd see
4:14 another one in my lifetime but I have
4:17 and its name is Chad gbt generative AI
4:22 um and yeah I think we're in a pretty
4:24 amazing thing so anyway I love that
4:25 shirt thank you very much appreciate
4:27 that crypto Joe
4:29 um all right let me let me talk about
4:31 what I got going on here what are these
4:33 things so if you're if you're brand new
4:35 to
4:36 generative AI if you're like all these
4:38 people keep talking about Ai and I don't
4:40 know what the hell they're talking about
4:41 I don't like it I just don't like it one
4:43 bit
4:43 if you're one of those people and
4:45 haven't tried this my strong strong
4:47 strong recommendation is
4:49 go to this top URL chat.openai.com
4:52 that's the official chat GPT website if
4:54 you've never been there before go there
4:56 right now
4:57 fire up your laptop or just leave me and
5:01 go there
5:03 um and start playing with that now all
5:05 these other things down below it are
5:07 also they're either variations of chat
5:09 GPT or competitors of chat gbt so if you
5:12 have played with this and you haven't
5:13 explored these things below
5:15 these are the seven right now that I
5:18 think are worth you at least
5:19 understanding what they do now there's
5:22 there's
5:23 they're roughly in order of what I think
5:26 is you know most to least useful
5:30 um but the bottom two are useful in kind
5:32 of different ways the bottom two are
5:33 kind of more think of the bottom two
5:35 more as kind of a bit more consumery pie
5:38 is very chatty Poe you you have Bots
5:41 there you can play with a lot of
5:42 different models um
5:44 the top five are are the real players so
5:48 you've got chat GPT you got Bing which
5:50 is Microsoft's version of chat GPT you
5:53 got Claude which is a competitor of chat
5:55 gbt has a really big context window so
5:58 you can you can put really big documents
6:01 in Claude and that's amazing then you
6:04 got Bard which is not embarrassingly bad
6:06 anymore and it can read images and so
6:09 it's worth understanding what it's good
6:11 at
6:12 um you got lab stop perplexity that's uh
6:16 perplexity hosting
6:18 um metas llama 2 models llama 7B 13B and
6:22 70b
6:24 um so that's worth playing with so all
6:25 these are worth playing with but if
6:26 you're new if you're brand new to this
6:28 [ __ ] and you're like but Kyle I don't
6:30 know what to do I don't know where to
6:31 start oops I just changed my I just
6:35 changed the chart into a one
6:37 um go to that top one
6:38 the top top one
6:40 chat.openai.com that's gonna be the one
6:43 you're gonna look at you're gonna look
6:45 at that one if you if you're down in the
6:47 Bible you're trying to figure out what
6:49 we're doing with this yo chat Jupiter
6:50 and you're gonna go to that one you're
6:52 gonna go to chat.openai.com
6:55 take it right out in the swamp with you
6:57 all right
7:00 now
7:02 if you're like but Kyle
7:04 I don't know why you'd say it like that
7:06 but if you do
7:08 you're like I don't know what to do then
7:10 I got I got that for you too so
7:12 prompts.chat that'll teach you how to
7:14 prompt these things it's like a little
7:16 primer
7:17 um
7:18 that that uh will tell you what the
7:21 large language models are how they work
7:23 how to talk to them and it's a whole big
7:24 long list of prompts that you can
7:26 literally copy and paste in the chat
7:28 will give you lots of ideas uh this this
7:31 other thing here uh prompt book is for
7:33 images and down here the three little
7:35 image things I would recommend playing
7:37 with mid journey is the best fireflies
7:39 the most commercially safe and Leonardo
7:42 is a really nice web front end sitting
7:45 on top of stable diffusion which stable
7:47 diffusion is good it's catching up to
7:49 Mid Journey but it's not quite there yet
7:51 all right so that's what those things
7:52 are
7:54 all right
7:55 so if you've got questions about AI pop
7:59 them in the comments below do my best to
8:01 answer them uh
8:03 I try to answer them all that causes
8:05 issues because
8:08 I don't get I just it causes issues
8:11 because I'm answering all the questions
8:13 from the past and you're down in the
8:14 present going why the [ __ ] isn't the
8:16 answer in my questions oh and I swear a
8:17 little bit too so if you don't like that
8:19 I'd get out get out get out he's gonna
8:23 be swearing he's gonna do it he'll do it
8:25 he'll do it right up he'll swear right
8:26 up he'll do it I've seen him do it I
8:29 come to these lives every night and
8:31 sometimes he's just nice as a church
8:33 church mouse he's just super nice and
8:35 then other times he's just got a filth
8:37 mouth
8:38 he's just got a filth mouth and then so
8:41 I'll tell you what sometimes he's so bad
8:43 they'll kick him right off the tick tock
8:46 they kick him right off the tick tock
8:48 they'll do it too I've seen it I've seen
8:51 him mid-sentence just Boop he's gone we
8:54 don't see him for weeks
9:01 all right
9:03 all right I don't know what I'm gonna
9:04 say tonight you know what I'm gonna do
9:06 I'm gonna answer some questions there's
9:08 Emilio's wife what's happening 8 14 in
9:10 Colorado it is
9:12 hello the bear
9:14 how are you
9:16 Lori 711 let's shake him
9:20 all right let me go grab a question or
9:23 two here and oh it's gonna be hard to
9:25 read on this shirt sorry about sorry in
9:27 advance I might need to go change the
9:28 shirt out hi Kyle I told a bunch of
9:31 people about you in a librarian
9:34 webinar yesterday so you might have some
9:37 Librarians I hope they don't mind
9:39 cussing
9:41 oh well well he's very I
9:44 he's very boisterous isn't he certainly
9:48 seems enthusiastic
9:49 all right and listen I'm not saying that
9:52 Librarians that's how they all are
9:54 definitely not just the ones that yelled
9:56 at me when I was in the library
9:59 Mr Shannon Mr Shannon
10:03 Mr that is completely inappropriate
10:07 we do not toss peanut butter at the
10:10 other patrons we don't do that here Mr
10:13 Shannon thank you thank you
10:17 all right beautiful
10:20 so you've got a video portal
10:24 I have a video technology company so
10:27 it's a uh storyvine is a platform that
10:31 we have an app that guides an end user
10:33 through telling their story so we do a
10:35 lot of work in healthcare and Pharma so
10:36 for example if um
10:39 if Pfizer wants you know one of their
10:42 patients for a certain drug to share
10:45 their diagnosis Journey our app will
10:48 guide that patient you know introduce
10:50 yourself here talk about what it felt
10:52 like when you were diagnosed it walks
10:53 them through the process those clips go
10:56 up to the cloud five minutes later you
10:57 got a fully edited video
10:59 and then I'll hold back in management
11:00 system to deal with all that
11:03 so that's what Story vines all about
11:05 Rick Olson fantastic thank you sir do I
11:08 believe in palantir answered that one I
11:10 believe in it I believe it's [ __ ]
11:12 scary but you know
11:14 I was so I had a choice to make in life
11:16 when I was a kid my my dad was a pilot
11:18 in the Air Force and I thought I think I
11:20 might want to be a pilot and then at
11:22 some point it was like do I want to be a
11:24 pilot or do I want to be an actor I went
11:27 the uh I went the actor route
11:29 hang on one second
12:09 all righty all right all right all right
12:12 all right all right
12:15 retail consultant by day
12:17 oh that's nice
12:18 retail retail is going to be really
12:21 interesting with all this AI stuff all
12:23 right hi what's the best for help with
12:25 python code I don't know if you I don't
12:27 know what you mean with help for python
12:30 code if you mean
12:32 um
12:34 You Want To Learn Python coding that's
12:37 different than if you want to do python
12:38 coding for you which is different than
12:41 if you want
12:42 to kind of pair program with AI if you
12:46 want to pair a program if you're like uh
12:48 you know putting projects together I
12:50 would say uh github's Copilot seems to
12:54 be
12:55 um seems to be popular amongst the kids
12:58 and uh
13:00 if you just want to use
13:03 um
13:03 this stuff to write python code for you
13:06 chat GPT is really good I hear that
13:08 Bard's good but I would say chat gpt4 is
13:12 of all of these things gpt4 is still the
13:15 best I'm holding stock with them who are
13:17 you holding stock with
13:18 oh palantir oh uh yeah I would
13:24 then then as an investment yeah I I'd
13:28 say they're it's probably worth holding
13:30 stock in palantir like whoever whoever
13:32 sort of figures out the [ __ ] they're
13:33 trying to do is not bad
13:35 I I mean it's not it's not it's not bad
13:38 or good uh whoever wins that that race
13:41 is gonna be they're gonna do good
13:43 financially I would think uh this will
13:46 definitely be more transformative than
13:48 the worldwide but yeah exactly the World
13:49 Wide Web super transformative uh but if
13:52 you think about what the technology was
13:53 it was actually quite simple right a
13:56 page with hyperlinks in it that connect
13:57 to other pages
13:59 that transformed a lot of things right
14:01 it made Communications immediate it it
14:04 you know reduced the need for travel
14:07 um it's the foundation of all the stuff
14:08 we do today it's the foundation of a lot
14:10 of this stuff the ability to use this
14:12 stuff
14:13 um that said
14:15 the the the basic product that the World
14:18 Wide Web was was quite simple this stuff
14:20 is quite sophisticated and it's moving
14:22 faster a lot faster
14:25 um so
14:26 the the work product of these things is
14:29 significant
14:31 love your shirt thank you two shirt
14:34 compliments tonight very nice it's hard
14:36 to read the text against but Pi dot AI
14:39 is very empathetic to talk to but asks
14:42 too many questions if you want to get
14:43 [ __ ] done totally agree with that what's
14:45 your name I am Aryan
14:47 all right wait I can't I can't read
14:49 against my shirt I might need to go
14:50 change it
14:51 um yeah Pi dot AI
14:53 if you get the PI app if you search on
14:56 Apple for pi personal assistant you'll
14:58 get that app you put it into speech mode
15:00 and you can talk to it it's really good
15:02 to talk to but if you want to actually
15:04 get [ __ ] done it's a pain in the ass so
15:08 so you want to get [ __ ] done connected
15:10 to the internet that would be barred or
15:12 Bing you want to get [ __ ] done not
15:14 connected to the internet gpt4 gpt4 with
15:18 code interpreter and Claude
15:21 um if you want to experiment and build
15:22 your own [ __ ] use the Llama 2 models
15:25 from from uh meta and if you want to
15:29 make Bots the PO the post site is pretty
15:31 good for making Bots that you can share
15:34 if we are familiar with promoting is it
15:37 worthwhile to find a job remotely
15:41 doing promoting with AI wait if oh if
15:44 you're familiar with promoting
15:45 is it worthwhile to find a job remotely
15:48 doing promoting with AI yeah I don't
15:51 know a lot about that world what I would
15:53 say is if you know a lot about promoting
15:55 get AI literate and start figuring out
15:58 how you can do what you've done
16:00 historically better faster and cheaper
16:03 right
16:04 and so it's not necessarily can you get
16:06 a job with AI what I would say is if
16:08 you're a good promoter become a way
16:11 better promoter leveraging these tools
16:13 right so you might use mid-journey to
16:15 make images for posters and you or you
16:17 know social Graphics you might these
16:20 things use these things to write the
16:21 copy you might use code interpreter to
16:24 generate you know CSV files full of copy
16:27 that you then take over to uh canva and
16:31 do bulk create and just create you know
16:33 tons of promotional content and put that
16:35 out in the world
16:37 um
16:40 you know you could use you could use the
16:43 fact that Bing or Pi or Bard are
16:46 connected to the internet to find
16:47 Opportunities I don't know I mean
16:50 the the short answer to any question of
16:54 you know I've got I'm in this kind of
16:56 profession should I be using AI the
16:58 short answer to it is yes and but I
17:01 wouldn't even necessarily limit yourself
17:03 to a given profession what I would say
17:05 is
17:07 if you're a promoter if you're good at
17:10 doing things in that world you have a
17:12 lot of skills right you can't be a
17:14 promoter and kind of be you know
17:16 singular in your focus you you got to do
17:18 a lot of [ __ ] and do it well
17:20 learning how to use AI to do those
17:23 things really well you may discover
17:24 something that this stuff amplifies what
17:27 you do in a way that you couldn't even
17:29 imagine
17:31 I'll share I I just wrote a if you go to
17:33 my LinkedIn profile I just uh wrote an
17:36 article today about I had a fairly
17:38 profound
17:40 um interaction with chat GPT today
17:44 so
17:45 so this guy I'm volunteering for a
17:48 United Way thing for a small business
17:49 thing and I'm going to go tell them
17:51 about storytelling
17:53 and so the guy running the program it's
17:55 a cohort for small businesses for
17:57 first-time Founders and he he wanted me
18:00 to put together like a little mad libs
18:03 thing for people to do their elevator
18:05 pitch
18:07 and so we talked and you know he said
18:10 the the people are very nervous they're
18:13 first time Founders like the idea of
18:15 pitching makes them really nervous
18:16 they're resisting it they don't want to
18:18 pitch their business they don't want to
18:20 feel sleazy
18:21 and so I thought sure I can you know
18:24 I've done a ton of those in my day so I
18:27 can whip up a little mad libs kind of
18:28 thing like here's what my business is
18:30 about here's here's the problem and
18:32 here's the solution and you know here's
18:34 what my business is about
18:36 um
18:38 and then
18:39 I went to chat gbt and I wanted to get
18:42 there's uh a guy named Simon sinek who's
18:46 got a famous Ted Talk called
18:49 um
18:50 um
18:51 what is it uh lead lead with why
18:55 so so instead of talking about what you
18:57 do talk about why you do it so I went
19:00 there just to get that quote and then I
19:03 started
19:04 kind of brainstorming with chat EBT I
19:07 was going back and forth about how I
19:08 could incorporate this quote into this
19:10 little mad libs thing
19:12 and before I knew it an hour had passed
19:15 and I completely like redefined I got
19:19 rid of the term elevator pitch and I
19:23 came up with the term your belief brief
19:25 and it's a it's a whole framework for
19:28 how someone can tell the why of their
19:30 story here's why I started it and and
19:32 things like that and it you know it was
19:35 just this amazing sort of I went into a
19:37 hyper focused tunnel with my ADD
19:40 and
19:41 you know I I went into it wanting to do
19:44 just a very simple kind of search and an
19:46 hour later I had created this really
19:49 sophisticated framework
19:52 to give to these first-time
19:53 entrepreneurs and so I shared that on
19:56 LinkedIn and and you know it wasn't so
19:58 much what I wrote that was the
20:00 remarkable thing it was the speed with
20:02 which I went from
20:05 you know one one intention and then you
20:08 know I got I gained a little bit of
20:10 insight into that and then it just
20:12 opened up this whole new world
20:14 and and what would have taken me a week
20:16 or two to really think about and really
20:18 write and really put together and
20:20 iterate and back and forth I did all
20:22 that back and forth within within about
20:24 an hour
20:25 and the piece that I wrote is it's like
20:28 a three-page document with a worksheet
20:30 and with tips and tricks on how to do it
20:33 and the actual framework itself and
20:35 examples and opening paragraphs that
20:37 describe what it is
20:39 um so just you know remarkable so so
20:42 back to the promoting thing you likely
20:45 have skills that are worth
20:49 exploring how you do those different
20:52 things with these tools right and with
20:55 the image generation tools and in the
20:56 case of music maybe with the music tools
20:58 the animation tools
21:01 um go to
21:04 um futurepedia.io and start geeking out
21:07 on what are some of the tools that are
21:08 out there you might have some fun doing
21:10 that but if you're new to this stuff
21:11 just start with chat GPT
21:14 and uh but yes I think I think it's
21:16 probably
21:18 you know Wise to get yourself AI
21:20 literate
21:22 all right
21:24 um
21:27 oh good you're here I am here
21:31 The Tick Tock why do older people say
21:35 hmm
21:37 say that before any technical
21:39 application I do it ironically as a joke
21:43 um
21:44 because I'm a bitter cynical Gen X are
21:47 you a little [ __ ]
21:49 but
21:51 but we used to say back when I started
21:54 agency.com in the mid 90s
21:56 um we used to have a thing like like uh
21:59 I need to get the internet installed on
22:02 my computer
22:06 oh yeah did you just start I did just
22:09 start I keep hearing about AAA AI
22:12 automation agencies is that a big
22:14 opportunity new here hey John how's it
22:16 going
22:17 um
22:22 I think AI automation agencies I the
22:26 fact that there's an acronym for it
22:27 already just blows my mind
22:30 um
22:31 basically if you can use zapier you can
22:35 get people to pay you to do [ __ ] right
22:37 now
22:38 um
22:38 I think AI automation agencies are a
22:42 short-term opportunity yes
22:47 I like here's here's my deal this this
22:50 shit's moving so fast that whether it's
22:53 an agents like I wouldn't I if I were
22:56 going to start an agency right now I'd
22:58 just call it an agency and and or I
23:01 wouldn't even call it that I'd call it
23:02 like the like I've got the AI learning
23:05 lab I I also put together a project
23:07 called the AI leverage lab which I was
23:10 just going to turn into an agency
23:12 um but you know I've got too much too
23:14 much other [ __ ] going on including the
23:16 company I'm CEO of
23:18 um
23:19 and and
23:22 I think right now if if you want to get
23:25 work right now you can do it in a couple
23:26 of ways you can teach people how to use
23:28 AI you could do workshops
23:31 um
23:32 you could build stuff for people
23:35 if you're not super technical you could
23:37 build stuff for people using automation
23:39 tools like zappy excuse me zapier
23:42 but I wouldn't lock yourself into any
23:45 particular flavor of how you're building
23:48 [ __ ] for people
23:50 or advising them because it's just
23:52 moving too fast right so if you say
23:54 we're in the AI automation agency and
23:56 all of a sudden autonomous agents get
23:59 their [ __ ] together and now instead of
24:01 having to hand program automations you
24:04 just go in and you say I want you to do
24:06 [ __ ] for me and the Machine just goes
24:08 off and does that
24:11 um your [ __ ] out of business right
24:14 but at that point then you might need to
24:17 understand how to use those agents and
24:19 how to string them together and that
24:20 might not be an automation agency that
24:21 might be an agent agency so so now I'm
24:25 an AI agent agency not an AI automation
24:28 agency still Triple A
24:30 so
24:32 so again I think that I think that the
24:37 the opportunity over the next 18 months
24:39 is so vast that that the only real
24:43 requirement right now is don't sit on
24:46 the sidelines like spend hours I'm
24:49 spending two to three hours every night
24:51 just
24:52 you know rambling my [ __ ] mouth here
24:57 just to keep my head engaged in this
25:00 stuff
25:01 um and and to get other people up to
25:03 speed on it and and start playing with
25:05 it so so if you haven't go here and
25:07 start playing if you're thinking about
25:08 doing that go go build an Automation in
25:12 uh
25:13 in in zapier like if you've got an
25:15 agency background a good way to start
25:17 like the thing things that I've built
25:19 over the past six months have been
25:21 content generation engines where I'll
25:23 create a Google form with some inputs
25:25 I'll take those inputs
25:27 incorporate them into a prompt send that
25:29 off to open Ai and then either post that
25:33 to a slack Channel send it in an email
25:35 post it to a Google spreadsheet
25:41 you can knock one of those out in an
25:43 hour
25:44 and and it's really good practice to
25:46 work on your prompting and work and
25:48 understand the concept
25:50 and then yeah go tell people you can
25:53 automate [ __ ] for them or just go find
25:56 out what problems people have and say
25:57 hey can I help you solve that problem if
25:59 you're just getting up to speed just go
26:01 offer to help people do a couple for
26:03 free help a neighbor
26:06 she's having a bake sale she wants to
26:07 organize [ __ ] she's overwhelmed hey can
26:09 I help you with that I'm learning AI
26:11 okay I don't know what you're talking
26:13 about but sure and then use these things
26:15 to make her project plans and
26:17 spreadsheets and all sorts of happy
26:18 horse [ __ ]
26:20 so so start at whatever level you're at
26:22 do it for other people let other people
26:25 know that you're doing this stuff and
26:26 then yeah I think it's a good idea
26:29 dude have you not learned about the
26:30 shirt issue yet yes I have
26:33 yes I've learned about it and I also
26:36 call it the tick tock I'm just old what
26:38 do you want from me people
26:40 we're not like that anymore oh librarian
26:43 we're not like that anymore oh spinstery
26:46 spinstery kind of oh Mr Shannon
26:50 yeah I know I I had I had just too many
26:53 of those Librarians in I had them in
26:55 high school didn't have any in College
26:56 college is fun
26:59 high school I had him
27:02 um all right woohoo so happy to be back
27:04 on your channel I was in Castle Rock at
27:06 my friend's house oh that's cool Winston
27:07 welcome welcome welcome back
27:10 um I think I'm gonna go change my shirt
27:11 I'll be back do me a favor share the
27:13 live tap the like screen maybe get some
27:15 people here and then when they come in
27:17 and like why am I looking at shelves you
27:19 can just tell them that's what this
27:20 channel is all about
27:48 foreign
27:59 all right
28:00 is that better who's bitching at me
28:02 about the shirt
28:04 uh um someone was
28:10 no costume change Mr Shannon just joking
28:14 I still can't read oh I can still read I
28:16 know I could sort of read but this is
28:18 better
28:19 all right we're good okay uh the answer
28:23 to the question coffee again
28:26 I haven't had whiskey in a while
28:28 agua
28:33 and a little bit of crack cocaine
28:37 we don't joke about drugs
28:39 we don't do that around here
28:42 all right Chachi PT is great for
28:44 debugging code yeah hey retro Punk
28:46 what's happening yeah chat GPT amazing
28:49 for debugging code amazing for
28:51 documenting code amazing for explaining
28:54 code I've done that before I took a
28:56 sequel of a fairly long sequel query and
28:59 I popped it into chat GPT and I said
29:02 explain what this is doing I don't know
29:03 SQL
29:04 and then it explained it and then I said
29:06 oh could you fix some [ __ ] and it did
29:09 and it fixed it wrong first and I only
29:12 found that out the hard way and look
29:14 here's the report it's done and they're
29:15 like these numbers don't look right it
29:17 was like right and I definitely checked
29:20 that
29:23 um one of my new friendships I made
29:25 while I was in Colorado was a Top Gun
29:26 pilot inspector uh instructor at the USA
29:29 AF that's pretty cool yeah those guys
29:32 they'll [ __ ] buzz Denver every now
29:34 and again you'll just feel like
29:37 like four jets flying in formation Blast
29:40 by and it's pretty cool
29:42 hey you got a backup account I do not
29:45 have a backup account
29:47 um I should have a backup count when
29:48 they for when they kick me off but I'm
29:50 only off for seven days they give me
29:52 like the slap on the wrist one the the
29:54 problem with it is like like I'll just
29:57 be doing my live I'll be talking about
29:59 whatever going back and forth and all of
30:01 a sudden just go you've been banned and
30:03 I'm like you know and then I uh whatever
30:06 I I uh
30:08 um
30:08 resist it what's that called appeal it
30:11 and then they say no [ __ ] you you're
30:13 still you're still an evil person but
30:15 they don't tell me exactly they tell me
30:17 what rule was broken but not what I
30:19 actually said to break the rule so I
30:21 don't know how I as a responsible adult
30:23 am supposed to learn if I don't know
30:26 what the [ __ ] I did
30:30 um I trade Futures what AI program
30:34 would you recommend
30:36 um I would recommend not taking any sort
30:39 of investment advice from a guy with an
30:41 acting degree that's what I would
30:42 definitely recommend
30:48 I'll tell you the I'll tell you the
30:50 things that I
30:51 that I think are interesting right now
30:53 and it the the here's the deal it's so
30:57 impossible to know this stuff because
30:59 things are moving so fast
31:03 and I think that the I think that the
31:06 the big step change Innovations are
31:08 likely going to come from unexpected
31:10 places
31:11 so I can tell you what I think is going
31:13 on right now so so Chachi BT
31:15 um they have their [ __ ] together Sam
31:17 Altman's really thoughtful I feel like I
31:20 feel like they are
31:23 steps ahead of the rest of the industry
31:25 both in
31:27 um how they're training their models how
31:28 they're thinking about alignment how
31:30 they're thinking about scale
31:33 um how they're thinking about
31:37 um
31:41 managing through the disruption that
31:43 these tools are going to cause
31:45 so if you look at Sam Altman's World
31:47 coin I think that's a really interesting
31:49 project
31:51 um and if you don't know what that is
31:52 it's basically is cryptocurrency and
31:55 identity blockchain project that and
31:59 payment system
32:01 um that is about
32:03 verifying human beings
32:06 with Iris scans with a thing called the
32:09 orb which is [ __ ] terrifying
32:11 but it's about it's about
32:14 redistributing some of the insane wealth
32:17 that Altman is predicting that these AI
32:19 the prosperity that these AI tools are
32:22 going to generate he's basically
32:24 committing to take a big chunk of that
32:26 from open Ai and and give it back to the
32:28 people that he's going to do that
32:29 through World coin so I think they're
32:31 really interesting Claude is anthropic
32:35 um a couple of things about anthropic
32:37 their ex open AI people so they're
32:39 smarter than [ __ ] I just watched the the
32:42 an interview with the CEO last night
32:45 um he he's wicked smart and um
32:50 but they've got a thing called the
32:51 Constitutional learning model which
32:54 openai uses humans to do their
32:57 um training
32:59 Claude is using this thing called the
33:00 Constitutional learning model feels like
33:02 it has future to me
33:04 um Google is way on their heels they're
33:06 trying to catch up but I wouldn't count
33:07 them out but I wouldn't I wouldn't lean
33:09 into them right now at all meta just
33:11 open sourced llama two pie is super
33:15 interesting
33:17 because of uh they just raised 1.3
33:20 billion dollars uh and it's a deep mind
33:24 founder is the one that created
33:25 inflection so PI right now looks like a
33:28 little toy but I think that they're
33:29 going to be a very significant player
33:32 um I think Runway ml is interesting I
33:35 think mid-journey is by far the best of
33:37 the image generation tools but their CEO
33:39 is a bit of a clown and he's like ah
33:41 [ __ ] copyright he's gonna get his ass
33:43 sued into Oblivion
33:44 so I'd sit on the sidelines with that
33:46 one
33:47 um I think stable diffusion stability AI
33:50 is really interesting
33:52 um they're open sourcing all their Tech
33:53 so I think they're going to get their
33:54 asses sued into the ground too but
33:56 because they're open sourcing things
33:58 that might be a hedge against lawsuits
34:01 uh
34:03 um
34:04 I don't know Microsoft is playing like a
34:07 startup again so they're interesting
34:10 Nvidia seems to have a [ __ ] Monopoly
34:14 um amd's trying to break into the into
34:16 the GPU Market but whatever the [ __ ]
34:20 barrier is for scaling up chip
34:22 production
34:24 um like like Nvidia seems to have the
34:27 entire developer Community locked and
34:30 you know but you know they got a
34:32 trillion dollar market cap so I don't
34:33 know what the upside is there but
34:35 they're [ __ ] killing it
34:37 um
34:38 Apple
34:40 so they're apparently so a couple of
34:42 things built into iOS 17 is voice
34:45 cloning it's in there right now in the
34:47 beta that you can play with
34:50 um
34:51 they've got they're rumored to have
34:54 created a a language or a framework
34:57 called Aja I think it's called Ajax uh
35:00 for large language model creation and
35:02 they've got something called Apple GPT
35:05 that they're testing internally and
35:09 um
35:10 they're spending the bulk of their 24
35:14 billion dollar r d budget
35:19 24 billion dollar r d budget which is
35:22 bigger than the market cap of open AI
35:24 Apple's spending like their couch change
35:28 all on AI on generative AI so I have a
35:32 sneak in suspicion that Apple's gonna
35:34 come out [ __ ] swinging
35:36 and you know they've got this spatial
35:39 Computing thing that's going to be heavy
35:41 in AI I think they're just going to
35:42 bleed out all that AI technology apples
35:45 Apple's interesting because they're
35:47 going to be able to
35:48 they've got credibility with privacy and
35:51 they've got the technology for privacy
35:52 and they've got the technology to run
35:55 machine learning [ __ ] locally you're not
35:58 gonna have to send it to Apple so
35:59 Apple's going to be able to legitimately
36:02 say
36:03 you can GPT the [ __ ] out of all of your
36:06 data and we're not going to steal it or
36:08 give it to the government and I think
36:10 they're the only ones that can say that
36:13 um they've they've taken a pretty hard
36:16 line on that [ __ ] so
36:19 um yeah I don't know
36:20 I wouldn't go invest in any of that [ __ ]
36:22 on my advice but those are the ones I'm
36:24 looking at
36:26 all right uh he was in the Air Force for
36:28 20 years as a fighter trainer now just
36:30 flying as a commercial pilot yeah a lot
36:32 of the a lot of the Air Force dudes end
36:33 up going commercial it's funny I was
36:35 talking to a commercial pilot once and
36:37 and we saw we we hit some some weather
36:40 and he he
36:43 like he flew through it fine and
36:46 and I was talking to him and he told me
36:48 he was in an Air Force fighter pilot
36:51 and uh I said what's it like flying one
36:54 of these things he said it's like
36:55 driving a bus he was like really
36:58 disappointed and I said well we hit all
37:00 that weather there like did that get
37:01 your adrenaline going at all and he was
37:03 like no
37:05 so
37:07 you know I
37:10 there's probably not much else short of
37:12 driving an F1 car that's like flying a
37:15 you know nf-16 or an F-18
37:18 so so yeah all those guys gotta go
37:22 commercial and
37:24 you know I suppose the the ones that are
37:27 still Adrenaline Junkies go do the air
37:29 shows and things like that but it's
37:31 crazy that's really funny
37:34 all right are there any decent local
37:37 hostable large language models you can
37:39 train yourself
37:41 um yes but I'm I haven't played deep in
37:44 that world
37:45 but go so where I would start is go to
37:48 Reddit and just you know search for open
37:52 source large language models I'm sure
37:54 there are two or three subreddits that
37:55 are just dedicated to that [ __ ]
37:57 the developer Community right now has
38:00 kind of lost their mind on llama 2 so
38:03 you can go play with the Llama 2 models
38:05 at labs.perplexity but if you want to go
38:07 train your own [ __ ]
38:09 um you know just just go hang out on
38:11 Reddit look at the projects that they're
38:13 doing
38:15 um there's a project called GPT for all
38:17 that lets you
38:19 um use lots of different large language
38:21 models if you want to start training
38:24 large language models on your own data
38:26 then I would go look at a project called
38:28 Lang chain
38:29 which which can chain different
38:31 components together including the
38:33 embeddings and Vector databases that you
38:36 search against
38:38 um Lang chain is quite powerful
38:41 um most of the developers that I know
38:42 are using it that are doing any any any
38:44 of this sort of stuff
38:47 um I'm I've got an instance of it in a
38:50 like a Sandbox for my company storyvine
38:53 I've got another project that I'm doing
38:55 and the developer there is using it
38:57 pretty extensively he's building some
38:59 really cool stuff with it so
39:02 um
39:03 but there are lots there are lots and
39:05 lots and lots of Open Source large
39:08 language models that you can train or
39:10 that other people have trained that you
39:11 can use
39:13 some of them you can install on
39:16 non-optimized computers like M1 Max and
39:18 up and uh you know like a simple GPU on
39:22 a PC and up
39:25 I think within a year we'll be able to
39:27 install GPT 3.5
39:31 um
39:32 equivalent models like just locally on
39:36 our phones locally on our laptops so
39:38 it's it's gonna be it's gonna be a
39:40 something it's going to be quiet
39:42 something
39:43 and again pay attention to things like
39:45 uh
39:46 like what Apple's doing apples Apple's
39:49 Gonna Come Out Swinging and and there
39:51 they're gonna they're gonna be
39:55 I wouldn't be surprised if Apple finds a
39:57 way to reinvent how we think about these
40:00 large language models
40:02 either with autonomous agents that you
40:04 know like personal assistants that do
40:07 you know imagine if Siri didn't suck and
40:09 don't listen to me
40:12 she keeps turning herself on every time
40:14 I say her name Simon is great Simon
40:16 cynic yeah he's amazing
40:19 um anyway I was very busy with about 30
40:20 people at my friends and I wished I
40:22 could have met with you too oh yeah if
40:25 you get to Denver drop me a line you can
40:27 come to we I do an AI Salon that's the
40:29 other thing if you're if you're into uh
40:32 into this stuff
40:34 um I started a group called the AI Salon
40:35 you can go to the salon.ai and that's
40:38 it's a link tree that's got our Meetup
40:40 our Discord server and our LinkedIn
40:43 group
40:44 and like our mission statement and [ __ ]
40:46 like that and we meet every other
40:47 Tuesday we have a meeting coming up this
40:50 coming Tuesday the whatever it is the
40:53 15th and John Knack from adobe's AI
40:56 group is going to be speaking there he
40:58 spoke with us back in January
41:01 and they hadn't done Firefly yet they
41:04 were just they were like we're not sure
41:06 if we want to do this AI thing like
41:08 we're not sure if we want to like lead
41:09 into this generative AI thing because at
41:11 the time it was very very like the
41:13 artists were [ __ ] pissed off like way
41:17 more than they are now like like I knew
41:18 some artists that were were doing some
41:20 AI generated art projects that were
41:22 getting [ __ ] death threats
41:26 um so Adobe early on was like we're not
41:28 sure if we're gonna do this we might dip
41:30 our toes in the water and then like two
41:32 months later they came out with firefly
41:34 and they've just been Off to the Races
41:36 they got themselves a Discord server
41:38 they got Firefly up and running they
41:40 rolled that [ __ ] straight into
41:41 generative fill within Photoshop
41:44 and uh and the the other thing about
41:46 Adobe right now is
41:49 all of their models are trained on
41:51 public domain and Adobe stock licenses
41:54 that they own
41:57 um so they haven't trained anything on
41:58 you know active copywritten artists and
42:01 they're they're creating a system to
42:03 allow artists to opt out of being
42:05 trained up into being trained
42:08 attribution models compensation models
42:10 so I think adobe's doing it right so
42:12 anyway John's gonna come to the salon
42:14 next Tuesday and he's going to talk
42:16 about
42:18 whatever he's allowed to talk about but
42:20 he'll probably give us a little insight
42:21 into like what I want to hear from John
42:24 is what have they learned
42:26 thank you David gold appreciate that
42:28 Gould probably not gold there's a u in
42:32 there
42:33 so
42:34 and I'm an old guy the David Gould The
42:37 Tick Tock
42:38 um
42:41 was I gonna say yeah yeah I wanna I
42:43 wanna hear from John like what have they
42:44 learned
42:45 because like they just sort of you know
42:47 hit the ground running and they didn't
42:49 seem to catch the Wrath of the art
42:51 community because they did it right like
42:53 they talked to all of the major issues
42:55 um so I'm just curious what they've
42:57 learned what they're excited about
42:59 any little sneak peeks they can give us
43:00 about what's coming
43:02 and John's great he's he's a guy I used
43:04 to work with him at agency.com he was
43:06 actually an intern at agency.com
43:08 and uh if anyone often became like a big
43:12 [ __ ] deal he was a product guy at
43:14 photoshopped for years then he went to
43:16 Google for a bunch of years now he's
43:18 back at Adobe doing AI stuff
43:22 so that should be interesting so yeah so
43:24 come to the salon and yeah if you're in
43:26 Denver ping me let's go grab a coffee or
43:28 something
43:29 all right let's see
43:33 moving on Tick Tock jumped me to the
43:35 bottom I'm trying to figure okay there's
43:36 Simon is great all right
43:40 um hello from Maria Lupita from the
43:43 Laguna Beach area welcome welcome Maria
43:47 glad to have you here nice Emoji work
43:50 there I'd say 9.5 out of 10 on the Emoji
43:54 work it's solid
43:55 the work acceleration from
43:58 oh the work acceleration from AI is so
44:00 exciting I know Aaron it's
44:02 [Music]
44:03 well it's funny it's funny like the
44:05 story I don't know if you heard earlier
44:06 when I talked about I just had this
44:08 little 10 minute project to do today and
44:10 I went to Chachi BT to do a little
44:12 research and then it just ballooned into
44:14 this sort of hour and a half long
44:17 um
44:18 brainstorming session with chat GPT and
44:21 I created something really remarkable so
44:23 in that case it actually slowed the
44:27 speed of the work but it accelerated the
44:29 the quality of the work dramatically
44:32 like like the guy that asked me to do
44:35 something was just expecting like a
44:36 little mad libs kind of paragraph and I
44:38 sent him a three-page [ __ ] document
44:40 that redefined what a what an elevator
44:42 pitch was for a first time founder I'm
44:44 like that's pretty cool
44:47 um
44:48 so so yeah it's I I just I I swear to
44:52 God man every day I'm just like really
44:54 really you can do that
44:58 I don't I don't know when that's gonna
45:00 go away I'm trying to think back in the
45:02 early days of the web
45:05 I mean in the early days of the web the
45:07 technology was so basic
45:10 that the things we got excited about
45:12 tended to be like incremental
45:14 technological improvements like so when
45:18 I first started building websites you
45:20 could center text but you couldn't
45:22 Center images the center tag didn't
45:24 apply to images
45:26 so so all images were left Justified so
45:29 that drove us [ __ ] crazy so we
45:31 invented a whole little series of pixel
45:33 High invisible images to scoochar
45:36 GE our images to the right so we could
45:39 center our images with the text
45:42 um and then when you know you know when
45:44 the center tag came out we were like
45:48 and I remember when when
45:51 um flash came out and we could do
45:52 animations because we had been we had
45:55 been writing uh Pearl scripts to to do
45:59 animations where the the animation was
46:02 actually a series of calls back to the
46:03 web server and that was the animation
46:05 and then animated gifs came out and
46:07 Flash came out and we were doing games
46:09 within banners it was it was crazy but
46:12 all of that stuff were like incremental
46:14 improvements that we got excited about
46:16 here it's kind of like
46:20 I I mean they're they're exciting
46:22 technological things happening but it's
46:25 it's almost more like
46:27 the more people and companies start to
46:29 get their head around this stuff they're
46:31 like well here's how we're using it and
46:33 then you look at it you're like oh [ __ ]
46:34 I've never even thought about that
46:36 so yeah it's it's just crazy times right
46:39 now it really is
46:42 I find that AI is helping me bring my
46:44 vision out of the ethers and into the
46:46 tangible Denise that that's I that to me
46:49 feel that sorry I I agree with you so
46:54 much I literally can't talk
46:58 um
47:00 I think that's the biggest thing for me
47:02 that chat GPT does is I'll have a
47:05 whisper of an idea and within three
47:09 minutes five minutes that Whisper of an
47:12 idea is at 80 percent
47:14 execution now it still takes me very
47:18 often two hours to get the last 20 done
47:20 but like that's the reality anyway we're
47:22 creative people right we're constantly
47:24 you know pushing the boundaries of
47:26 deadlines but
47:29 but it
47:30 that ability to get something out of
47:33 your head quickly and articulated
47:34 quickly and then and then and to have
47:37 that idea Spur new ideas that you can
47:40 bring those to life quickly like I find
47:42 myself knocking out five or six major
47:44 Concepts in 30 to 45 minute periods and
47:49 then I kind of curate those and some
47:51 will park some will toss oh that was
47:55 just a stepping stone that's I totally
47:57 that like that's it for me
47:59 that's it for me is it's it's
48:03 yeah it's it's a it's a
48:07 it's an idea amplifier an ability
48:09 amplifier Ian my wife is sleeping so I
48:12 can watch you
48:16 what do you mean you're watching Kyle
48:18 again
48:19 you don't want to watch TV with me
48:21 yeah my wife was like oh you're going on
48:23 Tick Tock again
48:25 so I get it I totally get it Claude and
48:29 Poe gave me the McAllister moment the
48:31 other day oh Nicholas tell me what
48:32 happened what
48:34 so Claude info so I'm guessing Claude it
48:37 was about the hundred thousand token
48:39 context limit
48:41 um and uh Poe I'm not sure maybe it was
48:44 the the Bots
48:46 um the the Kevin McAllister moment if
48:48 you don't know so so I I say on the
48:50 channel a lot that you should use these
48:52 tools long enough to have your Kevin
48:53 McAllister moment which is
48:56 the Home Alone moment
48:59 um another handsome shirt thank you
49:00 belief belief brief that's F and
49:03 Brilliant so the bear thank you I I'm
49:06 I'm really excited about the belief
49:08 brief thank you very much
49:10 um
49:12 yeah it started out as an elevator pitch
49:14 and then it I went to Simon cynics y
49:16 thing and then I was saying is there a
49:19 framework for how to talk about the why
49:20 and then it just evolved and evolved and
49:22 evolved and so I had this whole
49:24 framework and then it said here's your
49:26 elevator pitch uh framework or whatever
49:29 it was recipe
49:31 and I'm like elevator pitch still kind
49:33 of sucks like that still feels like this
49:35 intimidating thing and so I did a naming
49:38 session with chat GPT and I probably
49:40 came up with
49:42 60 or 80 names I was doing them in
49:45 batches of 10.
49:46 and it it was just like shitty batch of
49:49 names shitty batch of names but then
49:50 there'd be like one little phrase in
49:52 there I'm like oh that phrase is good
49:53 and I sort of pull that out and I do
49:55 iterations on that
49:57 and it was I don't know it was like idea
50:00 84 was belief brief and uh I was like
50:04 that's good
50:05 so yeah I know isn't that fun
50:09 um zapier versus make versus python
50:13 um
50:14 I would say
50:16 the bear are you stalking me at the AI
50:18 learning lab
50:20 the kids you kids you kids
50:23 no fraternizing amongst the amongst the
50:26 uh the audience
50:29 um I don't give a [ __ ] Fred nice all you
50:30 want
50:31 um
50:32 zapier versus make versus python
50:35 zapier's been around for longer which is
50:38 good and bad it's good because and I
50:40 don't know why I'm talking to you with a
50:42 pen but listen little mister listen
50:44 little Mr Ian
50:47 um zapier's been around longer so it's
50:49 got more
50:51 um connections more API hooks to
50:53 different Services it's got like more
50:54 than five thousand make is newer and and
50:58 has a sexier interface but they're
51:01 they're both kind of exactly the same
51:03 thing right you've you've got you know
51:05 it's kind of a linear flow that can
51:07 branch
51:09 um
51:10 and then python is if you know python
51:13 like here's the thing with no code the
51:16 the problem with no code is that
51:19 you have to learn their [ __ ] platform
51:22 right you have to learn all the
51:24 intricacies of the platform and what
51:26 works and what doesn't and like every
51:28 little module has its own [ __ ] rules
51:30 and
51:32 it's probably if someone is really
51:35 competent with python they could
51:37 probably do what you do in zapier faster
51:40 than I could do it in zapier but if I
51:43 don't want to [ __ ] Learn Python I'm
51:45 going to use zapier right and I just I
51:48 don't know make I've looked at make I
51:50 like how it looks but I went to build
51:54 something and it just gave me one of
51:56 those weird [ __ ] automation errors
51:57 like I whatever the [ __ ] it was it's
52:00 just some stupid [ __ ] error that if I
52:02 had read the manual
52:03 um I wouldn't have had the issue but but
52:06 um uh but I don't read manuals because
52:08 that's just not how I roll I've never
52:11 read manuals I have a son that reads
52:12 manuals he drives me [ __ ] crazy
52:14 I'm like there's three screws in a bolt
52:18 what are you reading the manual for what
52:20 could we possibly [ __ ] up
52:23 and then of course I [ __ ] some things
52:24 up with that attitude but there's other
52:26 times when I just build [ __ ] and it was
52:28 fine
52:30 and I can use my brain I don't need a
52:32 [ __ ] manual I can figure it out
52:33 anyway
52:35 thank you for the heart
52:39 um so I'd say whichever one you're more
52:41 comfortable in they're probably all in
52:43 the neighborhood of the same
52:44 morgie calling you all in thank you I
52:47 assume that means you shared this thank
52:48 you very much well it's been
52:50 blessed with good eh
52:53 oh I see you're talking to each other
52:55 cool wait oh Morgan in a few hours is
52:57 the all-in podcast
53:00 John interesting
53:02 crypto Joe lady that went viral on the
53:05 plane saying the guy wasn't real is said
53:07 to be a marketing exec yeah those
53:10 marketing exists execs dear little
53:12 cuckoo
53:14 advertising is full of special people
53:17 they're special
53:19 the big Egos and big crazies
53:23 but they're smart and fun
53:27 yeah ad people are nuts
53:29 marketing people are crazy grades and
53:32 crazy
53:33 but they're storytellers what the [ __ ]
53:34 do you want
53:36 your storytellers are gonna have a
53:38 little scroll boost every once in a
53:39 while how else could they tell the story
53:42 they're The Observers of society you
53:45 know what I mean we're all out there
53:47 doing the storytellers are sitting back
53:49 going huh very interesting very
53:52 interesting
53:53 well that Line's going into play
53:57 I use I'll tell you what if you've never
54:00 been to New York or if you haven't been
54:01 there in a while I [ __ ] miss New York
54:03 I haven't I haven't been there in
54:05 way too long
54:08 um one of my favorite things in New York
54:09 was just go to go [ __ ] anywhere
54:12 go to a diner go to the subway and just
54:15 have a notebook open and just listen oh
54:18 my God
54:20 oh the lines
54:24 the lines
54:26 all right I went to block and bottle
54:29 in Castle Rock they went to a pretty
54:31 large potty party at Casa Bonita how was
54:33 the food at Casa Bonita I haven't been
54:35 yet I hear it's good although although
54:37 all the servers went on ahead and sued
54:39 uh the South Park guys because I guess
54:42 they weren't paying them or they weren't
54:44 doing what they promised or they weren't
54:45 opening or something
54:47 this guy is cool I don't know who you're
54:49 talking about ah it's not me I say The
54:52 Tick Tock that's not cool any decent
54:55 locally hostable large language models
54:56 we already talked about that
54:59 oh by the way if I'm not answering your
55:02 questions it's because I'm reading all
55:04 the questions so I'm I'm now 38 messages
55:07 or more behind
55:09 so don't go hating on me and don't go
55:12 asking the same question 47 [ __ ]
55:14 times because as I scroll through it I'm
55:15 gonna see him we had a situation last
55:17 night with a guy with an unpronounceable
55:19 name ask me questions 462 [ __ ] times
55:22 it got ugly
55:24 I do not suffer fools lightly especially
55:27 when I can't read their name
55:30 I don't know what's going on with my
55:31 voice tonight
55:34 it ain't it ain't
55:36 something something is awry up here
55:41 and something is a pumpernickel
55:43 huh
55:45 it's comedy
55:50 all right
55:51 we have fun we have fun we like to have
55:53 fun I like that shirt but this record is
55:56 also good I know it's good right
55:59 albums the vinyl
56:02 spin those Tunes yes much better
56:06 you were bitching about the shirt Kyle
56:08 oh you were bitching about the shortcut
56:10 I know Emilio's wife
56:11 all right much better shirt for reading
56:13 thank you yo yo which AI can take a zip
56:15 file and read through it that would be
56:17 claude.ai and
56:20 um code interpreter not Claude
56:24 chat.openai.com has code interpreter
56:27 that can read a zip file I think Claude
56:29 can also read a zip file let me go check
56:33 I know Claude can do PDFs
56:42 where's an archive come on I got I now
56:45 got one in here somewhere
56:49 [Music]
56:53 wait any archive no archive
56:56 name contains
57:03 where
57:06 I don't get it I just don't get it
57:09 people
57:10 what's happening
57:15 oh maybe it can't
57:23 um
57:27 oh it can't okay
57:29 so
57:31 um
57:33 okay
57:35 chat gbt using Code interpreter you can
57:38 upload a zip file to that and it can be
57:40 full of a bunch of other files PDFs and
57:42 csvs
57:43 um
57:45 code interpreter also just you can now
57:47 do multiple multiple files up to code
57:50 interpreter
57:51 and I'm happy to show that here if
57:53 anyone wants to see it have I seen hey
57:55 Jen Avatar pre-release today I tagged
57:57 you on LinkedIn oh cool retro Punk yeah
58:00 I saw that I've seen hey Jen is is
58:03 clearly getting ready to release
58:04 something because I've seen them
58:06 mentioned like 10 times in the past two
58:08 days so their PR firm is is earning
58:11 their stripes
58:12 um so so something's up with hey Jen
58:16 um
58:17 I saw a tick tock today where a guy was
58:19 talking about hey Jen and canva
58:22 um
58:24 I've seen a couple of people talking
58:25 about it on X so yeah how can we ever
58:29 prove AI can be sentient if we can't
58:31 solve the hard problems of consciousness
58:34 well user 588-59-3919
58:38 um by the way 0.5 on the name come on
58:41 seriously
58:42 like like
58:44 Bob at least right anyway
58:48 um
58:50 Bob five eight eight five nine three
58:52 nine one nine
58:54 um
58:56 I don't know what you mean about if we
58:58 can't solve the hard problem of being
59:00 conscious we are I I would say certainly
59:03 in America we are largely unconscious
59:07 um Society right now we're all living on
59:10 our amygdalas just [ __ ] triggered
59:13 every minute of every day
59:15 um so it would be nice if we got back to
59:17 a little more
59:19 um consciousness
59:22 I
59:25 Here's my thought on sentience and AGI
59:28 and is it smarter than us or not all
59:32 that sort of [ __ ] from where I sit is
59:35 [ __ ] irrelevant like who cares
59:38 okay at the point at which we go okay AI
59:41 sentient now what changes nothing
59:45 right these things are already more
59:48 capable like these things already know
59:50 more than any of us individually will
59:53 ever be able to learn ever in our lives
59:55 if we dedicated the rest of our lives to
59:57 learning all of the [ __ ] we could not
59:59 keep up with where these things are
1:00:01 right now
1:00:03 we can't code as fast we can't write as
1:00:05 fast we can't write as comprehensively
1:00:08 and these things are in their shittiest
1:00:11 form that they will ever be
1:00:14 so at the point at which we say okay now
1:00:17 it's sentient now it's conscious now
1:00:19 it's AGI
1:00:21 it's just it like I feel like at this
1:00:23 point it's just a Continuum right we're
1:00:25 we're already at the point and it's even
1:00:29 funny like there's people that have got
1:00:31 there are all these different
1:00:33 um
1:00:34 like definitions of when these things
1:00:37 pass the Turing test it's like okay well
1:00:40 it passes the Turing test for five
1:00:42 minutes but I want to talk to it for
1:00:44 four hours and if it if it can convince
1:00:47 me that it's not a computer for four
1:00:49 hours that then it's past the touring
1:00:51 well okay like whatever like these
1:00:54 things are no longer Alexa in Siri right
1:00:57 these things are Way Beyond
1:01:00 anything that we have historically
1:01:02 called a chat bot Way Beyond
1:01:05 right these things are transforming
1:01:09 work and
1:01:11 jobs and all that stuff right now in
1:01:16 their shitty janky duct tape ass form
1:01:20 um so I agree with you we can't get our
1:01:23 [ __ ] together as humans and I don't
1:01:25 think it's relevant when we label that
1:01:28 these things are more human than human
1:01:30 or not
1:01:31 like it's going to be good for a cover
1:01:33 of Time Magazine but
1:01:35 how about IG account
1:01:39 Instagram what I don't know what you're
1:01:41 talking about
1:01:42 Ian don't take any advice from Jim
1:01:45 Cramer
1:01:46 you should get a backup account because
1:01:48 I think that subsequent bands can be
1:01:50 longer oh well they can be if well if
1:01:54 they if if they do it again I I'm trying
1:01:56 to be I'm trying to be a little careful
1:01:58 now I I think I know the areas that I'm
1:02:00 not allowed to dance around it basically
1:02:04 yeah I'm not even going to talk about it
1:02:06 because I can't even talk about it or
1:02:07 let's shut this thing down any any I any
1:02:11 AI can read zip files and interpret the
1:02:14 code okay answers yes
1:02:17 and what we'll do so here we are at chat
1:02:19 gbt
1:02:21 um
1:02:22 in order to use code interpreter you
1:02:24 need access to GPT plus when you go to
1:02:27 the gpt4 area you'll see code
1:02:30 interpreter and plugins if you pay for
1:02:32 chat GPT plus and you don't see these
1:02:34 that's because they're hiding it from
1:02:37 you because they're evil little bastards
1:02:39 so to turn that [ __ ] on go to your
1:02:42 settings
1:02:43 and go to Beta beta features
1:02:47 and then you turn them on
1:02:49 why are they doing that because they
1:02:51 want to protect you from yourselves I
1:02:53 don't know they're pretending like
1:02:54 that's safer than not turning that [ __ ]
1:02:56 on okay
1:02:58 so now we're in code interpreter
1:03:00 next to your little doodaddy here is a
1:03:04 plus button and I can now upload uh
1:03:07 anything videos images uh
1:03:12 uh zip files csvs PDFs
1:03:17 so let's go grab some new data so I'm
1:03:20 going to run over to a site called
1:03:22 kaggle
1:03:24 kagg
1:03:27 Le
1:03:30 wait k a
1:03:32 g g l e okay
1:03:36 and I'm going to go to data sets and
1:03:38 then we're gonna go find us a data set
1:03:41 oh we got one here oh it's only 7K
1:03:44 we we have a data set for McDonald's
1:03:46 nutrition that would be a fun one
1:03:49 let's see top 200 Spotify songs database
1:03:53 Magnus Carlson's
1:03:56 games you know what we're gonna do we're
1:03:58 gonna do a filter here we're gonna we're
1:04:00 gonna get some decent sized ones we'll
1:04:02 go 10 megabytes
1:04:07 to a hundred megabytes
1:04:12 and let's grab csvs
1:04:17 all right
1:04:20 things on Reddit
1:04:22 S P 500 Global education statistics all
1:04:26 indicators that sounds interesting
1:04:30 um TV and movie metadata with genres and
1:04:33 ratings
1:04:35 uh
1:04:37 um
1:04:39 Metro train data set let's do Global
1:04:42 education okay
1:04:44 so I'm gonna download this 18 megabyte
1:04:47 file
1:04:48 looks like it's down is it down
1:04:51 did this just come in
1:04:55 today at 9 10 yes okay
1:04:58 so we're gonna go back to your yield
1:04:59 chat GPT
1:05:01 you can make money
1:05:07 we're gonna grab archive two I'm gonna
1:05:10 open it okay so that's gonna now upload
1:05:12 that file so it's taking a little little
1:05:14 moment here
1:05:17 and
1:05:19 we're gonna type in here
1:05:24 [Music]
1:05:27 please open this archive
1:05:30 and examine the files within it and tell
1:05:34 me what might be interesting about them
1:05:38 so you can just do that you can just say
1:05:42 go do [ __ ]
1:05:44 without being very articulate and then
1:05:46 it goes and does it so so what it's
1:05:48 actually doing here
1:05:49 is
1:05:51 spun up a virtual machine inside this
1:05:54 chat session
1:05:55 and it's writing and executing its own
1:05:58 python code
1:06:00 and
1:06:02 something is up with the focus again
1:06:04 this new phone's been better on the
1:06:06 focus but this is oh it's because
1:06:08 hang on my hand is in there let me see
1:06:11 if I can get it so my hand's not in the
1:06:13 shot
1:06:15 and not really let's see can we do it
1:06:17 there get rid of that oh hey hang on
1:06:20 please just uh I'll be there
1:06:24 okay
1:06:27 okay maybe that'll work
1:06:30 so it's it's writing and executing its
1:06:33 own python code
1:06:35 um it says this archive contains 15 CSV
1:06:39 files their names seem to be mostly uuid
1:06:42 based
1:06:44 except for one that mentions education
1:06:47 statistics
1:06:50 um one what do you want me to do I can
1:06:51 check the first few rows of each file
1:06:55 investigate any unique patterns or
1:06:58 outliers would you like me to start by
1:07:00 examining
1:07:02 or is it uh yeah just uh
1:07:07 you decide that's what I'm gonna say you
1:07:09 decide
1:07:14 all right let's start by examining the
1:07:16 first few rows of a selection
1:07:18 of these files and so again it's going
1:07:21 off
1:07:22 so it's it's starting with the one that
1:07:24 it thinks is the most interesting
1:07:26 it's it's not starting with the uuid
1:07:28 files
1:07:33 and again what it's doing there is
1:07:35 writing its own python code so it says
1:07:37 here's a quick overview
1:07:39 unfortunately I encountered an error
1:07:46 there's literacy proficiencies
1:07:49 of adults
1:07:54 language proficiency of second grade
1:07:56 from the initial examination it seems
1:07:59 these files might contain educational
1:08:01 statistics possibly for various
1:08:04 countries and years the data is
1:08:05 potentially related to proficiency
1:08:07 levels blah blah to further understand
1:08:09 this data we could check the number of
1:08:11 unique countries
1:08:12 examine any trends
1:08:15 or examine files to the C so let's say
1:08:18 um trends
1:08:19 we want to look at Trends and so let's
1:08:21 see what it does so so now what it's
1:08:23 going to do is it's going to go like
1:08:25 what's
1:08:27 why this feels like the future to me is
1:08:30 my input here is very minimal including
1:08:34 my my knowledge of data analysis and how
1:08:38 to even deal with data sets right I
1:08:40 don't know how to do that but I can come
1:08:42 in here and with a handful of things
1:08:45 just turn it loose to start working so
1:08:48 this is an example of an autonomous
1:08:50 agent
1:08:51 um it's not really going out and you
1:08:53 know using other tools and things like
1:08:55 that but it's running this you know
1:08:59 virtual computer system within my chat
1:09:02 session and it's got access to python
1:09:04 libraries and it understands python code
1:09:06 and it understands how to problem solve
1:09:11 and so it is remarkable
1:09:14 um I'm in the middle of a project right
1:09:17 now where I've got an old Legacy
1:09:20 Excel spreadsheet that just isn't
1:09:23 working I can't get it to work and so
1:09:25 I'm doing my data analysis for this
1:09:27 report that this
1:09:29 spreadsheet was supposed to do I'm doing
1:09:31 it all in code interpreter
1:09:33 and you know it takes me like 10 minutes
1:09:34 I go in I grab raw data
1:09:37 I turn it into tables I turn it into
1:09:39 charts and graphs I give it to the
1:09:40 account people and they go deal with it
1:09:41 and and I don't have to do any coding I
1:09:43 don't have to learn SQL SQL queries all
1:09:47 that sort of stuff
1:09:48 so let's see what it's doing it seems to
1:09:50 be there's an error so they're just made
1:09:52 a chart with nothing on it
1:09:54 and it observed that there's no there's
1:09:56 no data available so it made an empty
1:09:59 chart
1:10:00 uh
1:10:01 I'm gonna say find
1:10:05 enter
1:10:07 esting
1:10:10 data
1:10:11 in one CSV
1:10:16 let's see what that does
1:10:18 I might have picked a a janky data set
1:10:21 here
1:10:25 again if I knew more what I was doing
1:10:26 with data and Analysis that's so so one
1:10:30 of the other things that I think is is
1:10:31 important to recognize
1:10:34 me being clueless about data analysis
1:10:37 I'm going to be able to use code
1:10:38 interpreter like just a little bit like
1:10:40 a wee little bit but if someone's
1:10:42 actually good at data analysis they're
1:10:44 going to be able to use this thing in a
1:10:47 remarkable way
1:10:50 um let's see the file contains 270
1:10:52 countries 140 unique series descriptive
1:10:55 statistics
1:10:57 given the area a few interesting paths
1:10:59 we can take country analysis series
1:11:01 analysis
1:11:03 or Trends over time let's do
1:11:06 trends
1:11:08 over time
1:11:11 so if you're good at a particular thing
1:11:15 you're going to use chat gbt or code
1:11:18 interpreter or Claude or mid-journey
1:11:20 you're going to use it way different
1:11:22 than I use it and that's a good thing
1:11:24 right and that's another reason I like
1:11:26 this channel is the more people that
1:11:29 come in here with different points of
1:11:30 view and perspectives
1:11:33 like teach us all
1:11:36 different ways of looking at these
1:11:37 things and it's that's what we need
1:11:39 right now right the only way we can
1:11:41 really learn what's possible with stuff
1:11:44 that's this different from how we've
1:11:46 done things historically requires lots
1:11:49 of lenses being looked through
1:11:52 all right and this is getting boring
1:11:54 because it's like it hasn't quite
1:11:56 figured out how to pull the data out of
1:11:58 these things
1:12:00 but anyway that's that's code
1:12:01 interpreter that's how it works there
1:12:03 just made Another Empty chart okay we're
1:12:05 gonna we're gonna bail you on you on you
1:12:07 stop generating stop just stop just stop
1:12:13 we have no use for you code interpreter
1:12:15 code interpreter is useless tonight
1:12:19 all right I'll get a different data set
1:12:26 wait you could rip the transcript of
1:12:29 your offering
1:12:31 oh of your offending video give it to
1:12:33 chatgpt along with Tick Tock rules oh
1:12:35 that's a good idea River
1:12:37 thank you David Gould what is that thing
1:12:39 oh it's a tick tock music
1:12:42 thing it's a 3D neon tick tick tock tick
1:12:47 tickety tock
1:12:49 um River that's a really good idea take
1:12:51 take so down download the offending
1:12:54 video
1:12:55 transcribe it put it into chat GPT with
1:12:59 the tick tock rules yeah in the middle
1:13:02 find it yeah I could do that I don't
1:13:05 really care thank you Kay appreciate
1:13:08 that
1:13:10 what about IBM's Watson I know
1:13:14 so so they did they did
1:13:17 um announce like six weeks ago Watson
1:13:21 Watson AI was it Watson AI
1:13:24 or Watson oh Watson X yeah they
1:13:27 announced Watson X yeah so they had
1:13:29 Watson and then they had Watson health
1:13:31 and that was going to be that was gonna
1:13:32 change the world right thanks David
1:13:33 Gould and um
1:13:37 um
1:13:38 I really do appreciate that
1:13:40 um and then it just flamed out they
1:13:42 screwed it up somehow I don't I don't
1:13:44 know what they did I mean but but you
1:13:46 know Watson Health was pre-transformers
1:13:49 right all these tools that we're playing
1:13:50 with now this is all possible because of
1:13:53 the Transformer that was invented in
1:13:55 2017. I think Watson Health was kind of
1:13:57 on its you know
1:14:00 on its downswing at that point wow David
1:14:02 thank you so much and and TM Tim
1:14:08 good name
1:14:10 um
1:14:11 yeah so all these things are are
1:14:13 leveraging Google's Transformer
1:14:15 technology and so it's mid-journey and
1:14:18 stable diffusion and so are the image
1:14:21 generation or the the video generation
1:14:23 tools and the voice synthesis tools and
1:14:25 the animation tools they're all using
1:14:27 the same [ __ ] right now which the other
1:14:30 good Lord David that's very that's
1:14:32 awesome thank you so much
1:14:34 um
1:14:36 um
1:14:37 what was I gonna say the uh oh so so
1:14:41 right now all of these things are using
1:14:43 um you know the Transformer model but
1:14:45 there's all these other different models
1:14:47 and there's models that have that are in
1:14:49 research right now that we don't know
1:14:51 anything about that all of a sudden are
1:14:53 going to spring up on the industry and
1:14:55 it's going to change everything again so
1:14:57 so I think we'll see
1:15:00 um something like the transformer comes
1:15:02 along that creates this step change in
1:15:04 capability and then all the developers
1:15:07 sort of rush to to sort of catch up with
1:15:10 it integrate it and then go beyond it
1:15:12 and then there'll be another one that
1:15:14 does something different better faster
1:15:16 more whatever and then same thing will
1:15:19 happen and
1:15:20 we are definitely in sort of the hockey
1:15:24 stick of innovation
1:15:26 um and we're just starting like we're
1:15:28 just this is nuts
1:15:31 and and it really is Chachi PT that
1:15:34 kicked off this whole frenzy because
1:15:36 that came out you know machine learning
1:15:38 had been around for a few decades
1:15:41 and Chachi PT came out and all of a
1:15:44 sudden 100 million people in six weeks
1:15:47 said oh [ __ ] this is cool
1:15:49 and those were not 100 million machine
1:15:51 learning scientists and researchers and
1:15:53 data people those were like human beings
1:15:56 Like Yours Truly here that went this is
1:15:58 pretty swell
1:16:00 I think I could even use the AI now and
1:16:03 all the scientists are like oh [ __ ]
1:16:05 really them
1:16:08 oh that's a cool one the hell is that
1:16:12 never seen that before
1:16:14 David Gould absolutely killing it with
1:16:17 the action
1:16:20 I'm gonna turn this into a goddamn MPC
1:16:22 Channel yet
1:16:24 I'll do it I'll twerk
1:16:27 and I'll go oh cowboy hat oh cowboy hat
1:16:30 oh cowboy hat
1:16:31 oh Tick Tock hat Tick Tock hat
1:16:38 I can't tempt David it's not it's not
1:16:40 good David it's not good it's not good
1:16:43 don't do it don't do it it serves no one
1:16:47 it serves no one
1:16:53 oh good lord Bagels Bagels
1:16:58 um
1:17:00 how you doing
1:17:02 okay
1:17:04 okay holy crud on Pi dot Ai and Nvidia
1:17:07 does have a good lock on GPS gpus and
1:17:11 likely on newer tpus yeah I just I like
1:17:15 it it baffles my mind
1:17:18 that
1:17:20 all the chip well you know I assume
1:17:23 there was a lot of consolidation in chip
1:17:25 manufacturers but like how did Intel
1:17:28 blow it like how did Intel not get in
1:17:30 the GPU game and and how did AMD blow it
1:17:34 and how did like there are
1:17:37 like it was just like they seated they
1:17:40 said okay [ __ ] it video games can go to
1:17:43 Nvidia
1:17:44 and then the Bitcoin [ __ ] came out
1:17:47 they're like well that's fine that's
1:17:48 just crypto
1:17:49 and now it's like all of AI and like
1:17:53 like
1:17:54 they've had the market cornered long
1:17:56 enough now that like all of the drivers
1:17:59 and all of the software infrastructure
1:18:01 and all of the development
1:18:02 infrastructure is all made for NVIDIA so
1:18:06 it's not even can you make the chips
1:18:07 like AMD can make the chips but they're
1:18:09 soft their their software is all [ __ ]
1:18:11 up if you listen to the there's an
1:18:13 interview with uh with uh Lex Friedman
1:18:16 and George hotz and George is talking
1:18:19 about wanting to use someone other than
1:18:22 Nvidia and he was trying to use AMD
1:18:24 chips and their [ __ ] drivers crashed
1:18:27 when he tried to do machine learning
1:18:28 [ __ ] on him the things that were
1:18:30 designed for machine learning and they
1:18:33 just [ __ ] crashed and he's like
1:18:36 seriously so
1:18:39 that's just amazing to me
1:18:41 just amazing I want to drive an F1 car I
1:18:44 would love to drive an F1 well I don't
1:18:45 honest to God I don't know if I would
1:18:47 want to drive an F1 car I I would love
1:18:49 to at least start with an F4 car like I
1:18:52 feel like the amount of stress just
1:18:55 getting in an F1 car and and what was I
1:18:59 don't know if you ever saw that Top Gear
1:19:00 episode where
1:19:03 um Hammond gets in gets into an F1 car
1:19:06 and he's and he's in the pits and they
1:19:08 train him for like two hours you know
1:19:09 how to do this thing and they get him in
1:19:11 the pit they start the car and he
1:19:13 [ __ ] stalls it like if I stalled an
1:19:15 F1 car I just I would just sort of melt
1:19:18 into the seat
1:19:19 so I think if I started with like an F4
1:19:21 F3
1:19:23 maybe go to an F2 car those F1 cars are
1:19:27 just they're just [ __ ] crazy
1:19:29 [ __ ] nuts be fun to drive a NASCAR
1:19:31 too because they're just [ __ ] Beastie
1:19:34 Wild
1:19:35 just [ __ ] blast that thing around the
1:19:37 two mile track
1:19:41 left turns only
1:19:43 um Lang chain and meta's llm are on top
1:19:47 use within EV uh within Enterprises yeah
1:19:49 I'm probably gonna set up we we've got
1:19:52 I'm going to New York Sunday to talk to
1:19:55 a big Pharma client I think we're gonna
1:19:57 do three pilot projects
1:19:59 um using llms
1:20:01 um quite frankly
1:20:03 some some
1:20:05 projects that are beyond what I
1:20:08 you know know exactly how to do it's
1:20:11 kind of exciting
1:20:12 it's good
1:20:14 um but I'm gonna probably end up going
1:20:15 with llama
1:20:17 uh just for data data safety
1:20:22 um asked yesterday but I didn't hear you
1:20:23 get to it sorry if I missed it what did
1:20:26 you ask oh if Steve Jobs were alive
1:20:28 would he partner with or compete against
1:20:30 Elon
1:20:33 hmm
1:20:37 that's an interesting one
1:20:40 um
1:20:42 I certainly think he would respect him I
1:20:45 you know I think that
1:20:49 you know jobs in Elon
1:20:52 he
1:20:53 you know are both both struggle on the
1:20:57 you know
1:21:00 capacity to be Human Side although job
1:21:03 jobs was pretty pretty
1:21:05 good on it publicly but not good on it
1:21:07 privately
1:21:09 um but but they're both relentlessly
1:21:12 committed to what they believe in right
1:21:15 and so I think I think at a minimum he
1:21:17 would respect Elon
1:21:21 um
1:21:25 I don't know that's a really good
1:21:26 question
1:21:30 I don't know I don't know
1:21:35 there was a someone did a uh was it with
1:21:38 Elon that was it was Joe Rogan and Steve
1:21:40 Jobs someone did a deep fake with Joe
1:21:42 Rogan interviewing Steve Jobs it'd be
1:21:44 it'd be fun to do one with Elon Musk and
1:21:47 Steve Jobs and just you know what would
1:21:49 they say to each other that'd be
1:21:51 something to be Pro decently easy to put
1:21:53 together that's a really good question I
1:21:55 don't know I think they I think that
1:21:57 they would
1:22:03 like musk isn't going after Apple at all
1:22:05 there's nothing there's nothing that
1:22:07 Apple's doing that threatens what musk
1:22:10 is about now if
1:22:14 so so one of the things musk wants to do
1:22:16 with X and why he called it X is he
1:22:18 wants to make it the everything app
1:22:19 right he wants he wants to make it like
1:22:22 WhatsApp and and your payment system and
1:22:25 your communication system and your
1:22:27 publishing system and your money making
1:22:29 system and like he wants to make it the
1:22:31 everything app Apple if they start
1:22:34 getting into autonomous agents they've
1:22:35 already got the payment system they've
1:22:37 already got credit cards they've already
1:22:38 got the entertainment piece they don't
1:22:41 really have a social network
1:22:43 but once they once they start getting
1:22:47 good with the AI stuff I could see
1:22:51 X and apple
1:22:53 um competing in the same space but I
1:22:56 don't see musk getting into Hardware
1:22:57 right and that's primarily what Apple's
1:23:00 about so
1:23:02 I don't know I don't know it's a
1:23:04 fascinating one
1:23:05 damn it I scrolled at the wrong time
1:23:07 heard you go play with llama and was
1:23:11 pumped there was a spot what
1:23:14 Oh no just did it just jump me to the
1:23:16 bottom oh Tick Tock why do you do that
1:23:19 Tick Tock stop here and you made me sad
1:23:22 you made me very sad
1:23:26 when I just start watching with the
1:23:27 misses we just started watching um or no
1:23:30 she she was watching it for a while the
1:23:32 idol
1:23:34 it's about all these bratty musicians in
1:23:36 La it's just very good acting but like
1:23:40 I just I couldn't relate I couldn't
1:23:42 relate there was this very there's this
1:23:44 very spoiled they're also spoiled
1:23:48 standing around in their skimpy little
1:23:50 negligee with their little skinny little
1:23:52 bodies someone needs a burger
1:24:01 I almost moved to LA I was gonna when I
1:24:04 was living in New York we were living
1:24:06 north of the city
1:24:08 and I decided I couldn't take
1:24:10 Westchester that area north of the city
1:24:13 is just weird
1:24:14 couldn't take it anymore and I was gonna
1:24:16 move back to New York City
1:24:18 come out to Boulder or go to La and so I
1:24:22 went to LA and I went to a bunch of
1:24:24 startup meetings and man they were just
1:24:27 [ __ ] weird La is a weird [ __ ] town
1:24:31 everything's about
1:24:33 positioning and Angling
1:24:35 I'd share an idea with someone like oh
1:24:37 man I had an idea just like that man we
1:24:39 should hang out man
1:24:41 everyone wanted to hang out
1:24:43 and what it really was was no one had
1:24:45 any [ __ ] ideas and they just wanted
1:24:46 to hang out with people they thought had
1:24:48 ideas or they thought had power
1:24:51 it was a weird town man she's weird man
1:24:54 you know man you know what I'm saying
1:24:56 man
1:24:57 I don't know you ever been to L.A man
1:25:00 it's just weird man
1:25:03 I'm just trying to buy time to figure
1:25:05 out where I was in the comments here so
1:25:07 just hang with me I'll get there
1:25:10 Ian I think IG is Kyle Shannon huh
1:25:15 oh what's my Instagram I have at Kyle
1:25:18 Shannon and I also have at Kyle Shannon
1:25:20 dreams which is my art project with my
1:25:22 stable diffusion
1:25:25 self-portraits
1:25:29 okay I think I'm almost back to my
1:25:31 questions here all right oh there's the
1:25:34 Steve Jobs one
1:25:38 I scrolled on the wrong time heard you
1:25:40 go play with llama and was pumped there
1:25:43 was a spot oh this is what I was trying
1:25:44 to figure out before
1:25:47 where's the salon I'm local and would
1:25:49 love to join you for the Adobe speaker
1:25:51 uh altitude Tactical Good Name by the
1:25:54 way I'll go 8.5 out of 10 on the name
1:25:57 um
1:25:58 we meet at the Catalyst building in
1:25:59 Rhino so it's uh 35
1:26:02 13 3513 Brighton and we meet from five
1:26:07 to seven and we're on the seventh floor
1:26:10 boardroom
1:26:11 and all that informations if you go to
1:26:13 the Meetup so if you go to the salon.ai
1:26:16 go to the go to the uh
1:26:19 the link tree here and sign up just
1:26:21 become a subscriber to the Meetup we
1:26:23 send out the information on the day of
1:26:26 the meeting as well
1:26:28 so yeah come by say hi
1:26:33 um John Knox not gonna be in person he's
1:26:35 going to be virtual because he's off in
1:26:38 Adobe land
1:26:40 I will make time to get on the salon.ai
1:26:42 good name thank you appreciate that
1:26:46 which AI to make pictures animated bring
1:26:48 them to life thank you all right let's
1:26:49 go do that
1:26:51 we'll go make an image and then we'll go
1:26:53 bring it to Life we'll do both
1:26:55 we'll do both you're getting a bonus
1:26:57 here this is bonus normally I charge
1:26:59 double for this but tonight just
1:27:02 1X not 2x just 1X you only pay the
1:27:06 single amount
1:27:08 of zero
1:27:10 all right
1:27:12 so we are gonna go
1:27:15 here
1:27:16 oh this was this was the the images that
1:27:19 I made for the for the uh article that I
1:27:21 wrote today for LinkedIn
1:27:23 and it was the title of the article was
1:27:26 um
1:27:27 chat gbt gave me superpowers is that
1:27:29 what it was called
1:27:30 hang on
1:27:32 let's have an 80d moment shall we
1:27:36 I'll get back to the other thing calm
1:27:38 down you people
1:27:40 um LinkedIn
1:27:51 um
1:27:52 yeah Chachi BT just give me superpowers
1:27:56 so that was the image I put in my
1:27:58 article
1:27:59 here's the article
1:28:01 Isn't that cool
1:28:03 cool old image and so I did that over
1:28:05 here in Discord so if you've never used
1:28:07 mid Journey you do it in Discord I'm in
1:28:09 the AI Salon Channel
1:28:11 and there's a mid-journey play Channel
1:28:14 and so that's where I made this thing
1:28:17 is that one that's good to animate
1:28:19 actually that one's not bad to animate
1:28:22 um but I'll
1:28:23 uh yeah let's let's do that one I have a
1:28:27 feeling it's going to make these things
1:28:28 in the sky Fly so let's
1:28:30 let's grab this one all right so I'm
1:28:33 gonna open in browser
1:28:35 I'm Gonna Save as
1:28:40 um all right we'll save as there okay so
1:28:42 now I'm gonna go to
1:28:44 runwayml.com
1:28:48 and by the way this is goblin.tools
1:28:50 which is a
1:28:52 I looked at this quickly last night I
1:28:54 played with it a little more today this
1:28:55 is a to-do maker a to-do list maker
1:28:57 where you give it a single to do and
1:29:00 then you hit the little magic wand
1:29:01 button and it'll break that to do down
1:29:03 into all of the requisite steps it sort
1:29:06 of is a project manager in a box
1:29:08 goblin.tools pretty cool
1:29:11 okay
1:29:13 um we're gonna go to runwayml
1:29:17 uh
1:29:21 M I don't know what's going on with the
1:29:22 old throat tonight
1:29:23 I might be getting a little tickle in my
1:29:26 throat
1:29:26 you might be getting it so I went to
1:29:28 Runway ml I logged in and then I went to
1:29:33 um what the hell was it
1:29:35 text to video
1:29:38 but
1:29:40 oh they added new interface here this is
1:29:41 cool
1:29:42 so you can either use a text prompt or
1:29:44 an image prompt so I'm going to use an
1:29:46 image prompt
1:29:48 and I can upload a file so I'm going to
1:29:51 go grab that file that I just
1:29:54 saved
1:29:56 is this it yep
1:30:01 right
1:30:05 all right so there's our picture
1:30:07 and now we're just going to say generate
1:30:10 and I I I saw right before I came on
1:30:15 Live tonight
1:30:16 I think Gen 2 you can now extend a video
1:30:20 up to 16 seconds so you generate four
1:30:23 seconds and then you can add another
1:30:25 four seconds three three more times so
1:30:28 you can do it videos a total of 16
1:30:31 seconds in length
1:30:34 it's exciting isn't it it's so exciting
1:30:39 oh the other thing
1:30:41 oh Manny so I'm showing you this this
1:30:44 might not be what you're looking for you
1:30:46 might be looking for how do I turn like
1:30:48 a talking head into an actual talking
1:30:50 head I'll show you that as well
1:30:54 but let's let's see how this thing does
1:30:56 with our with our Leo pitcher we'll see
1:30:59 how it does our little picture and then
1:31:00 I'll go show you how to make an animated
1:31:01 dude or dudette
1:31:04 dudes doodetts
1:31:06 skinny people fat people hairy people
1:31:10 bald people you can do it all you can do
1:31:13 it all people with no teeth people with
1:31:16 too many teeth
1:31:18 people with 17 fingers in one hand
1:31:21 all right here it is let's see if it got
1:31:24 anything interesting are we ready are we
1:31:26 excited all right
1:31:29 yeah
1:31:31 that was pretty cool
1:31:40 all right I like it you know what I'm
1:31:41 gonna let's extend it four seconds
1:31:44 20 credits per generation
1:31:49 I don't know how many credits I have
1:31:55 so now I should have an eight second
1:31:57 video this is pretty slick
1:32:05 um
1:32:06 yeah welcome to save the day
1:32:12 [Music]
1:32:17 welcome everybody if you're new here my
1:32:19 name is Kyle Shannon this is the AI
1:32:21 learning lab I just ramble on for hours
1:32:24 night after night after night I'm
1:32:27 insufferable welcome
1:32:29 [Laughter]
1:32:32 thank you
1:32:36 we also talk about AI stuff occasionally
1:32:40 I'll sometimes make fun of people
1:32:43 I'll sometimes be even a little mean a
1:32:45 little Saucy a little Saucy sometimes
1:32:48 I'm nice other times I'm saucy
1:32:51 I'm Saucy if you don't like the sauce
1:32:54 get out of the sauce kitchen you know
1:32:56 what I'm saying
1:32:58 you understand oh I think we have video
1:33:01 I think we have eight seconds of video
1:33:03 Let's see if there's any good
1:33:07 yeah I kind of died at the end there
1:33:12 oh wait sorry my hand's in the way
1:33:14 it's cool at the beginning but it kind
1:33:16 of died at the end okay anyway so that's
1:33:18 how you do runway ml image to video
1:33:20 which you know again if you think about
1:33:24 that
1:33:25 you know imagine if you've got an idea
1:33:27 for a film
1:33:29 and you go to Mid journey and you do
1:33:31 your storyboarding well you can come
1:33:32 here and turn those still images into
1:33:35 Cuts you could do a rough cut of a movie
1:33:38 um
1:33:40 probably over a weekend and people are
1:33:42 doing that you can see that stuff on the
1:33:45 YouTubes okay now we're gonna go
1:33:47 somewhere else the other tool to to
1:33:49 animate still images
1:33:52 that I use there's a number of these
1:33:54 that you can use is D Dash ID there's
1:33:57 Synthesia there's hey Jen I think is one
1:34:00 that like I said they're they're
1:34:03 getting their money's worth out of their
1:34:05 uh
1:34:07 PR firm these days I've been hearing
1:34:09 more and more about them okay so let me
1:34:11 log in here
1:34:16 we'll do it with the Google
1:34:18 since they have all my data anyway we
1:34:20 might as well just log into everything
1:34:23 with them
1:34:24 all right so if I say create video what
1:34:28 happens here is you can you can either
1:34:31 upload an image or you can pick from one
1:34:33 of their avatars so they've got some
1:34:35 avatars here
1:34:38 um and then I've got some that I did
1:34:40 like here's a dude that I did that's
1:34:42 like a Pixar version of me
1:34:45 um and then you can either upload audio
1:34:49 that you recorded audio that you
1:34:51 synthesized in some other program like
1:34:53 11 labs or you can just you know type in
1:34:57 copy so let's let's hop over to chatgpt
1:35:02 and I'll say
1:35:04 um
1:35:08 write me a humorous
1:35:11 script in the style of Pixar
1:35:14 that's a short paragraph
1:35:16 talking about
1:35:19 how fun it is living with a chubby dog
1:35:22 that barks too much
1:35:30 all right that's too big
1:35:33 foreign
1:35:33 can you make it shorter
1:35:40 right there you go I'm not even going to
1:35:42 read it we're just gonna go with
1:35:44 whatever the hell it wrote that's what
1:35:45 that's what we get
1:35:47 that's what we get you get what you pay
1:35:50 for
1:35:51 you ain't paid [ __ ] honey you got
1:35:54 nothing on me nothing
1:35:57 you know what I'm saying where is my oh
1:36:00 there it is okay so now so we come over
1:36:03 here pop in that script and then let's
1:36:06 go grab a character so there's the
1:36:09 ladies and here are the gents all right
1:36:13 guy Davis Brandon I guess I could look
1:36:16 like a Brandon Eric Jacob Roger Stefan
1:36:19 definitely not a Stefan Tony Tony
1:36:24 we're gonna go with Brandon
1:36:27 oh Brandon
1:36:28 okay so so we just typed in the script
1:36:31 that
1:36:32 Chachi PT gave us we've got an image
1:36:35 that we made in mid-journey and modified
1:36:37 in Photoshop with generative Phil I did
1:36:39 that blurred background using generative
1:36:42 fill I can show you how to do that too
1:36:44 that's pretty [ __ ] slick and then
1:36:46 we're gonna say generate video
1:36:50 I got 34 credits left I'm spending three
1:36:53 of my credits I'm spending 10 of my
1:36:56 credits here to show you how to do this
1:36:57 come on
1:36:58 tell me that's not something that's
1:37:00 something
1:37:03 it's crazy I think it's just [ __ ]
1:37:06 nuts
1:37:09 all right it's done we have a 31 second
1:37:12 video
1:37:15 in a world where barks and belly rubs
1:37:17 Collide thank you the pudgy perpetually
1:37:20 excited pup who treats doorbells like
1:37:22 rock concerts and breakfast like a
1:37:23 five-star feast
1:37:25 living with biscuit is a symphony of
1:37:27 surprises an unpredictable dance of Joy
1:37:30 that'll have you laughing till your
1:37:31 sides ache
1:37:32 who needs an alarm clock when you've got
1:37:34 biscuits booming good morning Serenade
1:37:37 sure they might be a little extra in the
1:37:40 noise department but every woof and
1:37:42 wiggle fills life with more love and
1:37:44 laughter than you ever thought possible
1:37:47 yeah that was cringy but you know
1:37:50 what did it take us 10 seconds no you
1:37:53 know 20 seconds to generate the script
1:37:55 and then we had to generate the thing
1:37:57 so here here's my thought on these
1:38:00 things
1:38:01 um
1:38:03 okay
1:38:07 my thought on these things is while they
1:38:10 are not ready for prime time now
1:38:13 if you go back a year
1:38:15 and look at what image generation looked
1:38:18 like a year ago and compare it to what
1:38:20 it looks like today
1:38:22 it's not in the same neighborhood
1:38:25 um actually with images it was probably
1:38:27 close to a year and a half ago but like
1:38:29 a year and a half ago like the images
1:38:30 were laughably bad
1:38:33 um all this stuff is gonna get
1:38:35 dramatically better dramatically quick
1:38:38 so
1:38:40 um yeah I mean and that's just using
1:38:41 like off-the-shelf low level [ __ ]
1:38:45 um
1:38:48 you know some of the uses for this stuff
1:38:49 that are super practical is
1:38:52 um
1:38:53 like I do a lot of work in healthcare
1:38:55 and we're going to be able to take
1:38:57 videos of patients speaking English
1:39:02 and have them speak any language that
1:39:05 our client wants in their voice in a
1:39:08 different language you know
1:39:13 um with with the right inflections and
1:39:15 all that sort of stuff
1:39:16 uh grammar all that sort of stuff in in
1:39:20 relatively short order and while it's
1:39:22 kind of weird now it's not going to be
1:39:23 weird two years from now
1:39:26 um so so that kind of stuff is coming
1:39:28 wild huh isn't that crazy and thank you
1:39:31 for whoever ever dumped a pile of coins
1:39:33 down the screen that was cool appreciate
1:39:35 that yep the old days it's crazy times
1:39:38 exciting times it is I so agree with
1:39:41 that too what AI rule
1:39:44 wait what AI rule will it be in digital
1:39:48 currency
1:39:52 Wasim I don't know quite what you mean
1:39:57 what AI rule will it be in digital
1:40:00 if you can clarify your question I'm not
1:40:02 quite sure what you're asking I like
1:40:03 llamas I like totals
1:40:06 I'm gonna go show that
1:40:09 living on the edge
1:40:19 back here live at the waterfront village
1:40:21 with my friend the zombie Jonathan
1:40:23 you're looking back Jonathan just got an
1:40:25 awesome face paint job what do you think
1:40:26 I like turtles
1:40:30 the Waterfront Village open for the next
1:40:33 11 days
1:40:34 so here are three
1:40:38 oh it cracks me up every [ __ ] time
1:40:47 good night all good night Denise uh who
1:40:50 is this Kyle guy anyway what are your
1:40:52 qualifications sir sir
1:40:54 I am demanding to know your
1:40:56 qualifications what makes you think you
1:40:59 can speak upon artificial intelligence
1:41:01 with any credibility whatsoever
1:41:05 none none good day to you sir good day
1:41:11 all right carrying on you mentioned Ai
1:41:13 and Retail and how it will be
1:41:15 interesting can you elaborate
1:41:18 how it will affect retail um
1:41:24 I think in lots and lots and lots and
1:41:26 lots and lots of ways
1:41:28 um
1:41:33 certainly on the e-commerce side
1:41:36 I think I think things like you know
1:41:40 virtually trying things on in in really
1:41:43 seamless ways
1:41:45 dynamically generating sites and
1:41:48 recommendations for products so I could
1:41:51 just be in some sort of conversation and
1:41:53 say oh that's kind of an interesting
1:41:55 thing or I kind of like that look and it
1:41:57 just goes and finds the products and and
1:42:00 identifies
1:42:02 all of the different sources where I
1:42:04 might get product from and and sort of
1:42:06 custom builds me a website from multiple
1:42:08 sources
1:42:10 um so I think it's going to be
1:42:13 amazing on that front
1:42:16 um in physical retail you know you're
1:42:19 starting to see things like fast food
1:42:22 chains experimenting with virtual
1:42:25 you know synthesized people at the drive
1:42:27 up window
1:42:29 um
1:42:31 you know imagine imagine going into
1:42:35 a store and having questions about
1:42:37 something and instead of having some
1:42:39 Clueless
1:42:41 you know temporary employee who doesn't
1:42:44 really want to be there you've got some
1:42:46 virtual employee that no can answer any
1:42:48 question about any product from any time
1:42:51 in history
1:42:53 um I think you could start to see
1:42:54 interesting interactions that way I
1:42:57 think if you if you look at
1:42:59 um
1:43:00 if you kind of play Apple's spatial
1:43:03 Computing you know their Vision Pro
1:43:05 goggles if you play that out five years
1:43:07 and assume that the form factor of that
1:43:09 gets smaller and actually wearable out
1:43:12 in public and things like that the
1:43:14 augmented reality stuff you could do in
1:43:17 retail is going to be insane that's
1:43:18 that's all predicated on machine
1:43:20 learning and AI
1:43:22 um
1:43:23 I don't know dynamic pricing
1:43:26 um personalization I I I think we've
1:43:29 we're just scratching the surface of
1:43:31 personalization
1:43:34 um I'm working on a project right now
1:43:36 doing personalized images and print on
1:43:39 demand stuff uh which is really
1:43:41 interesting so lots and lots and lots I
1:43:44 don't I don't know that world super well
1:43:47 um so it'll be interesting to see
1:43:53 [Music]
1:43:54 it's easier to learn a platform than
1:43:57 learning python
1:43:59 and you don't even really need to Learn
1:44:01 Python now
1:44:03 just have it write it for you now people
1:44:06 that know python are like well it
1:44:07 doesn't write it as good as I do it
1:44:09 writes it in a very sloppy way and that
1:44:11 may be true right
1:44:13 if you want something bulletproof hire
1:44:15 an engineer if you want to get something
1:44:16 done
1:44:18 you can have to do this this will get it
1:44:20 done it might not do it as elegantly but
1:44:24 it will
1:44:26 again you know I give these machines
1:44:29 Grace
1:44:30 your son reads manuals Gen X doesn't
1:44:32 need no sticking manuals exactly I don't
1:44:35 know how he got that like [ __ ] Lego
1:44:38 manuals reads them every page
1:44:41 Ikea manuals Ikea manuals you have to
1:44:43 read those things are crazy
1:44:45 but like I'm trying to think what it was
1:44:47 it was something it literally had like
1:44:49 two bolts and a nut
1:44:52 just like what are we read in the manual
1:44:55 for
1:44:56 uh I need you to record my outgoing
1:44:58 message from my phone you can make money
1:45:00 watch out every day
1:45:04 we could do that here just everyone
1:45:06 we'll we'll do a coordinated screen
1:45:09 recording and then I'll do you can make
1:45:11 money with chat gbt and you can have it
1:45:13 for your for your outgoing
1:45:19 for your outgoing voice call
1:45:23 that's really funny I paid a guy for a
1:45:25 year to create e-learning I built a
1:45:28 better version in a day with Chachi VT
1:45:31 crazy
1:45:33 that's hilarious so so that's actually a
1:45:36 really good point TM
1:45:38 um TMM wait TMM keep Rising I want to
1:45:42 know what that means I think that's a
1:45:43 cool name but I don't know what it means
1:45:45 um
1:45:47 so everyone [ __ ] about these things
1:45:49 well it doesn't write as good as a human
1:45:51 well it might not write as good as a
1:45:53 good human but like
1:45:55 the amount of marketing professionals
1:45:59 that I've hired in my day that have been
1:46:02 [ __ ] incompetent
1:46:05 is staggering
1:46:08 what's really interesting about that is
1:46:10 so you did it for for an e-learning
1:46:12 thing
1:46:13 um
1:46:15 it's gonna be everything
1:46:16 it's going to be every SAS platform it's
1:46:19 going to be every
1:46:21 um you know learning
1:46:24 course right all of the educational
1:46:26 courses that people have done you're
1:46:28 going to be able to just make your own
1:46:29 learning course dynamically on the Fly
1:46:32 customized for you at the level you're
1:46:34 at right now that is going to be better
1:46:36 than any of the [ __ ] out there
1:46:39 and and then
1:46:40 where I go with that is okay what are
1:46:42 the implications of that I don't know
1:46:45 like
1:46:48 who gets an area right is it all just
1:46:51 everything goes to 20 Buck a month chat
1:46:55 GPT like services that end up doing
1:46:57 everything
1:46:59 or are there Specialists out there
1:47:01 that's you know have specialized data
1:47:04 sets where you can make an e-learning on
1:47:05 very specialized data and they're and
1:47:08 they're charging you to the data set
1:47:10 rather than the mechanism to make the
1:47:12 course
1:47:14 but I just I just feel like the the the
1:47:17 economics and the customer dynamics of
1:47:21 and this goes back to retail as well
1:47:23 like every interaction is going to be
1:47:26 redefined by these tools because we're
1:47:28 all kind of assuming that the things
1:47:30 that we do today are the things that
1:47:32 we'll be doing in the future
1:47:35 what's personalization on a streaming
1:47:37 service well they kind of listen to what
1:47:40 you like and you can tell them
1:47:41 explicitly what you like and then they
1:47:43 give you recommendations for playlists
1:47:47 of music that other people wrote
1:47:50 okay
1:47:51 well
1:47:53 why does it have to be music that other
1:47:55 people wrote what if it's a playlist
1:47:57 that just plays music written for you
1:48:00 in the mood you're in right now
1:48:03 for what you've got going on today
1:48:07 and it's all original music and they'll
1:48:09 sprinkle in you know 10 of historically
1:48:13 recorded songs
1:48:18 songs from the Before Time
1:48:20 for a little nostalgia and then
1:48:23 everything else is personalized and they
1:48:26 can tell because your GLA glasses micro
1:48:29 movements in your irises and and your
1:48:33 earbuds are reading the micro
1:48:35 temperature fluctuations in your in your
1:48:37 ear canal it understands what music you
1:48:40 like and what music you don't like
1:48:41 before you even hear it
1:48:46 so all of the music you listen to gives
1:48:49 you exactly the feeling you were hoping
1:48:51 to have
1:48:55 I mean what's weird about talking about
1:48:57 that is we're not that far away from it
1:49:02 it's cuckoo it's cuckoo don't you think
1:49:10 all right how are you
1:49:12 oh how are you doing I like your
1:49:15 conversation thanks for your life thank
1:49:16 you James James come out I appreciate
1:49:19 that cool name come out
1:49:22 you have lovely nostrils wow thank you
1:49:25 very much I I was my earlier lives I had
1:49:29 my camera too low I've now got
1:49:31 everything set up on on large boxes and
1:49:33 things like that so my camera was lower
1:49:36 so it was shooting up so
1:49:38 um I I didn't get any you'll have lovely
1:49:40 nostrils most of what I got was you
1:49:42 should shave your nose hair
1:49:45 foreign
1:49:50 it did get me to shave my nose here
1:49:54 but thank you for the nostril comment I
1:49:57 don't know that I've ever had a lovely
1:49:58 nostril comment
1:50:00 this is how I spend my break from class
1:50:02 thank you Thelonious
1:50:05 oh why can't I screenshot your lives
1:50:08 security policy no I think that's a
1:50:10 local setting someone else asked that
1:50:11 the other night I think that's a local
1:50:13 setting on your side
1:50:15 um I'm not blocking it thank you for the
1:50:17 for the hearty Hearts
1:50:18 um I'm not blocking anyone from from
1:50:20 screenshotting this
1:50:23 I should put I should put a little
1:50:25 background back here screenshot my lives
1:50:27 one dollar
1:50:29 a little little coin slot where you can
1:50:31 drop in nfts
1:50:32 [Laughter]
1:50:34 oh good lordy lordy lordy oh my God
1:50:39 you've switched cams right about the
1:50:41 time you said they'll shut me down and I
1:50:44 thought yep oh I know what are you gonna
1:50:47 do gazuntite thank you wait zai
1:50:51 yeah I think Gesundheit yes
1:50:55 I'm sure wait it's sure been glitchy for
1:50:58 me the past several days yeah my
1:51:01 internet
1:51:03 the technical term is poo poo it'd be in
1:51:06 a poo poo my internet being a poo poo
1:51:08 poo poo
1:51:10 we can't watch the TV
1:51:12 we can't surf on the internet the boys
1:51:15 can't play the video games and I can't
1:51:17 do my Ticky talks
1:51:22 you know what I'm saying you know
1:51:26 you know and do you know what I'm saying
1:51:28 how do you
1:51:30 hey anyone know how to turn this thing
1:51:32 off
1:51:33 huh
1:51:35 how do you do this
1:51:41 NC all right
1:51:44 cool you are on the kaggle data sets
1:51:47 this live is the best awesome thank you
1:51:50 Winston
1:51:52 thank you for the tip on the chat gbd
1:51:54 plugins you're welcome not sure which
1:51:55 tip I gave but cool if I want to learn
1:51:58 how to code using chat GPT where should
1:52:01 I start
1:52:02 um I would start with chat you PT
1:52:11 like I so so I guess I mean I would I
1:52:14 would ask a couple of things if
1:52:18 I would start with what do you want to
1:52:20 accomplish
1:52:22 and then
1:52:24 why do you want to learn to code
1:52:27 because these tools are going to get
1:52:29 more and more sophisticated so you might
1:52:31 not
1:52:33 if you want to learn to code to
1:52:35 understand conceptually how coding works
1:52:37 and how you think about coding and you
1:52:41 know Frameworks and languages and syntax
1:52:43 and all that sort of stuff just so you
1:52:45 can get your head around it so you can
1:52:46 understand what's actually happening
1:52:50 um
1:52:53 that's one thing if you want to learn
1:52:55 the code to get a coding job I think
1:52:58 coding's jobs moving forward are likely
1:53:02 going to be
1:53:03 coders that are also AI literate right
1:53:06 they're going to want coders that are
1:53:07 increasingly efficient
1:53:10 and understand how to use these tools
1:53:13 so so I'd say step one would be get
1:53:16 really clear on what your goals are
1:53:19 and then I mean seriously so
1:53:22 if I were you if you've got the 20 bucks
1:53:24 a month pay for gpt4
1:53:27 at open AI
1:53:30 and and just start working in chat gbt4
1:53:32 if you can't afford that go to Bing so
1:53:34 Bing is gpt4
1:53:36 and you can you can do this the same
1:53:38 [ __ ] there
1:53:40 but but the open AI one is just it's
1:53:43 it's just less restrictive and it's
1:53:46 better
1:53:48 um until further notice and I'll at the
1:53:50 point at which one of these things gets
1:53:51 better than gpt4 I'll flop the order but
1:53:54 that's it right now
1:53:56 and then what I would do is I would go
1:53:58 there and I would say
1:53:59 you know sort of here's a little bit
1:54:01 about me
1:54:02 here's what I'm trying to learn here's
1:54:04 my goals
1:54:06 um
1:54:07 what's the best way for me to learn this
1:54:10 skill set and like what do I need to
1:54:12 learn what are the things that I don't
1:54:13 know that I need to learn
1:54:15 and have that thing put together a
1:54:17 learning plan
1:54:19 for you to do it and you could even say
1:54:22 I want you to teach me how to in fact
1:54:24 let's let's go do this because this
1:54:26 could be this could be interesting this
1:54:29 is something I haven't done before that
1:54:30 I think could be kind of fun
1:54:33 all right so you're gonna go to church
1:54:35 EBT you're going to go to Just default
1:54:38 um chat gbt and you're gonna say Okay
1:54:41 um
1:54:44 I'm just getting started with coding
1:54:47 and I want
1:54:49 to learn how to program python
1:54:54 but I literally
1:54:56 don't know what step to take first
1:54:59 period
1:55:01 could you tell me the broad categories
1:55:05 of things that I should learn
1:55:09 so that I can become a good python
1:55:12 programmer period
1:55:20 it can be fun and rewarding especially
1:55:22 with python basics of python introdu
1:55:25 introduction what is it and why use it
1:55:28 installing it and setting up the
1:55:30 environment running a simple hello world
1:55:32 program data types and variables numbers
1:55:35 integers floats booleans basic operators
1:55:38 blah blah intermediate python functions
1:55:41 okay so this is this is like a
1:55:44 programming manual right so for me
1:55:47 because I don't like reading [ __ ]
1:55:49 manuals
1:55:50 this doesn't work for me this is driving
1:55:52 me [ __ ] crazy so number seven is
1:55:55 projects and practical application so
1:55:58 what I'm going to say is I'm going to
1:56:00 tell it I don't like its approach
1:56:04 I have to say I don't like your approach
1:56:07 period
1:56:08 this looks too much like a manual period
1:56:12 can you teach me Python program with
1:56:15 Hands-On use case driven projects
1:56:18 rather than
1:56:20 functions
1:56:23 let's see what it does to that it should
1:56:25 apologize
1:56:33 certainly learning through Hands-On
1:56:34 projects is a great approach number one
1:56:37 calculator build a simple calculator
1:56:39 okay this is good
1:56:40 build assume for calculator that can
1:56:42 perform addition subtraction blah blah
1:56:45 blah okay
1:56:48 so I like this I like this approach it's
1:56:49 telling me what skills I'm going to
1:56:51 learn on these projects I can get my
1:56:52 head around it this feels more like when
1:56:54 you get an Arduino the the the books
1:56:57 that come with arduinos are like little
1:56:59 projects make the Blinky lights make the
1:57:01 this make the you know TV interrupter
1:57:05 thingy
1:57:06 so I like this all right this is a good
1:57:08 approach this is making me happy
1:57:10 all right so now I'm gonna say oops
1:57:12 sorry
1:57:14 happy coding okay
1:57:17 let's
1:57:20 start with number one period
1:57:24 I want to get right into the programming
1:57:27 but I want you
1:57:29 to teach me step by step
1:57:33 period I don't want you to move on to
1:57:35 the next step until I've said okay I
1:57:39 understand the previous one
1:57:42 do you understand
1:57:47 absolutely I understand your approach
1:57:49 let's jump in silver simple calculator
1:57:52 step one setting up
1:57:54 set up python if you haven't already
1:57:56 installed python you can download it
1:57:58 from the official website here
1:58:01 two your first python script okay
1:58:05 so this is this is not listening to what
1:58:07 I said
1:58:08 so I'm going to stop generating I'm
1:58:10 going to say
1:58:12 I don't want you to give me all the
1:58:14 steps at once I want you to give them to
1:58:17 me one by one
1:58:19 period
1:58:21 do not move on to step two until I've
1:58:24 completed step one period
1:58:29 I apologize for the oversight step one
1:58:31 setting up python check and see if it's
1:58:33 installed open your terminal or command
1:58:35 prompt type python version or python3 so
1:58:39 I'm going to say
1:58:40 um
1:58:42 I'm on a Mac running OS X period can you
1:58:47 customize these instructions for my
1:58:50 specific setup
1:58:54 of course
1:58:57 ensure Python's installed on your Mac
1:58:59 open your terminal
1:59:02 so so anyway
1:59:04 um that's how I that's how I would do it
1:59:06 right just
1:59:07 think of
1:59:13 think of these tools as the most capable
1:59:20 smartest
1:59:23 intern you've ever had and if you're
1:59:25 like I'm too young to have an intern
1:59:27 well you have one now
1:59:30 they will do anything you ask them to do
1:59:32 it's what they're programmed to do they
1:59:34 want to give you what you need it's it's
1:59:35 what they're all about
1:59:37 um so
1:59:39 I would just do some version of what I
1:59:40 just did there but but before you just
1:59:43 jump in and start learning [ __ ] I would
1:59:45 get clear on what what you want to learn
1:59:48 and why you want to learn it because
1:59:50 learning to code might not be
1:59:53 the best option but it might be so get
1:59:56 clear on that first and then you could
1:59:59 even use chat gbt as kind of a career
2:00:01 coach to say what should I be doing
2:00:03 given that all of this shit's going to
2:00:06 be automated does it make sense for me
2:00:08 to learn coding right now maybe it does
2:00:10 maybe it doesn't and it would probably
2:00:12 have some pretty good Insight on that
2:00:16 all right wow the spreadsheet example
2:00:20 shared
2:00:22 the uh oh the code interpreter thing
2:00:25 yeah it's crazy
2:00:27 all right this is the future you say you
2:00:29 don't know about data analysis but
2:00:31 you're teaching others AI Eda oh yeah
2:00:34 you know I'm fairly deep into the AI Eda
2:00:37 if you know what I mean
2:00:39 if I knew what AI Eda was I would tell
2:00:42 you I mean I definitely know but I don't
2:00:44 you know what it is I don't want to just
2:00:46 give away the answers I want I want you
2:00:49 to understand them on your own
2:00:52 that's how I roll
2:00:56 welcome everybody it looks like we got
2:00:58 some new folks here my name is Kyle
2:01:00 Shannon this is the AI learning lab
2:01:02 uh we have fun here we have fun we just
2:01:05 hang out we talk we talk we have talks
2:01:09 we we go back and forth we have we have
2:01:12 witty banter we have witty banter we
2:01:15 talk about all these here tools that
2:01:17 one's that's chat GPT have you used
2:01:19 Chachi BT I don't know if you knew this
2:01:21 but you can make money
2:01:25 right there is chat gbt and then you got
2:01:28 Claude this is a bunch of pesky nerdo
2:01:30 Wells that used to work here and they
2:01:32 left and said I don't like how you do
2:01:34 [ __ ] we're gonna go do our own
2:01:36 and so they did and they're pretty
2:01:38 [ __ ] good and then you got Bing
2:01:41 Bing was little Microsoft remember
2:01:44 Microsoft remember Bing how cute it was
2:01:47 that they were trying to compete with
2:01:48 Google for years
2:01:50 and then all of a sudden they teamed up
2:01:52 with old Sam Altman up here and kick
2:01:54 Google's ass
2:01:57 and Google's like wait we're losing to
2:02:00 these guys
2:02:03 I think you got barred that's Google
2:02:05 Google's starting to get their [ __ ]
2:02:07 together
2:02:08 after nine months
2:02:10 nine months of like oh [ __ ]
2:02:13 they released the Kraken what do we do
2:02:17 I guess we better use that technology we
2:02:19 invented huh
2:02:23 and then oh Mark Zuckerberg was like uh
2:02:26 We're Not Gonna compete on this [ __ ] why
2:02:28 don't we just give it away here you go
2:02:30 so that's that's that one that's llama
2:02:34 and then you got inflection and you got
2:02:36 quora doing their thing with question
2:02:39 and answers that Poe
2:02:42 just a wild time it's wild
2:02:45 it's a wildfire of innovation
2:02:49 why do I have a generically Southern
2:02:51 accent
2:02:54 why do I talk like this
2:02:57 no one knows no one knows
2:03:05 All Right Moving On
2:03:09 is this 3.5 or 4 I can't hear you hmm
2:03:14 this is the exploratory data analysis of
2:03:17 the future it is it's insane that's code
2:03:19 interpreter that was code interpreters
2:03:21 GPT 4 Forward
2:03:24 3.5 is free and it's super fast but it
2:03:27 doesn't have any of the fancy [ __ ]
2:03:30 four is the really good one it can code
2:03:33 really well it can write really really
2:03:35 well yeah why really well
2:03:38 it can wait really well and uh
2:03:41 it can do code interpreter and you can
2:03:43 do plugins and plugins are interesting
2:03:45 and I think both of them can do custom
2:03:47 instructions but maybe not that might
2:03:50 just be gpt4 are you a paid actor well
2:03:53 I'm an actor
2:03:54 and I've been paid
2:03:57 have I been paid for my acting
2:04:04 uh
2:04:06 I have
2:04:10 not in a while not a good long while
2:04:13 although I act like the CEO and I get
2:04:15 paid for that that counts doesn't it
2:04:21 which is the best for generating
2:04:23 animation I showed that earlier so I I
2:04:28 don't know I don't know
2:04:31 um
2:04:32 I'm not I'm not digging deep deep into
2:04:35 the into the animation world I I think
2:04:37 if if you're talking about video like
2:04:39 image to video
2:04:41 um then runwayml but but there's there's
2:04:44 a bunch of them out there and
2:04:46 um if you're talking about like the
2:04:48 synthesized voice to animation thing
2:04:51 um
2:04:53 I think did but it's expensive and so is
2:04:57 Synthesia
2:04:59 and so is there's another one out there
2:05:01 oh hey Jen
2:05:03 h-h-e-y-g-e-n
2:05:07 um I don't know how much that one is
2:05:08 though but I I would imagine it's
2:05:10 expensive what I'm waiting for is sort
2:05:12 of the equivalent of stable diffusion
2:05:14 and I know stable diffusion does some
2:05:15 animation but it would be nice if
2:05:17 someone came out with something like
2:05:19 runwayml or whatever that wasn't so
2:05:22 expensive but I the the reason they're
2:05:24 expensive is likely that it's burning
2:05:27 massive amounts of GPU cycles and it's
2:05:30 probably expensive for them to run it so
2:05:33 until that [ __ ] gets
2:05:35 um
2:05:36 optimized and efficientized
2:05:41 and cheap then it ain't going to be
2:05:43 cheap
2:05:45 we use all these GPU cycles and pass the
2:05:48 costs on to you it's the American way
2:05:51 welcome to capitalism
2:05:53 [Laughter]
2:05:57 shouldn't the government be giving us
2:05:59 access to this oh that's cute that's you
2:06:02 you're a card it's a government
2:06:07 oh good lord he set the government and
2:06:11 competence in this same sentence Oh
2:06:13 that's hilarious oh
2:06:16 oh good lord
2:06:19 all right quit trying to make Watson
2:06:21 happen I know it's so true it's so true
2:06:24 like why resurrect that brand why not
2:06:27 just call it like IBM X or IBM does AI
2:06:31 or I'm an AI or at IBM or anything other
2:06:34 than Watson X
2:06:37 Watson X is like trying to say I'm
2:06:40 geriatric hip
2:06:46 but whatever it's they they try
2:06:50 oh my God the early days of the web the
2:06:53 uh the the web group at IBM they started
2:06:56 a Consulting service in a in a web group
2:06:58 and oh my God they were so bad for so
2:07:00 long they got their [ __ ] together
2:07:02 eventually but the early days the early
2:07:05 days that was special I suggest paying
2:07:08 for gpt4 I think it's great I I do too I
2:07:11 think it's worth it GPT is worth the 20
2:07:13 bucks in my opinion what's that little
2:07:15 little wait a little Raven
2:07:21 shark nest
2:07:23 little Raven's heart Nest I like the
2:07:26 name I don't know what it means
2:07:28 um
2:07:29 yeah so I don't I'm not affiliated with
2:07:32 these guys in any way so I'm
2:07:34 when I tell you I think it's worth you
2:07:37 paying for chat EBT like why it's on my
2:07:39 lives for the entire time
2:07:42 um is is because these are the ones that
2:07:44 I think are worth
2:07:46 um paying attention to right now
2:07:49 um I'll remove I I'm close to removing
2:07:51 Poe and I I just but I but the the fact
2:07:55 that it does Bots and the fact that you
2:07:57 can play with multiple models on it
2:07:59 makes it kind of worth having there and
2:08:01 Pi is is
2:08:04 specialized enough right now that it's
2:08:06 it's almost not worth putting on there
2:08:08 but
2:08:09 they just raised 1.3 billion dollars so
2:08:11 I have a feeling they're going to be a
2:08:12 major player so I think all these are
2:08:14 worth playing with right now but I do I
2:08:16 do agree that and and I pay for uh open
2:08:19 AI chat GPT four if it feels worth it to
2:08:24 me
2:08:25 yeah is there an AI that does video
2:08:28 editing there's
2:08:30 um d-script is one so so the problem
2:08:33 with all the AI video tools right now is
2:08:36 they're all kind of like dscript where
2:08:38 you have it's like a page of discrete
2:08:40 tools that all do like specialized
2:08:42 things
2:08:44 they're starting to roll some of those
2:08:45 features into Premiere and
2:08:48 things like that but
2:08:51 not that I know but what I would tell
2:08:53 you is
2:08:54 go to
2:08:57 futurepedia.io
2:08:59 once once we start getting into
2:09:01 specialized tools like video AI assisted
2:09:04 video editing editing tools there are
2:09:07 probably a lot out there that I don't
2:09:09 know about but if you go to
2:09:10 futurepedia.io that's just a big old
2:09:12 directory of AI tools they have a video
2:09:15 section
2:09:16 just go start clicking on logos I would
2:09:19 sort by popularity for sure because
2:09:24 I would be willing to wager that there
2:09:26 are 50 video tools or more
2:09:29 and ninety percent of them are going to
2:09:31 be steaming piles of poo
2:09:34 and you might not even find out how big
2:09:36 a steam and pile of poo they are until
2:09:38 you pay your you know sign up for our
2:09:41 free seven day trial and then you sign
2:09:42 up and like oh [ __ ] this is a piece of
2:09:44 [ __ ] God yeah rigm frigm Rock them suck
2:09:49 em and then you forget to
2:09:53 cancel your subscription and three
2:09:55 months later you have to sell the
2:09:58 children it's just not good so just be
2:10:01 careful there be careful out there
2:10:02 people
2:10:04 there are people asking for 20 bucks a
2:10:06 month everywhere you look
2:10:09 by the way you can subscribe to AI
2:10:12 learning lab for 20 a month welcome
2:10:14 no you can't it's a joke it's comedy
2:10:17 it's comedy comedy
2:10:19 All Right Moving On Moving on something
2:10:23 is happening with my voice and I don't
2:10:25 know what it is it's not God it's bad
2:10:30 maybe I should start a cartoon wouldn't
2:10:32 it you know if I had if I had if I were
2:10:35 clever
2:10:36 I would actually like well no if I were
2:10:38 clever and didn't have ADD I could write
2:10:40 down all the voices I do and then match
2:10:42 them to little AI heads I stick on
2:10:44 myself and just quickly switch between
2:10:46 them so when I have a hairbrained voice
2:10:48 I want to do I like pop on the character
2:10:50 it'd be like a whole little show
2:10:53 my own little version of The Simpsons
2:10:55 without the comedy or the good voice
2:10:58 acting
2:11:02 all right
2:11:04 moving on
2:11:06 working in that cowboy hat would have
2:11:09 dire consequences I agree
2:11:12 I think I would get banned for life
2:11:15 it's like twerking on Elder talk
2:11:21 gear is a hoot yeah it is it's good uh
2:11:24 you have exactly right on the history
2:11:26 and what is going on in designing of
2:11:29 gpus oh that's interesting yeah I like
2:11:31 it
2:11:33 I just like how
2:11:39 corner of the market it's just
2:11:41 mind-blowing to me
2:11:45 like like was it a conscious choice that
2:11:47 Intel went well you know it's all about
2:11:50 CPUs and gpus gpus are the garbage of
2:11:55 the CPU business am I right Bob
2:12:01 until gpus are the thing we need
2:12:06 like how do you [ __ ] that up
2:12:10 we're not in the GPU business
2:12:12 that's the trailer trash hey what time
2:12:16 what's our tee time let's go
2:12:20 sorry I'm sorry they're they're uh their
2:12:22 market cap is what a trillion dollars
2:12:26 just I
2:12:30 when's the board meeting
2:12:32 I can't breathe
2:12:38 that was that was my Intel executive
2:12:40 finding about out about nvidia's
2:12:43 trillion dollar market cap in June
2:12:48 right before the golf game
2:12:50 oh
2:12:52 Tick Tock just jumped me to the bottom
2:12:55 oh
2:12:57 uh and gift fundraising uh can we talk
2:13:00 automations yes I will do an automation
2:13:02 for you I promised I would do that
2:13:04 um oh my God the turtle guy made a new
2:13:07 video
2:13:08 um let's see uh by the way my name is
2:13:11 normally stevo42 I remember you I'll
2:13:14 change it back in five days and counting
2:13:18 uh have you seen any accounting
2:13:20 bookkeeping AI products uh I have not
2:13:22 seen anything on bookkeeping I would
2:13:25 assume
2:13:26 that quickened everyone is gonna have uh
2:13:29 AI layers sitting on top of their [ __ ]
2:13:32 you know what's going to be amazing when
2:13:33 the tax preparation folks put an AI
2:13:36 layer on top and the AI has just read
2:13:38 all the documentation and you just it
2:13:41 just looks at your [ __ ] and goes let me
2:13:43 prepare that for you blimp
2:13:47 move this slider what's your risk
2:13:49 tolerance for deductions
2:13:52 um
2:13:52 so for bookkeeping I'd go to
2:13:54 futurepedia.io and see if there's
2:13:56 anything in financial services
2:13:58 what do you do for work I'm a founder
2:14:00 and CEO of a company called storyvine
2:14:06 um
2:14:08 all right I'm gonna I'm gonna jump down
2:14:11 to the automation one I lost my place in
2:14:13 the
2:14:14 um
2:14:16 Amelia's wife as a consumer I'd prefer
2:14:19 to just to talk to AI with a generated
2:14:22 video rather than just a text chat yeah
2:14:24 and I think we'll get there soon again I
2:14:26 have a sneaking suspicion Apple's gonna
2:14:29 come out with that sooner than sooner
2:14:31 than later if you looked at the if you
2:14:34 looked at the um
2:14:36 if you watch The Vision Pro
2:14:40 announcement
2:14:44 pay attention to the section where they
2:14:46 talk about being on a zoom call
2:14:50 it's wild
2:14:52 so when you first get the goggles you
2:14:54 flip them around backwards and you face
2:14:56 them it yourself and it scans your head
2:14:57 and you know you scan it at all angles
2:15:00 and so it makes a model of your head and
2:15:02 then you put on these giant stupid ass
2:15:04 goggles
2:15:05 that have screens on the outside so
2:15:07 people can see your eyes it like
2:15:09 projects it takes the cameras and
2:15:11 projects them onto the screen on the
2:15:12 outside of these things they're [ __ ]
2:15:14 wild Apple's just insane
2:15:16 and then
2:15:18 because they've got your synthesized
2:15:20 face and because they've got cameras
2:15:21 pointing down at your mouth and they've
2:15:23 got cameras pointing at your eyes when
2:15:25 you get on a zoom call
2:15:27 you see like giant screens of the other
2:15:29 people in your living room
2:15:31 and then they see you they see your
2:15:34 whole face talking and your eyes moving
2:15:37 and your eyebrows moving it's all
2:15:38 synthesized and they're sort of in real
2:15:42 time composing all of these different
2:15:44 camera views and reconstructing your
2:15:47 face with your voice with your lip
2:15:49 synced and everything
2:15:51 so
2:15:53 they've built that [ __ ]
2:15:56 so that's the complicated solution so
2:15:58 now they just need to take a scan of
2:16:00 your face with their little lidar
2:16:01 cameras that are all over these phones
2:16:03 now
2:16:04 and I think I think we'll be able to be
2:16:07 be able to chat with personas of people
2:16:12 probably soon I think Apple's gonna be
2:16:14 big in that game I think they're going
2:16:15 to do it well
2:16:17 all right
2:16:18 um
2:16:19 you wanted to look at automations let's
2:16:21 go let's go do an automation let me let
2:16:23 me get my head around what we want to do
2:16:28 because I've done this before but on
2:16:29 alive this could be boring as [ __ ] so
2:16:32 all right so I want this not to be
2:16:34 boring okay how are we gonna do this
2:16:36 hold
2:16:43 all right let me get let me get a couple
2:16:45 of tools set up here
2:16:53 [Music]
2:17:08 foreign
2:17:11 [Music]
2:17:14 all right
2:17:20 [Music]
2:17:23 who's hitting Hearts this is not good
2:17:26 enough singing for hearts this this this
2:17:28 deserves no Hearts he deserves no hearts
2:17:31 for that kind of singing a trash
2:17:35 incompetent off-key off temba good Lord
2:17:41 you people are simpletons he's not
2:17:43 entertaining there's nothing
2:17:44 entertaining about this
2:17:47 he says the tick tock
2:17:51 who says that
2:17:53 he does
2:17:55 okay
2:18:01 I'll flip the screen over in a second
2:18:02 let me
2:18:04 let me find my peace Parts first
2:18:07 that was the challenge engine
2:18:11 wait where is it
2:18:13 challenge engine no chat gbt
2:18:16 lesson plan maker
2:18:20 where's the where's my example
2:18:23 I lost my example
2:18:33 that's it I think this is it
2:18:41 yes okay
2:18:44 now visualize your zaps
2:18:47 zoom out and use that okay whatever stop
2:18:50 giving me new features
2:18:52 I know how to use it now don't give me
2:18:54 new [ __ ]
2:18:56 oh no what have they done oh visual
2:18:58 editor
2:19:00 classic editor okay we're gonna go to
2:19:01 the classic editor just because I don't
2:19:03 know how to use a visualizer
2:19:05 okay
2:19:10 okay
2:19:19 so when you come to zapier
2:19:22 okay so wait let me let me let me start
2:19:25 at the beginning
2:19:27 let me let me go uh
2:19:31 I'm just trying to figure out where to
2:19:32 start where where it's gonna make sense
2:19:35 and it's gonna be somewhat remotely
2:19:37 interesting so so bear with me as I try
2:19:40 to get my head around what I want to do
2:19:43 okay uh here
2:19:45 so the way you can think about
2:19:46 automation tools like zapier and make
2:19:49 make is another one that's like this
2:19:51 that's got a slightly different
2:19:52 interface but it's the same basic idea
2:19:54 is all of them function effectively the
2:19:58 same you've got what they call a trigger
2:20:00 so that's something that starts the
2:20:02 automation
2:20:03 and then you've got some series of
2:20:05 actions
2:20:08 um
2:20:09 and what I have on screen right here is
2:20:12 is kind of the most simple automation
2:20:14 you can do where you've got an input
2:20:16 some sort of processing and then an
2:20:18 output
2:20:19 so the input I have is a Google form
2:20:23 the processing I have is send a prompt
2:20:26 with
2:20:28 the data from that Google form to to
2:20:33 um chat gbt to gpt3 in this case
2:20:36 and then take the result of that and pop
2:20:39 it in an email and send that email so
2:20:43 the inputs here could be you add a new
2:20:47 row to a Google form
2:20:49 you upload a file into a directory on
2:20:53 Google Drive you create a new Google doc
2:20:56 there's all sorts of things that can
2:20:58 trigger this it can be someone posts
2:21:00 something
2:21:02 in a slack channel right so the trigger
2:21:06 is something happens somewhere that's
2:21:08 connected into your thing
2:21:11 and then that that starts this Cascade
2:21:14 and so how how you add another step is
2:21:16 there's this little plus button here and
2:21:18 then you you add it and and so so let me
2:21:21 go through this so so what I did was I
2:21:23 created
2:21:24 a really simple Google form
2:21:27 the most simple Google form
2:21:30 that basically just said what's your
2:21:32 favorite color and then your answer and
2:21:34 submit so so that was the form that I
2:21:38 created
2:21:39 then over in zapier when you create one
2:21:43 of these
2:21:44 um
2:21:45 one of these zaps you could you kind of
2:21:46 build them think think of these as Lego
2:21:48 blocks you build it a Lego block at a
2:21:50 time and you kind of start from the
2:21:52 input and work your way down so
2:21:55 I said I want to create a Google form so
2:21:58 if I change this so so you come into
2:22:00 this little thing which says you know
2:22:03 now it says change the trigger but if I
2:22:05 was creating this from scratch it would
2:22:06 say create a new trigger and then I
2:22:08 would say Google form
2:22:11 right there's Google forms right there
2:22:15 um and then I select well what do you
2:22:18 want to do on this Google form well I
2:22:20 want a new form response so so
2:22:23 depending on the module you choose
2:22:25 you've got different options that that
2:22:27 API allows right so I can either say
2:22:30 anytime someone put something new in the
2:22:32 form or if they put something new or
2:22:34 update their form I'm going to take this
2:22:36 action and then I hook my account up and
2:22:39 then I say what what form do I want So
2:22:42 based on my account I got lots of forms
2:22:44 I can choose from so I choose the I
2:22:47 chose the uh
2:22:49 the one that said favorite color so
2:22:51 that's the one with our single color
2:22:53 there and then I test it
2:22:55 and
2:22:57 um I have a form response from when I
2:23:00 created the form I did a little test and
2:23:03 I put in the color green and you'll
2:23:05 notice over here when I when I tested it
2:23:07 it pulled in the color green so pull in
2:23:09 a bunch of other [ __ ] about the form but
2:23:11 I'm just interested in can it see the
2:23:14 answer that I typed into that form it
2:23:16 can
2:23:17 so now what happens is now that I've got
2:23:20 a trigger and I've got some data input
2:23:22 from that trigger now I can use that in
2:23:25 anything that follows it and that's one
2:23:27 of the cool things about the automations
2:23:29 is the more modules you add you can use
2:23:32 elements from anything above it okay all
2:23:35 right
2:23:37 so now we're going to say
2:23:40 continue with that that record so now
2:23:42 it's moved us down to our open AI prompt
2:23:45 our openai module right and so
2:23:49 an important thing to know is there's an
2:23:51 open AI module and there's a chat GPT
2:23:53 module
2:23:55 um
2:23:57 I'm a little cranky about the chat GPT
2:23:58 one because the the the fields are not
2:24:01 the same as they are over in in opening
2:24:04 eyes playground but that's neither here
2:24:06 nor there
2:24:07 okay so so what's happening here is it's
2:24:10 it's a very similar thing
2:24:12 um to what we just did so we're doing an
2:24:15 open AI module we've got a list of
2:24:18 things we can do here so we can check
2:24:20 moderation to see if someone cusses at
2:24:23 us we can create a transcription we can
2:24:25 create a translation we can generate an
2:24:27 image with Dali we can send a prompt so
2:24:30 that's the one I selected so we're going
2:24:32 to send a prompt
2:24:33 then I hook up my account
2:24:35 then I
2:24:38 um now this is where I do my prompt so
2:24:40 here's the model that I'm using I'm
2:24:43 using text DaVinci 3 and here's my
2:24:46 prompt I want you to use the following
2:24:49 variables
2:24:50 color and so let me get rid of this so
2:24:54 so over in in the
2:24:56 um
2:24:57 GPT playground I sort of figured out the
2:25:00 prompt I wanted to do and this one's
2:25:02 just a stupid stupid example prompt but
2:25:05 I want you to use the following
2:25:06 variables color so
2:25:09 because
2:25:11 I've got that form above that I've I've
2:25:14 now got access to I can now take any of
2:25:18 the input fields from that form and put
2:25:21 them into my prompt so I'm going to find
2:25:22 the one that says green
2:25:25 which is the bottom one here says what's
2:25:26 your favorite color I click green and so
2:25:29 that's now in my prompt so so this is a
2:25:31 variable within my prompt so what that
2:25:33 means is anytime someone fills out a
2:25:36 form it's going to take that color and
2:25:38 it's going to stick it in my prompt and
2:25:39 it's going to send that to open AI okay
2:25:42 and then like here's the temperature
2:25:44 here's the maximum length so this is all
2:25:46 stuff that you play with at playground
2:25:48 the temperature if I move it to zero
2:25:50 it'll be more dry if I move it to two
2:25:53 it'll be
2:25:55 um more creative one is somewhere in the
2:25:58 middle
2:25:59 um
2:26:00 I ignore a lot of the other [ __ ] just
2:26:03 because I'm I'm generally just doing
2:26:05 fairly straightforward things and then I
2:26:07 test that and so so basically what the
2:26:09 prompt was write a write a poem about a
2:26:12 blank colored animal right and so it
2:26:16 gave me back this is the test response a
2:26:18 silly frog of green croaks and hops and
2:26:21 you know it so it's a green frog poem so
2:26:24 that worked and now I say continue and
2:26:27 my next module was Gmail so
2:26:30 I pick Gmail I hook my account up
2:26:33 I say we're gonna we're gonna send an
2:26:36 email
2:26:36 so you fill in all the fields of send
2:26:39 the email and then I'm going to say
2:26:41 here's your poem about a and so I pulled
2:26:45 the the green variable from the form
2:26:48 right so here's your poem about a green
2:26:51 animal that's the subject and then the
2:26:54 body of the email is hello there you
2:26:58 asked for a poem about a green animal a
2:27:01 blank animal here it is and then this
2:27:04 other variable is the response from open
2:27:06 AI So within this email it's it's using
2:27:10 both inputs from the form and from open
2:27:12 AI
2:27:14 and then when I go test that
2:27:18 retest action
2:27:22 did it do it it should send me an email
2:27:26 so I'll go over to Gmail
2:27:35 [Laughter]
2:27:39 and so here we go
2:27:41 me from me here's your poem
2:27:46 here's your poem about a green animal
2:27:49 hello there you asked for a poem about a
2:27:51 green animal and I don't have a space
2:27:52 there so I see there's a typo in my form
2:27:54 so I could go fix that and then there's
2:27:57 my poem right and so now anytime anyone
2:28:01 goes to that form and fills in a color
2:28:03 within 20 seconds they'll get an email
2:28:06 with a poem about something colored
2:28:08 animal
2:28:10 so that's automation
2:28:16 that help was that boring
2:28:18 we lost 20 people
2:28:21 we lost 20 people doing the automation
2:28:24 we lost them they're like I can't take
2:28:26 it he's just talking about forms and
2:28:29 variables
2:28:30 this is like program I didn't want
2:28:32 homework
2:28:34 I came here for entertainment
2:28:37 now he's talking about zapier
2:28:41 zaps and variables
2:28:46 I can't take it Sally
2:28:49 again
2:28:51 I'm done
2:28:53 I'm out AI learning lab my ass never
2:28:57 again oh I'll be back tomorrow or I'll
2:29:00 see you tomorrow all right bye
2:29:04 all right
2:29:07 I stay up to see you heart heart but
2:29:09 gotta go to bed good night Dawn thank
2:29:12 you
2:29:13 um hey guys by the way my name is
2:29:15 normally Stevo yep we saw that one
2:29:16 earlier
2:29:17 you must read some manuals that's how
2:29:20 the Challenger exploded
2:29:21 no the Challenger exploded well they
2:29:24 didn't read the manual but didn't the
2:29:25 dude come in and bang the thing on the
2:29:28 table and said this is a problem and
2:29:30 they're like sure
2:29:31 get out pee on with your engineering
2:29:35 skills
2:29:38 yeah I know I I'm not I'm just not gonna
2:29:41 do it I ain't never read a manual ain't
2:29:43 never gonna read a manual I ain't gonna
2:29:45 do it and now haha
2:29:49 wait who said I needed to do that
2:29:51 Lynette Lynette ha
2:29:54 these things will read the manual for me
2:29:56 do you know you can take full-on
2:29:59 documentation and pop them in these
2:30:01 things and just have it give you
2:30:03 instructions on how to do [ __ ]
2:30:06 you don't ever have to read a manual
2:30:08 again we got machines for that now I got
2:30:10 people for that
2:30:13 I don't really knows thinking manual
2:30:15 never I will never do it never
2:30:19 all right best hour of my day just spent
2:30:22 here on this chat thank you TM thank you
2:30:25 peace love it oh my God the turtle guy
2:30:28 made a new video I know I know
2:30:31 I got messaged by someone
2:30:34 impersonating you be careful there are
2:30:36 fraudsters lurking
2:30:38 I got really
2:30:40 that's terrifying
2:30:43 um Lomas camper if if you would DM me
2:30:46 and tell me specifically what happened
2:30:47 there
2:30:48 I'm I'm curious well first of all yeah
2:30:54 all right
2:30:56 I got I got nothing going on here
2:30:59 Commerce wise but I suppose someone
2:31:01 could say bye my class
2:31:03 if I buy my class on you can make money
2:31:05 watch out
2:31:08 oh my God
2:31:10 I didn't think oh
2:31:12 that's annoying all right well we'll
2:31:15 have to deal with that if I find them
2:31:18 I'll tell him to knock it the [ __ ] off
2:31:19 sorry about that that sucks
2:31:21 Kyle can you do automation so I hope
2:31:23 that helped Dan I hope that was uh that
2:31:26 gave you what you needed to learn there
2:31:28 um it's it's kind of a it's kind of a
2:31:31 Deep River to run in but once you get
2:31:34 your header like the basic idea with
2:31:36 automations like the biggest thing to
2:31:38 get is you've got inputs you process the
2:31:40 inputs and then you output them
2:31:41 somewhere like once you get your head
2:31:42 around that then it's all just
2:31:44 variations on the theme but some of
2:31:46 those modules are really janky and
2:31:49 complicated and just overly whatever
2:31:52 this is crazy I'm adding I I am now
2:31:55 adding monetization to my platform using
2:31:57 stripe oh that's great that's great with
2:32:00 chat GPT chat GPT is writing it all yeah
2:32:02 it's isn't it nuts
2:32:04 it's nuts so so this is it so what tm's
2:32:07 doing is really good
2:32:10 one of the toughest things to sort of
2:32:12 flip your head around that the thing
2:32:13 that took me a while is I would just
2:32:15 kind of sit and stare at these things
2:32:17 like they were gonna do something I'm
2:32:19 like all right what do I do with it I
2:32:22 should I I guess I could make another
2:32:24 poem uh
2:32:27 um
2:32:28 what do I is there a class
2:32:32 because they can do anything they can
2:32:35 get very overwhelming and and the best
2:32:37 way to get your head around what they
2:32:39 can do is is don't think about what can
2:32:42 I do with AI flip it around and say
2:32:45 what's something in my life that I'm
2:32:47 trying to do
2:32:49 that it would be interesting to see if I
2:32:51 could do with these things
2:32:53 and then you've got a then you have an
2:32:55 imperative you're like okay all right so
2:32:57 I wanna I wanna write a Mad Lib style
2:33:00 thing for first-time Founders okay that
2:33:04 that I can let me go here and see if I
2:33:05 can do that and then you'll start doing
2:33:07 it and then at some point it will give
2:33:09 you some sort of output where you'll go
2:33:11 oh God
2:33:13 I didn't know I could do that
2:33:20 when you have your Kevin McAllister
2:33:22 moment and then your life will never be
2:33:23 the same and then you just show up on
2:33:25 these [ __ ] tick tocks night after
2:33:26 night going why am I here what have I
2:33:28 done with my life
2:33:30 oh my God more chat ubt oh it's so good
2:33:34 it's so good
2:33:36 I used to like pudding but now it's Chad
2:33:39 gbt I could have pudding and chat EBT oh
2:33:43 no this is so good and then three hours
2:33:45 later you're like where's that pudding
2:33:48 oh
2:33:53 do you own a cat I am not a cat person
2:33:56 I'm a dog person I don't like the kitty
2:33:58 cats sorry sorry
2:34:02 kitty cats are fickle
2:34:04 I like I like unbridled love doggies are
2:34:08 just doggies are like you walk in and
2:34:10 like oh my God you're here this is the
2:34:12 best thing ever
2:34:17 I don't like to like please pet me
2:34:19 please pet me [ __ ] you
2:34:22 that doesn't do it for me
2:34:24 cats make me crazy like you want the
2:34:27 love you get the love
2:34:28 you want the love and then you scratch
2:34:30 the [ __ ] out of me um
2:34:32 Homie don't play that
2:34:36 all right I know coder's using chat gbt
2:34:39 to help them with bug fixes and I use it
2:34:41 for quick templates yeah that's a good
2:34:43 idea that's cool
2:34:44 yeah you can oh one of the things these
2:34:46 things are really good at if you haven't
2:34:48 done it
2:34:50 it's really good at mimicking stuff so
2:34:52 you can give it like for Story vine we
2:34:55 write these things called video guides
2:34:57 which are a series of prompts that show
2:34:59 up in our app that walks someone step by
2:35:01 step through telling their story they're
2:35:03 written in a very particular way that
2:35:04 you know because of the nature of how
2:35:06 the app works and and we've sort of
2:35:08 figured out over the years what works
2:35:09 and what doesn't and you start with an
2:35:12 introduction and then it's got a story
2:35:13 arc and just all sorts of [ __ ] in it
2:35:16 and so
2:35:18 I was trying for a long time to get it
2:35:20 to do our prompts correctly and I was
2:35:22 just you know writing these increasingly
2:35:24 complicated prompts and then it hit me
2:35:26 oh [ __ ] why don't I just give it an
2:35:27 example and then I said I I explained
2:35:31 you know I have a video company and we
2:35:33 write these story Frameworks called
2:35:35 video guides and here's what they look
2:35:37 like and so now write me a story that's
2:35:40 got five questions in it you know that
2:35:42 follows that format that tells this
2:35:45 story and then out it comes and it was
2:35:48 [ __ ] perfect
2:35:51 like Bang from day one so if you've got
2:35:54 something you have a particular kind of
2:35:56 writing style or you've got something
2:35:58 that's formatted in a particular way or
2:36:00 you've got in your case a code template
2:36:03 pop that template in there and tell it
2:36:05 to solve a problem within that template
2:36:07 or follow the tone of the email it's
2:36:09 [ __ ] brilliant for that
2:36:11 do you believe humans will use AI also
2:36:13 in weapons in the future someone was
2:36:15 asking about palantir earlier so yes we
2:36:18 already are one of the dangers in my
2:36:20 opinion totally agree
2:36:22 um Boston Dynamics has come out publicly
2:36:25 and said you know we probably shouldn't
2:36:28 put weapons on our robo dogs
2:36:33 thanks Boston Dynamics I'm sure that's
2:36:35 what they tell the public and then with
2:36:38 their defense contractors
2:36:39 they're like wait can is there any way
2:36:42 we can stick on six more guns paint
2:36:44 pointed in all directions
2:36:47 so you know they're doing that [ __ ] but
2:36:50 you know long as it's with the military
2:36:53 um yes I do think that uh I do think
2:36:56 that AI will be used in weapons and you
2:36:58 know bad guys will get their hands on AI
2:37:00 and good guys will get their hands on AI
2:37:02 and
2:37:03 well well yeah yeah and will there be
2:37:06 big news stories where something went
2:37:07 awry absolutely
2:37:10 we are we are predisposed right now
2:37:14 to be afraid of these things for a
2:37:17 couple of reasons one is any new
2:37:19 technology especially ones that are
2:37:21 going to be disruptive
2:37:23 scare the [ __ ] out of people and it's
2:37:26 happened throughout history
2:37:28 happened with the tractor it happened
2:37:30 with the camera
2:37:32 with the daguerreotype
2:37:35 it happened with the printing press
2:37:38 it happened with the World Wide Web
2:37:42 and
2:37:44 so so humans are you know resistant to
2:37:48 change they don't like change they're
2:37:49 afraid of new technology we've also got
2:37:52 50 years of history of Hollywood telling
2:37:54 us the robots are going to kill us
2:37:57 so we've seen all the movies
2:37:59 so the minute that we have a story
2:38:02 that's like oh that's like that scene in
2:38:04 Terminator
2:38:06 here we are you know it'll make the
2:38:08 cover of the New York Times and it'll be
2:38:10 on all the news channels for three days
2:38:13 and it'll probably knock this [ __ ] back
2:38:15 six months
2:38:17 so yeah that's gonna happen and is it
2:38:20 gonna suck yeah it's gonna suck but you
2:38:22 know
2:38:26 they ain't gonna be uninvented they
2:38:28 ain't going away I'm sorry about the
2:38:29 hair I don't know what just happened
2:38:30 there
2:38:32 it was normal and now it's not that's
2:38:34 the way she rolls it's just my lot in
2:38:38 life
2:38:39 all right
2:38:42 um it will be used a lot by experienced
2:38:44 coders it will be used a lot what I've
2:38:46 heard Winston from uh
2:38:49 the the all-in podcast Jason kalkanis
2:38:52 and those guys is that what what what
2:38:54 they were saying is that
2:38:56 um Tech shops in San Francisco that are
2:38:58 using
2:38:59 um GitHub co-pilot as a as a normal
2:39:03 course of business they're seeing 30 to
2:39:05 40 percent increases across the board in
2:39:08 those companies that's that's [ __ ]
2:39:10 significant and this is and this is like
2:39:13 the new raw not really working version
2:39:16 of this technology like imagine two
2:39:18 years out when you can you can just kind
2:39:20 of Imagine a sophisticated you know
2:39:24 subroutine and it just writes it for you
2:39:27 and integrates it and updates all your
2:39:29 API calls
2:39:30 imagine never having to update a library
2:39:33 ever again because the AI is just
2:39:35 sitting there waiting for Library
2:39:37 changes it just scans your codes looks
2:39:40 at all your libraries and dependencies
2:39:41 goes out
2:39:43 finds any updates
2:39:45 finds replacements for things that when
2:39:48 things have been deprecated
2:39:49 automatically update your code
2:39:53 what's gonna happen to all of the Ops
2:39:56 people that that's all they do with
2:39:57 their time I don't know
2:40:00 they can go right the Bots that go do
2:40:02 that [ __ ]
2:40:03 and then they can sit back and smoke a
2:40:05 big fat cigar going yeah
2:40:08 my machines are coding themselves
2:40:10 [Laughter]
2:40:13 bring me another cocoa
2:40:20 hey I will take software engineering
2:40:22 jobs accountants lawyers too so Jorge
2:40:25 Ramirez yes AI is gonna so any any all
2:40:32 all jobs that are intermediated with
2:40:35 computers
2:40:37 so
2:40:39 every knowledge work
2:40:41 job every creative job every
2:40:47 anything
2:40:48 Logistics
2:40:50 manufacturing anything with computer
2:40:52 mediation
2:40:55 the jobs
2:40:58 every single job is going to be
2:41:00 redefined by these tools every single
2:41:02 job
2:41:03 now
2:41:05 does that mean every single person gets
2:41:07 replaced no
2:41:09 what it means so so there's a there's a
2:41:12 guy I talk about a lot on here named
2:41:13 Paul rates or he's got the marketing AI
2:41:15 Institute and he he said
2:41:17 AI will not replace your job but someone
2:41:19 using AI will
2:41:21 and so one of the things that this
2:41:23 channel is about is just encouraging
2:41:25 people like why I have these things up
2:41:27 on screen for the entire [ __ ] thing
2:41:28 is if you're not playing with these
2:41:31 things go play with them go right now to
2:41:33 chat GPT if you haven't done it yet
2:41:36 and start playing start getting AI
2:41:38 literate because
2:41:41 when those disruptions start to happen
2:41:44 in every business
2:41:46 the powers that be at some point are
2:41:48 going to go oh [ __ ] we got to get our sh
2:41:51 our [ __ ] together with AI
2:41:53 who knows anything about AI uh Jorge was
2:41:57 talking about Chachi PT in the lunchroom
2:41:59 the other day uh bring Jorge to my
2:42:01 office
2:42:02 Jorge the [ __ ] going on with this AI
2:42:05 stuff I don't know it's you know
2:42:08 would you like me to write you a poem
2:42:10 about quantum mechanics here you go
2:42:14 um
2:42:16 so so get literate get literate start
2:42:20 figuring out what these things do what
2:42:21 they can't do what they're good at what
2:42:23 they're bad at
2:42:24 um even if you don't have a goal in mind
2:42:26 it's gonna be it's it's gonna serve you
2:42:28 well
2:42:29 um
2:42:30 so yeah now here's the good news so
2:42:33 here's what this is just my gut right
2:42:36 this is all this is all speculation
2:42:39 right now we're just starting to see
2:42:41 impact on the workforce you know we're
2:42:44 starting to hear oh so and so laid off 4
2:42:47 000 people and they're replacing them
2:42:48 with AI and this company is not going to
2:42:51 hire any more people because of AI and
2:42:53 we got the SAG strike and the writer
2:42:54 strike right now
2:42:57 this stuff is kind of at the periphery
2:42:59 of the conversation it's sneaking in a
2:43:01 little bit but it's at the periphery of
2:43:02 the conversation
2:43:03 over the next two to three years
2:43:08 big swaths of jobs are going to be
2:43:10 impacted and it's they're just going to
2:43:12 get bowled over and I don't think we're
2:43:14 frankly prepared for it so I think
2:43:15 there's going to be a lot of pain and a
2:43:18 lot of disruption
2:43:19 what we don't know is
2:43:21 on the other side of that pain and
2:43:23 disruption all of these Technologies as
2:43:26 they get more sophisticated are going to
2:43:28 generate lots of prosperity
2:43:30 and they're going to generate lots of
2:43:32 new jobs and lots of new job demand
2:43:35 that
2:43:36 the workforce won't be trained for
2:43:39 because it's going to be entirely new
2:43:40 things
2:43:41 so who's going to be first to fill those
2:43:43 jobs people that at least understand
2:43:46 something about AI right and so as it
2:43:49 does this and the workforce kind of
2:43:51 diminishes and does this it's the people
2:43:54 that sort of you know hop out of
2:43:56 whatever they were doing before start
2:43:59 learning this AI stuff and they're going
2:44:00 to be able to ride that that wave up
2:44:04 um so I think five years out we start to
2:44:07 see all sorts of New Opportunities and
2:44:09 jobs
2:44:10 but the two to three year period you
2:44:14 know sort of probably starting 2024
2:44:18 I think it's going to be painful I've
2:44:21 used AI to help program an idea I have
2:44:24 but it doesn't do certain languages that
2:44:26 I guess I need
2:44:29 um
2:44:30 it's possible
2:44:33 um
2:44:34 I would just keep asking it questions
2:44:37 because there's with programming there's
2:44:39 always many ways to to solve the problem
2:44:44 um so I would keep uh keep uh
2:44:48 I would keep on trucking I would I would
2:44:51 tell it that you're frustrated and you
2:44:53 know if it says it doesn't do certain
2:44:56 languages say well what languages do you
2:44:58 do like I would just keep pushing on the
2:45:00 edges of it just keep having it educate
2:45:02 you about what you need so here's here's
2:45:05 an example when I first started using
2:45:08 chat EBT I thought oh I'm gonna have it
2:45:10 write me some python code I don't know
2:45:12 python
2:45:14 and so it wrote python code
2:45:16 I was like cool
2:45:18 and then I realized I don't know how to
2:45:20 run python code
2:45:22 so I told it I don't know how to run
2:45:23 python code and it said oh sure you just
2:45:25 go to your terminal and you type in this
2:45:27 command line and and I was like oh no I
2:45:29 don't want to type in command lines I
2:45:30 don't know I'm not yeah I don't know
2:45:32 it's that feels like work to me I said
2:45:34 isn't there a website that I can use
2:45:36 that'll just do this for me and it gave
2:45:37 me five websites and I picked one
2:45:40 and then it taught me how to use the
2:45:41 website to run python code so there
2:45:44 there there may be ways for you to to
2:45:48 get around that the other thing to do is
2:45:51 um
2:45:52 talk to people in your life that you're
2:45:54 doing this that you're playing with AI
2:45:56 even if they think you're weirdos
2:45:58 someone in there might be playing with
2:46:01 this stuff also and maybe they've got
2:46:02 coding skills and could support you on
2:46:05 that so I would sort of get generous
2:46:08 about talking about
2:46:10 how you're exploring it and by the way
2:46:12 9.5 out of 10 on the name verified mom
2:46:16 is a solid name very solid like it my
2:46:20 internet is bad today [ __ ] Xfinity
2:46:22 Chris you too
2:46:24 Xfinity it's so bad
2:46:27 you got another account I did not I just
2:46:30 came back they only booted me for seven
2:46:32 days can we talk automations we just did
2:46:36 um
2:46:37 prompt engineering
2:46:40 oh by the way there's 90 folks here I am
2:46:44 more than 99 messages behind I
2:46:47 because I'm an idiot I like to read
2:46:50 every comment from every person but it
2:46:53 means
2:46:54 as the I've been doing this now for like
2:46:57 two hours two and a half something like
2:46:59 that I don't know way too long what time
2:47:01 is it oh I gotta go to bed it's [ __ ]
2:47:03 11.
2:47:05 um
2:47:06 I'm just I'm behind oh I really I gotta
2:47:08 go I've got a workshop I Gotta Give
2:47:10 tomorrow and I I am behind I haven't
2:47:12 done my homework for my little workshop
2:47:14 yeah
2:47:17 it's an add thing
2:47:20 as long as it's an emergency I can get
2:47:22 it done
2:47:26 um anyway I'm really behind on the on
2:47:29 the comments so I'm gonna you know what
2:47:31 I'm gonna do I'm gonna jump down to the
2:47:33 bottom
2:47:34 because I need to go soon so I'm just
2:47:36 gonna skip a bunch of questions so if I
2:47:38 missed your question sorry about that so
2:47:39 I've now jumped down to the bottom and
2:47:41 I'm gonna just go back a few and just
2:47:44 grab a couple more questions Kyle's hair
2:47:46 is tired now says Kyle
2:47:49 um I'm a software engineer use it every
2:47:51 day to help write code that's cool Sam
2:47:54 the name wait and the name is a letter
2:47:57 or so off that's what you were saying AI
2:48:00 learning lab 2ls
2:48:02 huh
2:48:06 did I misspell something somewhere
2:48:11 Maybe
2:48:12 gotta pivot to architecture or cyber
2:48:15 security architecture is going to get
2:48:16 really cool in the next couple of years
2:48:18 what's gonna happen when planned
2:48:21 obsolescence is thwarted by
2:48:23 self-updating programs
2:48:25 I think everything's gonna get [ __ ] up
2:48:28 I you know
2:48:35 I think it I think it's going to be
2:48:37 really fascinating to watch
2:48:40 the power infrastructure not not
2:48:43 electric power but the powerful
2:48:45 corporations
2:48:49 as these tools start [ __ ] with the
2:48:53 power centers
2:48:54 the death throws of those power centers
2:48:57 are going to be fascinating they're
2:48:59 going to try to shut this [ __ ] down and
2:49:01 they're going to try to they're going to
2:49:02 try to box it in and I think Meta Meta
2:49:06 releasing the Llama 2 models
2:49:07 commercially is a really interesting
2:49:10 hedge against that kind of control
2:49:14 um
2:49:16 it's going to be fascinating like I have
2:49:18 no idea what it's going to look like but
2:49:19 but
2:49:21 chaos Generation by the way 10 out of 10
2:49:24 on the name I like chaos generation is a
2:49:27 really good name
2:49:28 um
2:49:31 yeah I
2:49:33 just just keep paying attention just
2:49:36 like like here's the deal if if you look
2:49:38 at this [ __ ] from the outside it it
2:49:41 looks a certain way and it's like oh
2:49:42 it's over hyped and it's Geeks and it's
2:49:44 this and that
2:49:45 when you get on the inside of this stuff
2:49:47 and actually start playing with it and
2:49:49 see what it makes possible and start to
2:49:51 just imagine even two years in the
2:49:53 future if this if this thing I'm doing
2:49:55 right now got just a little bit better
2:49:58 who would that threaten
2:50:00 start paying attention to that [ __ ]
2:50:03 and if you start seeing the who would
2:50:05 that threaten people starting to make
2:50:07 noise
2:50:08 you're starting to see the death throws
2:50:10 of a of a change in power air traffic
2:50:13 controllers let's replace them with AI
2:50:15 my big my biggie right now is uh
2:50:18 customer support call centers they need
2:50:20 to go
2:50:22 they need to go and be replaced with
2:50:24 competent AI
2:50:26 and phone answerers that talk to me and
2:50:29 can actually solve my [ __ ] problem
2:50:31 without putting me on hold for 45
2:50:33 minutes and then hanging up with me or
2:50:35 hanging up on me
2:50:37 after my fourth person I'm talking to
2:50:40 where I've had to re-explain my problem
2:50:42 to each person every single time that
2:50:46 [ __ ] needs to go so you take your air
2:50:48 traffic controllers I'll take the
2:50:50 customer support people let's get these
2:50:52 things moving on
2:50:54 what's it going to do to digital
2:50:56 marketing completely totally [ __ ]
2:50:59 obliterated and upend the entire agency
2:51:03 structure
2:51:05 time and materials is a relic of the
2:51:08 past but no one knows it yet
2:51:14 the whole ad industry is but what if I
2:51:17 like turtles whole ad Industries
2:51:20 what I mean what's cool is as as the ad
2:51:24 industry starts to make things more
2:51:26 efficient and let go of people as they
2:51:29 find efficiencies they're still going to
2:51:31 struggle to hold on to their time and
2:51:32 materials model and what's going to
2:51:34 happen is the 10 of the people that they
2:51:37 cut due to efficiencies or twenty
2:51:39 percent the Smart Ones of those are
2:51:42 already learning this [ __ ] if they
2:51:44 haven't already learned it and they're
2:51:45 going to go start 10 person agencies
2:51:47 that can come kick the [ __ ] out of 200
2:51:50 person agencies
2:51:52 the holding companies are in trouble and
2:51:54 they're being [ __ ] arrogant about it
2:51:56 right now
2:51:58 I'm talking to the clients like we've
2:52:01 got big ass clients storyline's a small
2:52:03 company but we have big ass clients and
2:52:05 we have some pretty trusted
2:52:07 relationships right now
2:52:09 and they're telling us what they think
2:52:11 of their holding companies and it ain't
2:52:12 pretty
2:52:15 they're in trouble
2:52:17 they don't know it yet
2:52:19 because they're you know
2:52:21 full of themselves because they're big
2:52:23 powerful people that have been big and
2:52:24 powerful for a lot of decades
2:52:27 but they got some [ __ ] coming
2:52:31 we're incorporating touch EBT into our
2:52:33 six-figure blueprint digital business
2:52:35 that's super cool
2:52:37 I like that
2:52:39 I guess I'll be staying oh styling hair
2:52:42 forever if you're dealing with me we got
2:52:45 we got some issues to deal with
2:52:47 I hope you go work for Trump
2:52:50 huh
2:52:53 okay
2:52:55 I don't even know what that means yep
2:52:57 I'm the weirdo in my circles
2:53:00 that's weirdos gotta stick together
2:53:01 that's what that's what this channel is
2:53:03 all about
2:53:04 this is this is a weirdo collection Bowl
2:53:07 [Laughter]
2:53:10 oh my God somebody is impersonator You
2:53:14 by a similarly named account Kyle I'll
2:53:17 message you oh okay got it all right
2:53:19 we'll get them whoa
2:53:21 get them uh whatever you call it
2:53:24 we'll get them we'll get them
2:53:26 tick-tocked out
2:53:28 like what
2:53:31 why why oh are they like stealing my
2:53:35 content and doing like I don't what
2:53:38 all right I'll have to go check it out
2:53:41 there's no there's somebody
2:53:44 impersonating you and scamming people
2:53:46 it's AI learning lab with two L's
2:53:51 oh ail learning lab oh I see okay
2:53:55 all right I'll go check it out
2:53:58 and how are they scamming people
2:54:01 oh that's exhausting okay can we stop it
2:54:05 with the [ __ ] scamming like really
2:54:08 isn't it just as easy to start a [ __ ]
2:54:11 business as it is to like scam people
2:54:13 and just try to [ __ ] sleep at night
2:54:15 never understood that
2:54:19 can you try the Vizsla plug-in the link
2:54:21 is in response where the video outcome
2:54:23 is I tried it the other night deadly and
2:54:25 it didn't work it was broken so I'll try
2:54:28 it at some point I got to go to bed now
2:54:29 but I actually got to go do some work
2:54:31 now
2:54:32 um I think they were telling you about
2:54:34 the impersonator with a misspelled
2:54:35 screen name I will find them
2:54:37 uh Richard feyman came in and dropped an
2:54:40 O-ring into ice water and showed it
2:54:43 became brittle
2:54:45 oh that yeah that's what it was yeah
2:54:46 exactly
2:54:48 exactly that was the The Challenger
2:54:50 thing yeah they said this is bad
2:54:53 all right this this stream makes me
2:54:55 anxious why
2:54:57 all right oh my God I'm so ready for AI
2:54:59 to replace customer call centers I know
2:55:01 right exactly
2:55:04 all right people I gotta go
2:55:08 it's time good night Kyle do you have
2:55:11 homework to do I do I have homework to
2:55:12 do all right peace everybody like
2:55:15 honestly AI makes it easier to scam than
2:55:18 start a business that's a big fear yeah
2:55:20 I know we'll get after him thanks Eric
2:55:22 growth geek appreciate it peace out
2:55:23 everybody peace out the bear Dr J chaos
2:55:29 generation
2:55:30 Rebel Alex peace out everyone thank you
2:55:33 talk to you tomorrow
2:55:36 peace