aiLL 8/28/23 - The Great Renaissance: How AI Will Unlock Human Potential

Description

Chapters

Transcript

0:05 foreign
0:29 welcome to the AI learning lab Tobias
0:31 right in there I feel like number two
0:34 what's happening
0:38 good to see you good to see you
0:42 all right
0:44 got the Hamilton shirt on tonight
0:51 um happy Monday happy Monday hey Maria
0:53 do you it's shaking
0:54 welcome welcome welcome welcome to the
0:57 EI learning lab
0:58 got a lovely little graphic here we got
1:01 more Graphics we got look
1:03 I should go make more I only made four
1:06 but these are with that new tool
1:08 ideogram ideogram
1:12 pretty slick pretty slick
1:14 you can make letters and pictures now
1:17 didn't be you could do that before but
1:19 now you can I got a little pimple right
1:21 there it's just it's not pleasant love
1:25 this shirt
1:28 Hamilton
1:33 that was a fine musical although I saw a
1:37 road and truck show of it here in Denver
1:39 um
1:40 bad casting not so good
1:45 but what are you gonna do
1:47 welcome everyone my name is Kyle Shannon
1:49 this is the AI learning lab we have this
1:51 lovely graphic here
1:53 um also if you are new to the world of
1:56 generative AI what we talk about on the
1:57 channel is generative AI you ask
1:59 questions I I answer them I wax
2:02 philosophic I make funny voices I cuss
2:05 like a sailor sometimes then other times
2:07 I don't uh but but I was called
2:09 classless the other night or something
2:11 like that because of my uh tacky I think
2:14 because of my cousin but it happens so
2:16 if you don't like cussing I wouldn't
2:18 stay uh the Channel's popular enough we
2:21 have a scammer that's cloned our Channel
2:23 and stolen all our videos so that's fun
2:25 uh if you haven't played with chat gbt
2:28 go here right now like while I get this
2:31 up and running we're waiting for people
2:32 to come in
2:33 chat.openingi.com that's the official
2:35 chat EBT website I don't work for them I
2:38 this is more here these are things that
2:40 I think are not a waste of your time
2:41 that's why these are here
2:43 um chat gbt is really the best of these
2:45 things right now um there's there's some
2:48 some getting close some getting close
2:51 Claude is getting close claude's got
2:53 some advantages that chat CPT doesn't
2:55 but just go there
2:57 start talking to it if you're like but
2:59 Kyle I don't know what to say what would
3:01 I say it's just like a Google box it's
3:03 like the Google box you just do a Google
3:04 search is it like the Google boxes not
3:06 like the Google box well it is like the
3:08 Google box but it's not like the Google
3:09 box
3:10 much more different than the Google box
3:11 deeper way deeper
3:14 uh if you go to this thing prompts.chat
3:17 that will teach you how to talk to that
3:19 thing if there's a primer in there and
3:21 there's sample prompts of like you can
3:24 tell it to act like certain things
3:26 a lot of people also come in here and
3:28 they're like but Kyle how can I make
3:30 money with uh with AI and and what I
3:33 always say is
3:37 no I don't say that
3:40 avoid those people
3:41 anyone is like give me money won't judge
3:44 every day
3:45 no it's I mean you can you can
3:48 but it's not about the tool it's about
3:50 what you do with the tool like any tool
3:52 so so we can talk about that uh Zach you
3:55 are so totally awesome thank you please
3:56 tell us about your T-shirt today that is
3:58 from Hamilton from the road and truck
4:00 show of hey Hamilton do you remember
4:02 movie hidden figures Catherine replaced
4:05 by man with IBM computer same today yeah
4:09 yeah oh yeah hidden figures yeah exactly
4:12 um yeah that had a lot of other issues
4:14 in it that movie that was a good movie
4:16 um
4:17 yeah lots going on Lots going on so
4:19 anyway if you're new to this stuff go go
4:21 start with that I'll take a look at some
4:23 other things pop your comments and
4:25 questions down below I do try to answer
4:28 all of them a little bit about my
4:30 background I am an entrepreneur I am a
4:33 uh
4:34 the founder and CEO of a company called
4:36 storyvine which is an automated video
4:38 storytelling platform we do a lot of
4:40 work in healthcare and Pharma
4:42 it's an 11 year old company our our
4:45 automated video platform actually does
4:47 not use AI to do the video automation
4:50 the editing automation believe it or not
4:53 we did it the old-fashioned way with
4:55 programming and code
4:58 um it's not just magically assembled but
5:01 we're rolling a lot of AI into that
5:03 platform and then the other the other
5:04 thing that that you know is relevant in
5:07 my background is I was an entrepreneur
5:09 in the early days of the World Wide Web
5:11 so in 1994 I co-founded an agency called
5:15 agency.com which is one of the very
5:17 first digital agencies and the parallels
5:20 between the early days of the web in the
5:22 early days of this stuff are insane it's
5:25 crazy it's cow cow it's just like it you
5:28 wouldn't believe it except this is
5:30 moving faster and it's significantly
5:32 more profound in my humble little
5:35 opinion uh so anyway yeah that's that's
5:37 what's going on there what else happened
5:39 today chat GPT announced chat GPT for
5:42 enter prize today so they're rolling it
5:45 out over the next few weeks trying to
5:46 get it to as many people as they can
5:48 there's going to be a self-service
5:50 version for small business big
5:52 businesses are going to have whatever
5:54 they have uh it's likely going to be
5:57 more than 20 bucks a person but I'm
5:58 hoping it's kind of like 30 bucks a
6:00 person at least for small companies
6:03 um it allows uh safety and privacy
6:06 they're not going to train on your data
6:08 they will shortly be rolling out your
6:10 ability to train on your own data so all
6:13 of the startups over the last six months
6:15 that were like but chat gbt it's really
6:17 hard to train on your own data so I'm
6:19 going to make a company to compete with
6:20 them well
6:23 they're gonna just turn that [ __ ] on and
6:25 uh so you'll be able to train uh chat on
6:29 your own data it'll be secure
6:31 apparently you get API credits so if
6:34 you're an Enterprise customer you can
6:35 use the chat GPT front end but you also
6:39 get API credits for the back end so
6:41 interesting fascinating move
6:45 um someone in my comments asked me to
6:47 talk about Watson X which is IBM's entry
6:50 into this I don't think I saw any super
6:52 new news it looks like
6:54 um the most recent thing they've talked
6:57 about with Watson X was
6:59 um about six days ago they announced
7:04 a kind of code assistant for Enterprises
7:07 that kind of specializes in taking
7:09 Mainframe Cobalt
7:12 code and converting it to Java which if
7:15 you're an Enterprise I.T you still have
7:18 to find people that know Cobalt just you
7:20 know like Cobalt was the thing when I
7:22 when I was going into college in the in
7:24 the late 80s and early 90s so uh I think
7:28 it was tired back then
7:30 so so that's interesting
7:33 um but but you know what's an X in my
7:37 opinion is is IBM trying to kind of
7:40 resurrect a brand that was damaged a bit
7:42 because with Watson Health that sort of
7:44 purported to change everything in Ai and
7:47 then and then didn't really go that well
7:48 didn't wasn't a success
7:51 um so yeah so I don't know if you have
7:53 any questions about that
7:55 um yeah Pete Mater Cobalt exactly
7:59 like AI it's gonna turn your Cobalt into
8:02 Java that's some exciting stuff right
8:05 there boy I'll tell you
8:07 although if you're on an I.T group and
8:09 you're trying to manage some Legacy
8:11 Cobalt applications fairly not bad
8:16 uh so zapier interfaces is now obsolete
8:20 um yeah although zappier because it
8:23 connects to other things
8:25 yeah you know
8:28 I I would say the zapier interface this
8:30 thing is probably still valid I assume
8:33 you're not talking about Watson x uh I
8:36 assume you're talking about uh chat GPT
8:38 Enterprise uh anyway let me let me jump
8:41 in and show you
8:44 oh Siri
8:48 I I never said that those two words
8:50 together that invoked her I didn't do
8:53 that I didn't do that you were here
8:55 you listen to me all right so this is
8:58 what I was talking about earlier if
8:59 you're new to this AI stuff uh go to
9:02 these things go to this top one that's
9:04 the official chat GPT website Claude is
9:07 from a company called anthropic and it
9:10 has a hundred thousand token context
9:12 window
9:13 so what that means is you can put in
9:15 really long documents into Claude and
9:17 interact with them like you could put in
9:18 a 50 000 word novel
9:20 and just start asking questions about it
9:22 and it'll tell you Bing is Microsoft's
9:26 entry into this space Bing chat
9:30 um okay Farmers here
9:32 Stevo's here
9:34 um what else we've got Bard perplexity
9:38 pie and Poe are also four things that I
9:41 think are worth exploring Labs not
9:43 perplexity lets you play with metaslama
9:45 2 models uh that was something that I
9:48 saw
9:49 um about Watson X also
9:52 is that they announced I don't know two
9:54 weeks ago no maybe a week ago
9:57 you're going to be able to
10:00 um play with llamas llama 2 models on
10:03 Watson X so uh they're getting into the
10:06 open source game or or they're enabling
10:08 meta to get into the open source game
10:11 um or or be hosted on on their platform
10:13 uh Pi dot AI uh we like to call that
10:16 here at the AI learning lab chatty GPT
10:18 it will talk your little ear off but if
10:22 you want to have a conversation with an
10:23 AI and pretend like you're in a Sci-Fi
10:25 movie go to pie and put it into voice
10:27 mode
10:29 and then poe.com is uh there's lots of
10:32 bots there you can play with lots of
10:33 different models so those are worth
10:34 playing with then in terms image
10:36 generation
10:38 check these bad boys out feel free to
10:40 screenshot this it's like it's a little
10:42 big hang on all right come on come on
10:46 there you go there you go there you are
10:49 all right uh so you got mid-journey
10:52 which is the best you got idiogram which
10:54 is brand new and can do text so if you
10:58 want to make images with actual text in
11:00 them that's how this image was made
11:03 with ideogram
11:06 and then leonardo.ai and clipdrop are
11:09 both stable diffusion prompt book this
11:12 openart.ai prompt book is a primer
11:15 that'll teach you how to talk to these
11:16 things and then futurepedia.io is a big
11:20 old directory of AI apps and finally
11:24 not
11:26 last last but least uh we have a scammer
11:30 and the scammer has stolen all of my
11:33 content and the name and put two L's in
11:37 front of learning so if you see someone
11:39 that looks like me especially after
11:41 these lives
11:43 he tends to reach out to people and say
11:45 hey I just saw you in the live let me
11:47 try to scam your ass
11:49 so if someone reaches out to you trying
11:51 to tell you CR sell you crypto [ __ ] or
11:53 befriend you
11:56 um I would say we should have a safe
11:58 word but the fucker's probably in here
12:00 so he's gonna know the safe word so
12:03 um but you know what's funny he copied
12:04 all of my videos except for the video
12:07 where I talk about him huh isn't that
12:11 funny
12:12 anyway I've reported it to tick tock
12:14 many many times if you're from Tick Tock
12:16 and you haven't taken it down what the
12:17 [ __ ] uh
12:20 and uh anyway for those of you that have
12:23 reported him and unfollowed him please
12:26 thank you for that and and if you uh if
12:29 you have followed this person or that
12:33 person has reached out to you just be
12:34 really careful and do not buy any
12:36 multi-level marketing [ __ ] for me or
12:38 anyone else for that matter all right so
12:42 let's get this party got it let's do
12:44 that let's do this like one of the
12:46 things I do I'll answer a question you
12:48 put this question questions in the
12:50 kitchen in Nevada and then I answer them
12:52 okay
12:58 I like turtles
13:00 all right
13:02 so yeah pop the questions into the
13:05 comments below there I got the tongue
13:06 working good Lord all right
13:09 welcome everybody Silver Fox good to see
13:12 you hair growth cake what's back Dr Jay
13:14 Stevo Thelonious it's a party by the way
13:17 the folks that come in here night after
13:19 night after night we call them the
13:20 Irregulars
13:21 why because they're little uh they're a
13:24 little touched they're a little touched
13:25 but
13:27 consider the source consider the channel
13:30 we're we're a little left to Center here
13:33 and I don't mean in politics get off
13:35 that good Lord everyone's so freaking
13:38 sensitive these days
13:40 how about you hug a puppy
13:42 hug a puppy
13:44 we don't talk all that crap in here I
13:47 don't like it oh if you're gonna troll
13:49 me I don't mind trolls trolls are cool
13:51 trolls
13:52 at some point trolls will get a hug and
13:55 they're like that was kind of an [ __ ]
13:56 back then but you know right now you do
13:58 what you need to do but
13:59 Be Clever about it if it's just a put
14:02 down like last night trim your nose hair
14:04 I mean okay sure I needed to trim my
14:07 nose here but that's not the point I did
14:09 by the way it's solid now it's there's
14:11 nothing good look look I'm I'm quaffed
14:13 I'm I'm trimmed I Fly Away hairs they're
14:16 all handled but anyway it was not a
14:19 solid troll
14:21 and just you know calling me an old man
14:23 old man teaching AI get off Tick Tock
14:27 you [ __ ] lose it like really
14:31 doesn't mom have some SpaghettiOs
14:33 heating up for you or something
14:34 now if you're gonna troll me here here's
14:37 the kind of trolls I like they shouldn't
14:39 at first seem like a compliment and then
14:42 and then the more you read it you're
14:43 like ouch that kind of stings
14:46 so I'll let you figure out how to do
14:48 that you can have chat GPT help you with
14:49 that anyway I want to encourage trolling
14:55 and breathe everybody welcome welcome to
14:59 the AI learning lab
15:01 we're gonna do a little ads admsr 88 SMR
15:05 amsr am a a
15:09 welcome thank you Robert Robert Rossi
15:11 thank you for the donut I like donuts
15:14 donuts are tasty do you like donuts
15:17 I'm sorry Dave I'm afraid I can't do
15:19 that yes welcome welcome welcome welcome
15:22 all right let's get to some questions
15:25 let's do it
15:28 um when and where is the meet up next
15:30 week the Meetup is actually tomorrow
15:32 and if you go to the if you go to
15:37 the salon dot AI there's a link tree
15:40 there and there's a Meetup link the
15:43 second link down is is the Meetup so if
15:46 you join the group and you'll see the
15:49 wait did I I might not have put the
15:52 featured tomorrow's meeting on the
15:54 Meetup site hold on let me see if I'm a
15:57 dum-dum
15:58 it's possible it's very possible
16:01 meetup.com
16:06 all right there we are
16:09 Tuesday August 29th
16:12 feature the event
16:14 confirm
16:15 yes so go to meet up
16:18 and
16:20 let my not now we don't want that
16:23 yeah so the the information is there so
16:25 just basically RSVP to the salon
16:30 um tomorrow night and then tomorrow
16:32 we'll send login information if you're
16:35 not in Denver and we'll send physical
16:38 location if you are in Denver and you
16:40 can come hang out and we can talk about
16:42 AI stuff
16:43 and uh yeah that's fun it's a fun time
16:46 it's a fun good time how about all
16:49 right all right all right all right all
16:52 right all right all right
16:55 I asked chat GPT where I might like to
16:57 live in Italy with certain conditions
16:59 and how did it do
17:02 zero like Zack zero Pippins oh shared
17:05 the life oh that would be nice actually
17:06 if you're willing to
17:08 um share this live let people know
17:10 you're an AI weirdo watching another AI
17:12 Weirdo And if you like but Kyle I don't
17:15 want people to pick on me I don't want
17:17 them to know that I like AI well that
17:19 makes you a secret cyborg
17:21 don't be a secret cyborg come out of the
17:24 Shadows yeah you curious little George
17:30 um
17:31 by the way if you are a secret cyborg
17:33 you should probably go read the article
17:35 by Ethan Malik called
17:37 um detecting the secret cyborgs which is
17:40 about people at work that are getting
17:42 secretly AI literate while their
17:44 companies remain Clueless
17:47 pretty special it's cool it's a really
17:50 interesting article uh it was from if it
17:53 was from a bit ago thank you Stevo
17:55 appreciate that speaking of similarities
17:57 do you think we are actively in an AI
18:00 bubble
18:01 that's an interesting one it's there's
18:04 certainly a lot of hype here
18:06 um
18:07 so so with the World Wide Web
18:10 I started agency.com in late 1994 and it
18:15 wasn't until
18:22 probably late 97.
18:25 that the real hype like the hype train
18:28 it was like mid to late 97. so in that
18:33 case it was three years from when it was
18:35 kind of simmering and every you know the
18:38 the the the Geeks were kind of talking
18:40 about oh this is cool hey you know that
18:42 and and you know it was sort of it was
18:44 like inside baseball I still feel like
18:47 with this AI stuff we're we're still a
18:48 bit in that inside baseball I don't I
18:50 don't think it's really hit yet for a
18:52 couple of reasons I think these tools
18:54 right now are still very spit and duct
18:57 tape
18:58 um
19:00 you know it was about the time Microsoft
19:03 decided to go all in on the internet and
19:05 and like like there there was there was
19:07 just a lot more momentum a lot more
19:09 players
19:10 now this stuff why AI is is moving
19:13 faster is
19:15 it in 94 we still had to build all of
19:18 the infrastructure for the internet to
19:20 take off right in 94 we were still using
19:22 dial-up modems 13-inch cathode ray tube
19:26 monitors
19:28 um
19:29 um you know if you were lucky you had a
19:31 T1 you know internet line into your
19:34 business building
19:36 um
19:37 just none of the infrastructure was
19:39 there right so so that took years to get
19:42 all of that all of that that up and
19:44 running
19:45 um so as as these tools get more
19:48 sophisticated and more consumer friendly
19:51 um the infrastructure is already there
19:53 right and and it's not even just are
19:55 these tools gonna get more friendly it's
19:59 going to be things more like Microsoft
20:01 is going to roll chat GPT into office
20:04 365. so everyone that uses Microsoft
20:07 Word is all of a sudden going to have
20:08 that button that says write it for me
20:10 and so people that don't even really
20:12 know what chat GPT is are going to start
20:15 using that little button right Ethan
20:18 Malik did a piece on that as well like
20:19 are you going to push the button when
20:21 these when these things are in all of
20:22 the tools that we use are you going to
20:25 push the button and does it matter if
20:26 you did
20:30 is it more important for you to take two
20:32 hours to craft that recommendation email
20:35 for that student to get them the job
20:37 or if you press the button and it was
20:39 done in 30 seconds and they get the job
20:42 is that the goal
20:44 which one's the goal
20:47 um
20:48 so
20:49 I think what else
20:52 are we in a bubble I don't think so I
20:55 don't think we are yet
20:56 um
20:57 I think it's very very hyped
21:01 um I think it I think I think there's
21:03 more awareness of this so Chachi PT got
21:06 to 100 million users in six weeks the
21:09 World Wide Web it took six years
21:12 six years to 100 million users for the
21:14 World Wide Web six weeks for that tool
21:17 fastest adoption of software in history
21:21 um
21:23 so I think the adoptions dramatically
21:25 quicker I think the tools are
21:27 dramatically more powerful
21:36 I think it's gonna be next year I think
21:38 it's there's gonna be some combination
21:40 of
21:41 the 2024 election here in the states the
21:44 presidential election there's going to
21:45 be a lot of AI generated crap I think
21:47 that's going to make the news and so
21:49 people are going to start talking about
21:50 it more
21:52 um
21:54 I mean we're we're still kind of in this
21:57 phase
21:58 they're still kind of in this phase
22:11 Mark with the A and then the ring around
22:14 it at see that's what I said
22:17 um Katie said she thought it was about
22:19 yeah oh but I've never heard it around
22:21 I've never heard it said I don't even
22:23 see it in the mark but never heard it
22:25 said and then it sounded stupid what I
22:26 said it violence at NBC
22:31 there it is violence at NBC GE com and
22:36 then look the at symbol they didn't even
22:37 have the at symbol in their in their
22:40 Graphics package so they drew a circle
22:42 with an a in it
22:44 violence at nbcg Ecom
22:48 I mean well Allison should know what is
22:52 the internet anyway what is internet
22:54 anyway you know I think we're still
22:57 we're still kind of in this phase where
22:59 where
23:00 people are very you know
23:03 miseducated uneducated about it like I
23:06 think a lot of people have heard of AI
23:08 but most people assume that you know AI
23:12 means the robots are gonna kill us
23:13 because that's what Hollywood's told us
23:18 um
23:20 now I think it's gonna be late next year
23:23 early 2025 I think it's gonna get frothy
23:26 and insane and I think it's I and I
23:28 think it's like the uh the markers are
23:32 gonna be things like they're gonna
23:33 there's gonna be either big news about
23:36 good things that have happened or bad
23:38 things that have happened or layoffs
23:39 that happen
23:41 um
23:42 I think it's going to get much much
23:44 frothier as as as sort of Peppy as it is
23:46 right now oh thank you for the crown
23:49 very nice Stevo
23:51 very nice
23:53 um and what's what's Curious to me is
23:56 yeah it's hyped it might be over hyped
23:59 it might be under hyped right I think
24:02 that these things are what they do and
24:05 what they are are profound on a pretty
24:07 deep level they change the way we
24:10 interact with computers fundamentally
24:11 they fundamentally alter how Engineers
24:15 deal with computers how human beings
24:17 deal with computers how business people
24:18 deal with computers
24:21 um we can't see it yet because they're
24:22 often these crappy tools off in the
24:24 corner but as these things sort of
24:26 assimilate into the the sort of fabric
24:28 of all the tools we use
24:30 um I think that it's going to be pretty
24:32 profound so so really fascinating meta
24:35 AI released a new translation demo oh
24:37 interesting did they need to go see that
24:41 meta
24:43 hey oh yeah that was yesterday wasn't it
24:46 translation
24:52 seamless m4t meta has created an AI
24:55 model seamless m4t that can translate
24:57 and transcribe up to a hundred languages
25:00 across text and speech
25:03 seamless m4t is the first all-in-one
25:06 multilingual multimodal AI translation
25:09 and transcription tool multilingual
25:12 multimodal AI translation and
25:14 transcription tool everybody
25:16 multilingual multimodal AI translation
25:19 and transcription model good Lord could
25:23 you please get at least one more
25:24 catchphrase in there please
25:26 could you make it maybe make it an
25:28 acronym
25:29 it's the world's first mmattm
25:33 [Laughter]
25:37 yeah that's pretty outrageous I I mean
25:39 so
25:41 just like the so this is a really good
25:43 example right of
25:48 if you geek you kind of know what that
25:51 means right if you're a geek like okay
25:54 multilingual multimodal multimodal
25:57 meaning it can listen and say words
26:02 in a hundred languages and it translates
26:05 and transcribes
26:07 well
26:08 they're giving this away they're open
26:10 sourcing this so developers are going to
26:13 be able to make like the little thing
26:15 you stick in your ear that just listens
26:17 to someone talk to you in whatever their
26:19 language is translates it to you you
26:22 talk back to them in in my case in
26:24 English and their little earbud will
26:26 translate it back to them like those
26:28 this is the foundation for those kind of
26:30 tools like all those star treky kind of
26:32 things that when we were kids we were
26:34 like yeah that's never going to happen
26:35 in our lifetime uh [ __ ] yes it is
26:38 right now like it happens with text
26:41 right now you can
26:43 I can do like even in here if someone
26:46 gives me a comment in the in the text I
26:49 can literally just hold it down and it
26:50 translates it good enough that I can
26:52 read it it's not that weird janky-ass
26:54 translation of your
26:56 remember like five years ago when
26:58 translation and transcription sucks so
27:00 bad you couldn't use it not anymore
27:03 crazy crazy it's crazy I tell you it's
27:06 just crazy it's crazy crazy
27:11 you know what else you know what else do
27:15 you know
27:21 the joy of it all
27:23 we Google will have some fun
27:26 announcements oh paint modern PhD this
27:30 week at next
27:32 oh Pate connect with me you work at
27:35 Google do you know Ben Jones there he
27:36 works over in the YouTube side on the
27:38 advertising group
27:40 um he's a good buddy of mine he's a
27:41 great guy
27:42 um
27:43 yeah I is it could it be related to
27:47 Gemini by any chance paint we any Gemini
27:51 dude can you break any news here in the
27:53 AI learning lab tonight that would be
27:55 fantastic we slash Google that that's
27:59 solid I like it it's solid we have some
28:02 fun announcements this week at next all
28:03 right I'm gonna pay attention to that
28:06 so so one of the things those of you
28:08 that have been in here for a long time
28:10 you know
28:11 that for many many months
28:16 I unapology
28:19 unapologetically made fun of Google
28:21 because Bard was a piece of poo poo and
28:24 then I put Bart on my list like a month
28:26 ago because all of a sudden Bard got
28:28 kind of better and I initially thought
28:29 it got better enough paint that I
28:32 thought
28:34 um
28:35 that I thought it was Gemini because it
28:37 like I don't whatever you guys did Bard
28:39 went from being like laughably bad to be
28:41 like oh this is actually pretty good and
28:43 it's connected to the internet
28:45 and how you're incorporating hyperlinks
28:47 into it is actually pretty solid and
28:48 really interesting using web previews
28:50 and things like that so I'm like okay
28:51 something's up and then I started
28:54 hearing about Gemini I'm like it's got
28:55 to be Gemini and then I said that in a
28:57 video and a bunch of people corrected me
28:59 and said it's not Gemini you're a big
29:01 dumb dumb
29:03 so anyway uh yeah Gemini and then what
29:07 else did you do that impressed the [ __ ]
29:10 out of me style drop
29:12 if style drop is in what you're
29:14 announcing this week holy crap that's
29:16 impressive if have you all seen style
29:17 drop is it style drop let me make sure
29:19 I'm not putting my stupid foot in my
29:22 stupid mouth
29:24 style drop Google yes
29:28 this thing is insane it's just crazy
29:32 it's too crazy
29:34 I don't know if you knew this but it's
29:36 just crazy
29:38 is this it yeah this is it this is it
29:42 right here
29:43 so what happens with style drop is you
29:47 upload one image that looks like this
29:49 like this little kid's drawing and then
29:52 all of these are simple prompts in the
29:55 style of that drawing
29:58 and so what I'm going to do is I'm going
29:59 to flip through some different styles
30:01 and just check out like pay attention to
30:04 the Moose the hat and the robot these
30:07 three as as we change so not only is it
30:10 picking up like the texture of the paper
30:12 and that these are cartoon drawings but
30:14 like the style of the like the shape of
30:16 the thing itself
30:18 Alters check this out
30:21 so there's like a van van Gogh a van
30:24 Gogh one right so the robot down here is
30:26 just like a dude the Hat's floating in
30:28 the little sky
30:30 there's a bunch of these
30:33 look at that one that's like that that
30:34 90s kind of you know icon style look at
30:38 the Moose it's like hanging on a wall
30:41 and then here's like a flat art style
30:45 here's a moose like standing up with the
30:47 little
30:48 graphics on it the little leaves
30:51 look at the stickers that's [ __ ]
30:52 insane you put up the little tree with
30:54 the sticker and it does all this stuff
30:57 so this is insane but we can't play with
30:59 it yet pay do we get to play do we get
31:02 to play
31:04 I'm all excited now I hope you're saying
31:06 stuff down below
31:07 anyway welcome welcome to the AI
31:10 learning lab see we've been learning I
31:12 try to keep it honest in here
31:13 you know I bash the crap out of uh out
31:16 of Google for the first six months of
31:18 the year and now I've been nicer because
31:20 quite frankly
31:23 you got your [ __ ] together it's like
31:26 Chachi BT was a bit of a wake-up call I
31:28 know you got some smart people there you
31:30 invented the [ __ ] Transformer
31:33 it's like gotta be something in the
31:36 mothballs back there
31:39 so I'm excited about that cool Stevo
31:42 what's up dude hi everyone what's
31:43 happening so zapier interfaces is now
31:46 obsolete I don't think it is actually
31:48 woo cobal
31:51 um
31:53 oh I was already here wait oh oh just
31:56 Tick Tock just jumped me to the bottom
31:58 welcome everyone my name is Kyle Shannon
31:59 this is the AI learning lab
32:02 and see if it weren't the AI learning
32:06 lab could I do that
32:08 I have a scammer he can't do that he
32:10 doesn't have that graphic although he
32:11 probably just screenshotted that he's
32:13 like oh got it I got it now you want to
32:15 buy my multi-level marketing crypto
32:17 thing it's going to be fantastic
32:22 one of the Irregulars started
32:25 interacting with him and he goes uh yeah
32:27 but your name's not Kyle Shannon he goes
32:28 it's Chris Kyle Shannon yeah Kyle's my
32:31 middle name yeah that's it
32:34 [ __ ]
32:36 Winston Winston is here from early
32:39 adopters life tends to follow wait where
32:42 am I I lost my place I'm going back
32:44 going back hold please hold please I'll
32:47 be answering all of the questions all of
32:49 them is the way I like to do it in here
32:50 frankly Madam I don't believe frankly my
32:53 damn I don't give a deer
32:55 um
32:56 anyone looking at nose hairs has issues
32:58 I know maybe they're attracted to my
33:00 little nose they were like oh those
33:02 little hairs are sticking oh it's very
33:04 sexy yeah I don't know what was going on
33:07 there there was a situation
33:11 this whole Channel's a situation
33:15 oh joker I'm here for a bit not working
33:18 tonight but I need to chop up a tree at
33:21 10 pm Joker dude
33:24 I I'm assuming you live in a cabin with
33:27 no neighbors because chopping up a tree
33:29 at 10 pm I assume that's going to piss
33:31 someone off unless it's just the
33:33 raccoons
33:36 all righty all right all righty all
33:39 right all right all right
33:41 um
33:42 wait his trading platform is real weak
33:45 oh oh yeah the the scammer there's a big
33:49 surprise
33:50 he's got a he's got a shitty trading
33:52 platform that he tossed together
33:53 probably with chat GPT code
33:56 no I just the the the the the way the
33:59 scammer is interacting right now and by
34:01 the way if you're here hey how's it
34:02 going go [ __ ] yourself
34:04 um
34:05 the way he's interacting it feels very
34:08 very much like nft crypto bro kind of
34:11 stuff so and so it and he's selling
34:13 crypto stuff so I have a feeling it's
34:15 probably some [ __ ] tossed together
34:17 piece of [ __ ] wallet draining
34:21 whatever you do don't give this guy your
34:23 wallet
34:24 don't don't do any transactions with him
34:26 with crypto
34:28 because you know I don't know if you
34:30 played around with the crypto but if you
34:32 interact with the wrong contract zip out
34:34 goes the [ __ ] in your wallet
34:36 it happened to me once it's [ __ ] yeah
34:40 it's awful
34:42 um what do they call a damaged pair of
34:44 pants at Sears Irregulars oh that's
34:46 right the Irregulars so so now we've got
34:49 we've got we've we've got
34:53 um gastro gastro meaning of it we got
34:57 you know the neuro spicy meaning of it
34:59 now we've got the damaged goods meaning
35:01 of it see I'm loving the Irregulars by
35:04 the way
35:05 um if you're new here if you're just
35:07 like wait who's this guy what is he
35:08 talking about and why are people liking
35:10 the channel like this it's be it's not
35:13 because they like me don't don't mistake
35:15 this for flattery no no
35:17 the Irregulars have decided that well
35:20 they're going to hang out and listen to
35:21 me just Babble but they have to
35:23 entertain themselves so they've said
35:26 internal goals to give me get me up to
35:28 the the record so far is like a hundred
35:31 and one thousand likes and like a two
35:32 and a half hour meeting then they all
35:34 injured their wrist they got all carpal
35:36 tunnel we haven't hit it since so but
35:38 you know
35:40 listen they're having a good time just
35:42 let them be that's what that is it's got
35:44 nothing to do with me so
35:47 uh but it is it's a lot of fun and it
35:49 makes me smile it makes me very happy
35:51 hello Irregulars
35:54 spaghetti
35:56 yeah see I control people back I asked
36:00 chatty bot not to be so long when did
36:02 she agreed with a very long-winded reply
36:04 I've done that too Joker I was like
36:06 don't talk anymore just give me the list
36:09 I asked for oh boy I can sense your
36:12 frustration I you know I would be
36:13 frustrated too if if someone just talked
36:16 on and didn't give me the list that I
36:17 wanted I'm like lady
36:20 Ah that's crazy anyone looking in those
36:23 hairs has issues we did that one
36:24 dungarees even Huskies exactly do you
36:28 remember the episode of Taxi where Louie
36:34 um
36:35 I think he was talking to Mary Lou
36:37 Henner the red-headed woman on Taxi
36:40 and they were having this this moment
36:42 and he got really emotional and he
36:44 talked about
36:45 how humiliating it is for him to buy
36:48 pants but because he has to go to like
36:50 the the
36:51 um
36:52 the boys section at Sears and ask for
36:56 Husky pants because he's so short and
36:59 fat that he had to get husky boys pants
37:01 I remember because that was the funniest
37:03 show like I would just cry laughing at
37:05 that show that was one of those episodes
37:07 where I was just like oh [ __ ] like the
37:10 just it was just you know you kind of
37:13 fell in love with all these characters
37:14 and they they kind of you know
37:16 occasionally revealed the humanity of
37:18 the characters that was an amazing
37:19 episode
37:22 um like [ __ ] seared into the brain uh
37:25 I walked my mom to the library and back
37:28 home they know her there
37:30 they hold books for her oh that's cool
37:32 that's very nice
37:34 see okay here's the other thing you know
37:37 what we like to do here at the AI
37:38 learning lab like oh thank you very much
37:41 share bear
37:44 um
37:45 we like to be nice to each other like as
37:48 salty and smartasses I get like I like
37:51 this whole channel is about trying to
37:53 help people figure this [ __ ] out and
37:55 like the people that are in here
37:57 regularly like they help each other out
37:59 in the comments they support each other
38:00 they've been supporting me trying to
38:02 like you know help an old man get better
38:04 at Tick Tock and have the channel be
38:06 better and deal with this [ __ ]
38:08 scammer like it's super nice so it's
38:11 like you know Thelonious is awesome and
38:13 he walked his mom to the library
38:16 we need more of this in the world
38:21 less of the divisive derisive [ __ ]
38:27 you know every time we do that divisive
38:30 [ __ ]
38:31 we're playing their game
38:37 you know that right that's by Design
38:40 that we hate each other and not our
38:42 design
38:44 so yeah I broke chatty Bot toys today
38:47 that's good oh no I bought a 69 ticket
38:50 for the AI Salon in Des Moines tomorrow
38:52 that's that's a shame it's a different
38:54 Salon uh you know that that one's a you
38:57 know it's a bit more Hands-On that one
38:59 but but it might be fun you might enjoy
39:01 it so
39:03 Salon meet is tomorrow that's correct
39:07 Thelonious you're a model son felonious
39:09 is they won't give her books she doesn't
39:11 request because as the librarian says
39:13 she's probably already read it oh it's
39:15 really funny
39:18 a 286 for only nineteen hundred dollars
39:21 19.99 286 that's a desktop 286 CC
39:27 what's that line in in uh Mr Mom
39:30 he's like yeah you're gonna run at 220.
39:32 I had 220 221 whatever it takes they
39:35 sound like a real Handy guy there
39:37 uh by the way I'm quoting movies that no
39:40 one will know the references to I worked
39:42 in a movie theater when what what played
39:45 Scarface Mr Mom
39:49 um I don't know whatever is playing with
39:50 when those two were playing I saw those
39:52 movies I saw Scarface probably at least
39:56 a hundred times
39:57 I knew every line in that film
40:00 fairly major crush on Michelle coked out
40:03 Pfeiffer
40:04 Bill Gates David Letterman internet
40:06 Youtube short yep that was pretty funny
40:09 uh I began with the p IBM PC Junior then
40:12 the XT then the at see I was a Mac guy I
40:16 had a Mac 512 ke then I had a
40:19 se-30
40:21 with the color monitor upgrade
40:23 I had an external 13-inch CRT with my
40:27 se30 had that for years
40:30 then I upgraded to a Max 660 AV it was
40:33 like a pizza box and that was where I
40:36 coded the first issue of urban desires
40:38 which was one of the first digital
40:39 magazines back in 94.
40:42 and then uh I don't know I've had a
40:44 bunch since then at one point I had the
40:46 20th anniversary Mac that was pretty
40:48 slick but it was kind of useless but it
40:50 was it was really cool looking
40:52 um I want to even get into interrupts
40:54 that really a lot of money is riding on
40:56 AI a lot of money is riding on AI right
40:59 like just just chatty GPT chatty GPT
41:03 inflection the company behind Pi dot AI
41:05 they just raised 1.3 billion dollars
41:07 open AI is raised well
41:10 they've raised more than 10 billion
41:12 Microsoft gave them 10 billion I think
41:14 most of that was in compute but still
41:17 that's a lot that's a lot of [ __ ]
41:19 compute
41:20 uh
41:21 um
41:22 we're in the beginning of stages of
41:25 acceptance and development
41:27 um yeah couldn't speak confused her this
41:29 is not really an answer to how long it
41:31 would take
41:32 classic clip wait I pick an audible
41:35 audiobook each month the one I chose
41:38 today was called super intelligence is
41:40 awesome oh interesting
41:43 um trying to intro my mom to AI I said
41:46 you can scan terms of service why would
41:48 I want to do that
41:51 that's awesome
41:53 life tends to follow sci-fi oh yeah yeah
41:57 well we're definitely entering a
41:59 post-sci-fi world aren't we I mean
42:03 I did I did a fun thing I went I went to
42:06 three different I went to Chachi PT
42:08 Claude and
42:12 one of the other ones maybe it was bard
42:15 and I said
42:16 um
42:18 list list
42:20 um
42:23 the top 20 Sci-Fi movies
42:26 um that that mention anything about Ai
42:29 and then and then make a column for like
42:32 have we achieved it yet or if we haven't
42:34 achieved it when do you think we'll
42:36 achieve it that was pretty interesting
42:37 little exercise because
42:39 a lot of those movies we can sort of
42:41 check the box like yeah we can do that
42:43 now or like these tools are close enough
42:46 that we're going to be able to do that
42:47 very shortly
42:48 um if you're new here everyone and and
42:50 not quite sure where to start with this
42:52 stuff and what I'm talking about like
42:53 what are you going on about this stuff
42:56 um go to chat.openai.com that URL right
42:59 there that's that's the official chat
43:00 GPT website if you haven't played with
43:02 that gbt
43:04 go play with it and you're like but my
43:07 friends keep talking about Ai and it's
43:08 really annoying and I don't want to and
43:10 it's going to steal my job and the
43:11 robots are gonna kill us yeah if you
43:14 like that
43:15 just
43:17 take a breath in through the nose out
43:20 through the mouth
43:22 and go there
43:25 and start playing
43:27 and you're like Makai I don't know what
43:28 to type okay then go there
43:31 prompts.chat that will teach you how to
43:34 talk to these things
43:36 I'm not affiliated with any of this [ __ ]
43:38 so I don't really care if you do it or
43:40 not but it is in your interest to go do
43:42 it
43:42 in my opinion it it is worth every human
43:45 being's
43:47 time on this planet to go play with
43:50 these things long enough to have a Kevin
43:53 McAllister moment
43:55 remember Home Alone
44:01 it won't take that long when you're
44:02 playing with these things that all of a
44:04 sudden you realize oh this is different
44:07 oh
44:08 it can do what I do
44:10 oh I didn't know it was that fast I
44:12 didn't know it was that deep I didn't
44:13 know it that it knew that
44:16 ask it to write you a country song ask
44:18 it to write you a computer program
44:22 ask it to give you a table comparing two
44:26 obscure topics that you know something
44:27 about
44:29 watch how fast it does it and how
44:31 quickly it does it and how deep it does
44:33 it
44:35 and just go start playing
44:37 and then if you have questions about AI
44:38 pop them down below I'm very far behind
44:40 I'm 90
44:42 questions behind because I just don't
44:43 shut up
44:45 all right
44:47 this is just the beginning for sure
44:51 CNET has an AI glossary basic terms all
44:54 chat GPT users should know that's cool I
44:58 like it go there that sounds right
45:01 streams on a whole different level
45:03 well yeah this stream is this is yeah so
45:08 here's what happens when you get a guy
45:10 with an acting degree who started out as
45:12 a computer science guy who knows enough
45:14 about computers to be dangerous and
45:16 enough about acting to be dangerous you
45:19 get this [ __ ] insanity and you toss
45:21 in a little ADHD as the as the operating
45:24 system this is like like uh like like a
45:27 PC Junior running Windows 92.
45:32 up here
45:34 but someone put a Mac interface on it so
45:38 it's like it's yeah this Channel's it's
45:41 it's something it's
45:44 you know it's gonna be good though all
45:46 right did you cut your chin shaving what
45:48 color is the face paint wait what color
45:51 is the paint fence um no that's a zip
45:55 I discovered that earlier and it's it's
45:57 annoying
45:58 um it's just it's just it's just a zit I
46:01 did trim my nose hair
46:03 the porches of some sort of shitty like
46:06 1992 green you know that's sort of like
46:10 it's like a less saturated avocado
46:13 not our choice
46:15 and then the
46:17 um fence is
46:19 semi-transparent stain in a lovely Stone
46:23 Hedge color it and it is lovely it's
46:25 like a washed out beautiful sort of
46:27 thing but uh yeah but yeah that's what's
46:30 going on here
46:32 start calling it juggler I am a juggler
46:34 by the way I was a professional juggler
46:36 for a while
46:38 it's so funny it's like whenever we
46:41 drive to Costco there's a there's a
46:43 juggler dude that just at the stoplight
46:45 he's always juggling and so whenever we
46:49 pull up to the stoplight I'm like well I
46:51 got off I got career options honey this
46:53 storyline thing doesn't work out
46:55 it's not funny
46:58 it's funny to me where's my t-shirt
47:01 that's all I'm asking for I know I gotta
47:03 get a t-shirt I gotta get a t-shirt I
47:05 mean we gotta get tea we'll get
47:07 something going listen I'm busy I just
47:10 I'm trying to keep up with the AI and
47:12 then I gotta make a workshop because I
47:14 said I'd make a workshop and then I said
47:16 I'd make t-shirts and then I got a zit
47:18 under my chin and my nose hairs were too
47:20 long it's not easy
47:23 okay
47:25 all right
47:27 uh
47:28 um kathalonius I have ADD but listen to
47:31 audiobooks over and over if I like them
47:33 oh yeah so so that that 80d thing where
47:36 where I will when I get into a song I'll
47:39 just listen to it sometimes for months I
47:42 did it with Martin Sexton with with like
47:44 three or four of his songs I'm doing it
47:47 right now with Jason Isbell
47:49 with uh if we were vampires
47:53 um I did it with Brandy Carlisle the
47:56 joke
47:57 and and then the the eye of the
48:01 hurricane one
48:02 yeah I do that a lot
48:04 I can't read and I can't even audio
48:07 books to me I can listen to a podcast
48:09 occasionally if I'm really occupied with
48:12 like fence painting then I can listen to
48:14 one but if I'm just sitting there
48:15 listening to an audiobook my my brain is
48:18 just like yeah and move on next what do
48:20 you what else you're gonna do you're
48:21 gonna do anything else why don't you
48:22 look at your computer why don't you look
48:23 at computers the TV on why is the TV not
48:25 on shouldn't the TV be on in the
48:27 computer on well wait and can't you look
48:28 at your watch as well so
48:30 it's just it just doesn't work
48:33 style drop style drop is the bomb I
48:36 don't think style drop is live let me
48:40 see if it is
48:46 I think it's just a paper still
48:49 but maybe that will be announced we had
48:52 a little googler in here earlier it
48:53 seems like
48:55 and okay the other time I stumped it had
48:58 to do with mRNA shots and scientific
49:00 method not gonna can't say anything oh
49:03 can't say anything
49:04 pay
49:07 pay why can't you say anything
49:10 listen you got a PhD
49:12 you work at Google you teased us
49:16 hang on one second I'll be right back
49:55 [Music]
51:08 okay
51:52 all right I'm back and back the boys
51:57 were fighting the dog was jumping around
51:59 good Lord oh y'all been tapping that
52:02 screen a lot
52:05 okay I hate demos never that never go
52:08 live I know
52:09 that one in particular the style drop
52:11 thing like I want that live yesterday
52:16 but what that says to me is that
52:17 Google's gonna come out with some stuff
52:20 even though Pete can't tell it's
52:21 anything I don't like it Pete I don't
52:24 like it won't be it I don't like it one
52:26 be it
52:28 all right imagine took forever to come
52:30 to become ideogram
52:33 oh I didn't know that was imagine before
52:34 idea Graham
52:36 retro Punk do you know if that's is that
52:38 built on like a stable diffusion core
52:41 did they train their own model
52:46 am I late no you're a little late but I
52:49 started at like eight like or maybe like
52:52 three but three minutes before eight
52:54 I started right before I ate Emilio's
52:56 wife but welcome
52:58 welcome welcome
53:04 um
53:08 all right let's see
53:11 meta victory Yo cousin D Google got
53:14 their [ __ ] together because they'd be
53:16 late to the AI learning lab
53:20 well you know
53:24 I'm not saying
53:26 that that Google got their [ __ ] together
53:28 because
53:30 we told them to but
53:33 this [ __ ] wasn't together and you know
53:37 we kind of trolled them for a bit and
53:39 then they got their [ __ ] together and
53:40 now they got exciting stuff coming out
53:42 you know
53:44 the lab's got some clout I don't like to
53:46 brag
53:48 but you know when you got it you got it
53:53 it's fantastic
53:55 all right I'm fairly certain Google
53:57 doesn't know who we are but from
53:59 Google's here that's pretty cool
54:01 welcome Pete it's cool having you here
54:04 I'd like to have someone here from Tick
54:06 Tock so I could say get rid of my
54:08 [ __ ] scammer
54:11 would you like to report someone who's
54:12 impersonating you yes ah yeah we don't
54:15 see a problem with that
54:19 okay then why is it a choice
54:23 all right Tesla's 300 million dollar AI
54:25 cluster going live today yes they're 10
54:28 000 gpus that I were ironically ordered
54:33 a week after Elon Musk signed that thing
54:36 that said all AI companies should pause
54:39 their development
54:42 I think what he was saying was hey guys
54:44 wait for me
54:46 everyone should pause order the 10 000
54:48 gpus when do they get here when do they
54:50 get here I'm going to call the company X
54:52 let's do a bunch of AI stuff we good I
54:55 just want to understand the universe
54:56 that gonna work for everyone else
54:59 so yeah yeah Tim so so who's not on on
55:04 on the wacky little list here the wacky
55:06 little AI learning lab list we got you
55:08 know open AI Microsoft anthropic
55:13 Google
55:15 meta
55:17 inflection 1.3 billion dollar raise
55:19 that's quora they're not as big a player
55:22 but that's an interesting Tool uh we we
55:25 still have yet to hear from cohere
55:29 um
55:30 Apple
55:31 X
55:35 is there's some stuff coming
55:38 there's some stuff coming it's gonna be
55:40 good sounds like we're going to get
55:42 something this week go down to the Bayou
55:44 get some of us some of that Google stuff
55:46 they're gonna come out and they're gonna
55:47 go oh open air are you so cute
55:50 you're so cute with your little chat gbt
55:56 we just cut a hundred million dollar
55:57 deal with the New York Times isn't it
56:00 funny that they're threatening to sue
56:01 you now it's such a curious little thing
56:04 hmm
56:06 I wonder why that is
56:08 [Laughter]
56:13 oh Joker Matt it's an electric chainsaw
56:16 it's kind of quiet Joker's out cutting
56:18 down trees at 10 o'clock at night he's
56:20 he's really listened to us with an
56:23 earbud and he listens to us every now
56:25 and then in between shops Jenna how do
56:27 we sign up for access to style drop well
56:29 you're gonna have to pay ask paint that
56:31 Pate said he can't tell us anything
56:33 um the style drop thing is not available
56:35 yet that I know let me let me just go
56:37 confirm that
56:39 uh
56:40 uh
56:41 Siri is there something else please
56:45 oh God
56:47 it's just not right she is just she
56:51 really doesn't understand how humans
56:53 work
56:54 just the problem go talk to pie you want
56:57 to talk to one of these things that's
57:00 fun go talk to pie uh style drop uh
57:04 let's see demo
57:10 um
57:13 there's a GitHub of it
57:16 yeah no this is just the
57:18 this is just the announcement
57:22 videos of it
57:37 yeah it's all videos it's all videos of
57:39 people like me going style drops gonna
57:41 be so cool it's gonna be so awesome you
57:44 heard it here first
57:45 and then you can't play with it yeah so
57:48 you heard it here first but where is our
57:50 store I want my merch I know one love I
57:54 know I know
57:56 and I gotta do something I got work to
57:59 do
58:00 I gotta prepare for a I'm speaking to a
58:04 very large Pharma company on Wednesday
58:06 morning
58:08 so Wednesday morning at 7 30 a.m send
58:12 kind love my way I will be waxing
58:15 philosophic about AI to a big Pharma
58:18 company
58:21 and what else I got I don't know we got
58:23 a lot we're preparing for right now so
58:26 but but it's listen it's all on the list
58:28 the good news is I have a list
58:30 the bad news is the list is still
58:32 existing
58:34 Tesla's about to flip the switch on its
58:36 new AI cluster yep they just did it or
58:38 they're doing it today
58:41 if anyone likes crypto analysis
58:43 cryptography David Wong real world
58:45 cryptography is really great thank you
58:47 Winston that's awesome
58:50 um thanks again most great Innovations
58:53 come from artsy drug using living on the
58:56 Fred Fringe lifestyle it do it's well
58:58 that I mean
59:01 one of the reasons I'm so excited like
59:05 I'm excited enough about all of this
59:07 stuff that I started a [ __ ] Tick Tock
59:10 Channel at 58.
59:13 it ain't cause I'm a genius
59:15 is
59:16 the inflection point
59:19 after a tool like chat gbt or after a
59:22 tool like the World Wide Web is a really
59:24 exciting time thank you Stevo
59:27 to Gmail beaver
59:29 that was a lovely Little Beaver there
59:31 Stevo thanks I like a beaver everybody
59:33 likes a good beaver
59:36 um
59:40 prior to the World Wide Web the internet
59:42 was this command line research
59:44 scientists universities kind of tool
59:48 and then the world wide web made it this
59:51 simple right
59:52 and machine learning was similarly
59:54 complicated and you know using it for
59:56 all sorts of specialized things and you
59:58 had to be a mathematician and but today
1:00:00 did it
1:00:01 and then chat gbt comes along and kind
1:00:03 of reveals
1:00:05 this incredible power
1:00:07 in a single little ask me anything box
1:00:09 and um
1:00:11 yeah it's it changes everything because
1:00:14 what happens is
1:00:16 the researchers and scientists and
1:00:18 mathematicians look at these tools and
1:00:20 they're going to use them in a
1:00:21 particular way
1:00:23 and you get some wacky ass artist or
1:00:25 some you know ex-actor that fancies
1:00:28 himself computer savvy
1:00:31 um
1:00:31 people are going to start to do some
1:00:33 [ __ ] you get architects in here and
1:00:36 product designers and
1:00:39 you know wordsmiths
1:00:42 of all kinds they're gonna be some
1:00:44 amazing stuff happening and and just
1:00:47 really unexpected stuff and that's like
1:00:49 that's this is right now's the time
1:00:51 right this is like revenge of the
1:00:53 liberal arts Majors like bring all of
1:00:55 your ideas bring all of you
1:00:58 because these tools can amplify even the
1:01:02 smallest little piece of talent you have
1:01:05 and blow it out and that's what that's
1:01:07 why I think we are entering
1:01:15 that
1:01:18 the great Renaissance
1:01:20 that started in my humble opinion on
1:01:23 that date November 30th 2022.
1:01:27 that's the day Chachi PT launched
1:01:31 and that's the beginning of that in my
1:01:34 opinion
1:01:36 like I think the next 5 10 15 20 30
1:01:40 years are going to be insane
1:01:44 and I and from where I said in a good
1:01:47 way is there the risk of lots and lots
1:01:49 of bad [ __ ] yeah but that's been all
1:01:51 covered I feel like that's it's easy to
1:01:53 go there
1:01:57 what if it gets really good
1:02:02 what if way more humans are fully
1:02:04 self-expressed
1:02:06 and less frustrated and less Souls
1:02:09 crushed
1:02:11 because their jobs are these repetitive
1:02:13 mindless numbing Cog in the wheel even
1:02:17 knowledge work
1:02:18 I have to update my data in the
1:02:20 spreadsheet my TPS report TPS report TPS
1:02:24 report TPS report
1:02:26 that no one [ __ ] reads what if all
1:02:29 that [ __ ] was automated
1:02:32 well then what would I do I don't know
1:02:34 what do you want to do uh
1:02:36 I kinda always wanted to do needlepoint
1:02:39 okay
1:02:40 go do that
1:02:42 yeah I used mid-journey and I come up
1:02:44 with these images to kind of blow my
1:02:46 mind and then I put them in needlepoint
1:02:47 and that's how I do these things and now
1:02:49 I'm in the Smithsonian like I don't know
1:02:51 that kind of [ __ ]
1:02:53 that kind of shit's coming it's coming
1:02:56 it's coming
1:02:57 it's coming it's crazy to come he's so
1:02:59 good
1:03:01 I sent you a DM about a regular
1:03:03 moderator oh awesome Zach I appreciate
1:03:05 that that's very cool
1:03:07 um yeah any anyone who said that they'd
1:03:09 be willing to mod I'll reach out this
1:03:10 week I've got some time this week I'll
1:03:12 reach out and we can organize a time to
1:03:14 talk and figure out what that actually
1:03:15 looks like
1:03:17 um but I think it I think it could help
1:03:18 the channel to have you know some some
1:03:20 actual management so it's not just me
1:03:22 here acting like an idiot and getting
1:03:24 people frustrated that I don't answer
1:03:25 their questions which by the way if
1:03:28 you've been here and you're frustrated
1:03:29 that I haven't answered your questions
1:03:30 that's because I don't have mods yet but
1:03:32 we're working on it man it's gonna be so
1:03:33 good
1:03:35 you don't even know how good it's gonna
1:03:37 be good Lord that's like just a it's
1:03:39 like a mid-journey
1:03:41 um face on the bottom of my chin
1:03:45 taxi was awesome wasn't it the best taxi
1:03:48 was great might have to hit that again
1:03:49 exactly thanks for sharing uh who is Sam
1:03:53 and what did you learn from him I I
1:03:55 assume the Sam that I was talking about
1:03:57 was Sam Altman
1:04:00 what's up
1:04:01 what I was talking about uh Sam Altman's
1:04:04 the CEO of
1:04:06 um openai
1:04:09 wait where is it hold hold hold hold
1:04:13 there now
1:04:16 there
1:04:18 that guy
1:04:20 Sam weltman that guy
1:04:24 he's a he's a very thoughtful fella it's
1:04:27 funny someone was in here the other day
1:04:28 and they're like I met Sam Waltman not
1:04:30 impressed I'm like okay no you know the
1:04:35 dude ran y combinator for a while so
1:04:36 you've seen a lot of startups
1:04:39 um he got Elon Musk to invest 50 million
1:04:42 dollars and then walk away from open AI
1:04:44 he got Microsoft to invest 10 billion
1:04:48 dollars
1:04:50 you know as someone who's 11 years into
1:04:52 a startup
1:04:54 I haven't gotten close to raising 10
1:04:56 billion dollars
1:04:59 our seed round was [ __ ] almost
1:05:02 impossible
1:05:04 so at a minimum
1:05:07 he did that
1:05:10 he says you know Microsoft's about to
1:05:14 announce a billion dollar investment in
1:05:16 us we don't really have anything we can
1:05:18 show the public
1:05:19 why don't you take that that idea we
1:05:22 have in the documentation of gpt3 that
1:05:26 that chat GPT thing
1:05:28 you got two weeks why don't you code
1:05:30 that up
1:05:32 so Sam Altman two weeks before they're
1:05:35 going to announce the billion dollar
1:05:36 investment from Microsoft whips up chat
1:05:39 GPT
1:05:41 and in six weeks it gets to a hundred
1:05:43 million users and it doesn't [ __ ]
1:05:45 explode
1:05:47 and it puts Google on their heels and it
1:05:50 puts Microsoft back on the map
1:05:57 I mean
1:06:01 you can say what you want about him but
1:06:03 so far so good
1:06:05 now you know
1:06:08 is is he gonna be the winner forever and
1:06:10 always of course not like shit's gonna
1:06:12 happen if if pate's correct and Google's
1:06:15 gonna announce something Saucy this week
1:06:18 that puts Microsoft and open AI on their
1:06:21 heels well then we gotta now we got a
1:06:23 boxing match right
1:06:24 you ever watch Boxing
1:06:26 one guy's just [ __ ] pummeling the
1:06:28 other one I can't believe this I can't
1:06:30 believe he knocked the champ down he
1:06:31 knocked the champ down the champ doesn't
1:06:33 have his legs under him and champ
1:06:35 doesn't have oh
1:06:36 the champ just woke up
1:06:38 you know
1:06:40 so so this is going to be a fantastic
1:06:42 fantastic
1:06:45 uh thing Emilio's wait wait we signed up
1:06:48 to help an old man on his channel you
1:06:50 did that's the deal I'm the old man
1:06:53 this is the channel this is it
1:06:56 old man on Tick Tock old man on Tick
1:06:58 Tock yeah you can't get off my lawn
1:07:00 where do I click on it how do I turn it
1:07:03 Martha how do I turn this tick tock on
1:07:06 I'm on the tick tock
1:07:09 I'm on a tick tock
1:07:11 now I'm on The Tick Tock how do I turn
1:07:13 Tick Tock up what oh it's already on
1:07:15 he's done
1:07:17 yeah that's this channel
1:07:22 all right divide and conquer when I
1:07:25 click
1:07:26 the like a lot I keep accidentally
1:07:28 scrolling off the live I know so so what
1:07:32 we did discover the other night if if
1:07:34 those of you with carpal tunnel are like
1:07:36 we're up to 30 000 likes in like an hour
1:07:38 so you guys are amazing is that right
1:07:39 yeah
1:07:40 um so uh apparently from Amazon you can
1:07:45 order an auto clicker an auto screen
1:07:47 Tapper
1:07:48 so or you can just get a raspberry pie
1:07:50 and a little actuator and just put like
1:07:52 the end of you know a little one of
1:07:55 those little uh screen styluses
1:07:58 [Laughter]
1:08:02 um we are one people we are the same
1:08:03 blood yeah exactly yeah we're all just
1:08:06 us we're here's the deal
1:08:10 the comment when I was talking about
1:08:11 like let's stop being so divisive and
1:08:13 [ __ ] to each other but like let's
1:08:15 get back to some empathy and connecting
1:08:17 as humans
1:08:18 here's the deal we're all [ __ ] trying
1:08:21 to figure it out life's hard Life's a
1:08:23 pain in the ass
1:08:27 it's painful
1:08:30 you know when it's more painful when
1:08:32 you're mean to other people and they're
1:08:33 mean to you
1:08:35 so baby just do less of that
1:08:38 that's what I'm in for
1:08:40 all right and I think these tools are
1:08:42 going to help with that oddly enough I
1:08:44 think the robots are going to help us
1:08:46 ReDiscover our Humanity I think it's in
1:08:48 a great twist of fate we are going to
1:08:51 become better people because of the
1:08:53 tools
1:08:54 and you're like but Kyle why
1:08:57 do you find yourself saying please and
1:08:59 thank you to chat GPT I do
1:09:03 I've tried to stop
1:09:05 and then it'll be like nice to be and
1:09:07 I'm like oh thanks a lot I'm like I
1:09:09 don't have to say that but I'm like [ __ ]
1:09:10 it
1:09:12 and then when I go to write a note to an
1:09:14 actual human being you know what I do it
1:09:16 I go oh I should thank them
1:09:18 please do this oh
1:09:22 it's like this shit's reminding me to
1:09:24 like suck less
1:09:29 this is the best Channel lots of cuz
1:09:31 O'Donnell's fun humor because because
1:09:34 cousin O'Donnell that's nice that's nice
1:09:37 I like it
1:09:39 I'm telling you man we're gonna get
1:09:41 Rosie in here one night she's gonna see
1:09:43 me she's gonna be like oh my God am I
1:09:45 looking in the mirror
1:09:47 exactly
1:09:51 we'll hang out we'll party
1:09:53 do you have a Discord Channel I have the
1:09:55 the AI salon
1:09:58 has a Discord Channel
1:10:00 wait that's the AI salon
1:10:04 go to
1:10:07 the salon.ai
1:10:10 and that's a link tree and the second
1:10:12 link goes to our meetups our our next
1:10:15 meeting is tomorrow night and then
1:10:17 um the link below that is a link to the
1:10:19 Discord and so you can go into the
1:10:21 Discord and there's a mid-journey
1:10:22 channel in there and there's all sorts
1:10:24 of different channels the the AI Salon
1:10:26 is something I created back in December
1:10:27 as a as a way to just create a community
1:10:30 of people who are curious about Ai and
1:10:33 trying to figure it out and they're on
1:10:34 the adventure and you know some people
1:10:36 come there and they were you know
1:10:38 machine learning people and now they're
1:10:40 trying to figure out this generative AI
1:10:42 stuff some people are come in because
1:10:43 they're like uh I heard about this AI
1:10:46 I'm scared is it gonna be okay tell me
1:10:48 it's gonna be okay it's not gonna be
1:10:50 okay is it it's going to be okay tell me
1:10:52 it's gonna be okay
1:10:53 sometimes it's that sometimes it's
1:10:55 people like I heard it was cool man I'm
1:10:58 looking for a job Anybody hiring an AI
1:11:01 um
1:11:02 but yeah but what's cool about where we
1:11:04 are with this industry right now I
1:11:06 wouldn't even call it an industry yet
1:11:09 I suppose it is
1:11:10 um
1:11:11 is is
1:11:13 um
1:11:15 well you can get a degree in machine
1:11:19 learning and the science side of AI
1:11:23 the generative AI stuff is so new that
1:11:25 everyone's trying to figure it out no
1:11:27 one knows how to use it no one knows how
1:11:29 to incorporate it into businesses no one
1:11:30 knows where the boundaries are
1:11:33 you know do I marry or do I kill her is
1:11:35 like that that that old joke is like
1:11:38 like you want to avoid this stuff but
1:11:41 you can't so what do you do
1:11:44 and that's just where we are right now
1:11:46 and so people are coming from all over
1:11:48 so you know we have artists and lawyers
1:11:51 and
1:11:52 um you know software engineers and you
1:11:55 know super ultra Geeks and then Super
1:11:57 not Ultra Geeks just trying to figure
1:11:58 this stuff out
1:12:00 you know people come in and going I was
1:12:02 a copywriter and then I wasn't and now I
1:12:05 am again but sort of not really and I'm
1:12:07 confused and I'm having an existential
1:12:09 crisis and I don't know what I enjoy
1:12:10 anymore
1:12:11 and everyone's trying to figure it out
1:12:14 and it's really exciting so that's what
1:12:15 that group's about the the salon the AI
1:12:18 salon
1:12:20 um what was your line in New Jack City
1:12:22 it was
1:12:40 I can't remember the exact it was it was
1:12:42 something like
1:12:43 I'll take an eight ball so I so it's I'm
1:12:46 white boy number one and white boy
1:12:48 number two was in the passenger seat so
1:12:50 I'm driving the car so we pull up to the
1:12:52 Block in Harlem and a black dude walks
1:12:55 up to the car and I say something like
1:12:57 like
1:13:01 oh I don't think I said no no no
1:13:04 I pull up to the block
1:13:06 and some dude walks up to the car and
1:13:08 says something like like what the [ __ ]
1:13:10 are you doing in the neighborhood and I
1:13:12 was like go go go go go go and I floored
1:13:14 the car we pull out we zoom out
1:13:17 and then we shot another scene on a
1:13:19 bridge and I don't know if I forget what
1:13:22 the scene on the bridge was about that
1:13:23 got cut out
1:13:24 but it was just I was just a like a like
1:13:27 a white kid from Connecticut coming into
1:13:29 Harlem to buy drugs it's like
1:13:33 uh hey man how can technically wait
1:13:37 how technically intense do you think it
1:13:39 will be to set up GPT for Enterprise I
1:13:42 don't think it's going to be that
1:13:42 intense at all I think it's gonna be
1:13:44 like turn the [ __ ] on
1:13:46 I think training your own data on it is
1:13:49 probably going to be a bit more intense
1:13:50 but but what it sounds like like it kind
1:13:53 of sounds like like the zapier thing
1:13:55 like like basically you're gonna log
1:13:57 into your system so
1:13:58 so I would assume it's going to be like
1:14:00 I want to train chat GPT on these
1:14:04 directories on my Google Drive or I want
1:14:07 to do them on Dropbox or on like you
1:14:11 know AWS or wherever your content is
1:14:14 residing it sounds like you're just
1:14:16 going to give access to GPT to to go in
1:14:19 there and I assume they've written some
1:14:21 software that's just going to go in
1:14:22 embed your data toss it into
1:14:25 a vector database in some sort of
1:14:28 intelligent way
1:14:29 and then you're going to play now what I
1:14:32 don't know is how sophisticated they've
1:14:34 gotten with the permissions thing right
1:14:37 so if you're even a even a medium-sized
1:14:40 organization you know you don't want
1:14:43 Sally The Intern to have access to all
1:14:45 the financials that the CFO has access
1:14:47 to that would be bad right so all of
1:14:49 that permissioning stuff Microsoft has
1:14:52 announced that they're they're handling
1:14:54 that like they've got that solved
1:14:56 so I would think that one of the deals
1:14:59 that Sam Altman and the openai team cut
1:15:02 is hey when it comes to Enterprise [ __ ]
1:15:05 Microsoft help us solve that so if if
1:15:09 they're calling it for Enterprise that
1:15:12 says to me that they've got their [ __ ]
1:15:13 together if it were like chat GPT for
1:15:17 business or for Biz Biz then it would
1:15:20 probably be less stringent but if
1:15:22 they're calling it for Enterprise
1:15:25 um I would assume they've got it pretty
1:15:26 dialed in but the minute you get into
1:15:28 permissions and [ __ ] like that it's
1:15:30 going to get more complicated but I
1:15:31 would think that just turning it on is
1:15:33 going to be
1:15:35 probably very straightforward
1:15:38 um
1:15:39 but apparently you get faster gpt4 you
1:15:41 get a 32 000 token context window for
1:15:45 gpt4 you get free API credits to build
1:15:48 stuff on their API
1:15:50 um and then you get all the security and
1:15:52 privacy stuff and and they won't train
1:15:54 on your prompts and any of your data and
1:15:56 then someone even said on Twitter I saw
1:15:59 earlier that they think there might also
1:16:01 be a prompt or Rachel Woods of the AI
1:16:03 exchange on Tick Tock said
1:16:06 um there's even a hint of a prompt
1:16:08 library in there so that as you get your
1:16:09 prompts dialed in you can share those
1:16:11 with the rest of the organization that's
1:16:13 pretty slick
1:16:15 do you run metaslama locally I don't
1:16:18 looking for friendly
1:16:20 installation tips no that's why I have
1:16:24 um
1:16:24 that's why I have
1:16:26 labs.perplexity.ai on here because
1:16:28 they're hosting it so you can go play
1:16:30 with the models there for free
1:16:32 um but but I don't have them running
1:16:34 locally the the local thing
1:16:37 what I would probably do is I would
1:16:39 probably run it on a virtual machine
1:16:42 um with a GPU
1:16:43 capability so that I could run the 70
1:16:45 billion parameter model just running
1:16:48 things locally here here's what I'm with
1:16:51 the open source stuff
1:16:52 because things are changing so quickly
1:16:55 and it it really does require you to
1:16:57 spend a bit more time on the technical
1:16:59 side of things it's
1:17:02 while I could slog my way through it
1:17:04 what's going to end up happening is I'm
1:17:06 going to take 15 or 20 hours to get
1:17:08 everything dialed in I'm going to be
1:17:10 really happy with it and then two days
1:17:11 later someone's going to announce just
1:17:13 double click here to install this one
1:17:15 that's half the size and twice the power
1:17:17 and you just wasted all your time so I'm
1:17:20 just kind of avoiding the open source
1:17:22 world right now and just waiting for
1:17:24 someone to put together thank you for
1:17:26 the little llama llama Lucy the llama uh
1:17:31 uh to to I'm just waiting for someone to
1:17:33 put it together so I have I got nothing
1:17:35 for you what I would say snacky is just
1:17:37 go to Reddit and and go to the open
1:17:40 source large language model just search
1:17:43 for open source large open source llm on
1:17:46 Reddit
1:17:46 find some subreddits where they're
1:17:48 talking about this stuff and they just
1:17:50 start hanging out in there and and just
1:17:52 you know they'll point you to GitHub
1:17:53 repositories that make it easy so I'm
1:17:56 sure someone's already done it
1:17:58 it's the AI Salon with two s's yeah
1:18:01 exactly
1:18:02 uh
1:18:04 Ted looks with the panda thank you Ted I
1:18:07 like to Panda I appreciate Panda I go
1:18:09 visit my mom twice a year this visit is
1:18:12 scheduled a hip transplant wow
1:18:14 what are you gonna do
1:18:16 yeah my my wife's mom just got
1:18:19 um the big piece of surgery
1:18:21 and just came home today so all is well
1:18:24 there yay
1:18:27 um they are a eyes wait
1:18:31 chorizo for the ruse by any other name
1:18:35 would smell is sweet
1:18:38 what's that called
1:18:41 that's called overacting
1:18:45 or as they call it in the biz chewing
1:18:48 the scenery
1:18:52 oh wait hang on
1:18:55 I think that's me
1:18:58 yeah
1:19:01 I just set off set off the old yield car
1:19:03 alarm
1:19:04 had keys in the pocket
1:19:07 um
1:19:08 they are eyes that will wait do for you
1:19:10 use check wait
1:19:13 oh there are AIS that will do it for you
1:19:15 that use chat gbt let me give you one of
1:19:18 a few seconds oh cool One Love Is Gonna
1:19:20 help whoever was looking to install
1:19:21 llama
1:19:23 is going to help one love is going to
1:19:26 help that person awesome thank you
1:19:28 that's super cool dad bought us a
1:19:30 Commodore 64 first that's awesome
1:19:33 fantastic start the Commodore 64. that
1:19:35 actually hung hung in there for a good
1:19:37 long time
1:19:40 oh wow
1:19:42 Joker that's intense
1:19:44 sorry for that
1:19:46 um I was there I mean I was down we had
1:19:49 offices right right near there if if it
1:19:52 was if that was the cause
1:19:56 AI startup
1:19:58 intense eye raising funding from light
1:20:01 speed at 300 million
1:20:04 what's it called
1:20:06 intense eye
1:20:10 intense I
1:20:21 here we go
1:20:24 oh out of 300 million dollar valuation
1:20:26 so would they raise 20 or 30
1:20:31 series B
1:20:34 and what do they do workplace AI
1:20:36 software
1:20:39 huh cool
1:20:41 cool yeah it's starting get started I
1:20:43 mean the investments in this space are
1:20:46 are pretty eye-watering
1:20:48 um that one's kind of normal only a 300
1:20:51 million dollar valuation
1:20:53 wait they don't have a hundred billion
1:20:55 dollar valuation and they're a year old
1:20:58 what is wrong with them
1:21:01 um just sign up your company with
1:21:03 crunchbase and you'll see how ai's
1:21:04 changing business
1:21:06 thank you kind of stoic about death
1:21:10 life energy is energy it can't be
1:21:13 destroyed exactly I agree with that from
1:21:16 New York Times AI brings
1:21:18 the robot wingman to Aerial combat
1:21:22 oh interesting
1:21:24 out of curiosity do you or any other on
1:21:27 run in chat run things on tpus I do not
1:21:33 um
1:21:35 someone was talking about tpus in here
1:21:37 the other night
1:21:39 um
1:21:41 but I'm I know I know within the AI
1:21:45 um Salon there are some folks doing some
1:21:47 interesting stuff
1:21:49 but in here
1:21:51 I mean there might be some people doing
1:21:53 that not not talking about it a lot I
1:21:55 don't I tend not to talk too much about
1:21:56 the the the more technical side of
1:21:59 things just because I'm I'm kind of
1:22:00 Clueless there
1:22:03 um ml perf results should be published
1:22:06 in a few weeks oh interesting
1:22:10 all right I saw that net glossary link
1:22:12 today it's good CNET
1:22:15 and you toss in a heavy helping of ADHD
1:22:17 exactly we call it landlord green it is
1:22:20 totally land lord green yeah exactly
1:22:23 it's this is landlord green and then he
1:22:25 also has
1:22:27 um sort of landlord purple trim so it's
1:22:31 like that awful 90s sort of faded green
1:22:34 and Faded purple
1:22:36 oh it looks like a bad handbag
1:22:42 uh I can help with the blemishes
1:22:45 I know her growth geek right and I'm not
1:22:48 I'm not normally a a blemish dude but
1:22:52 just one popped up
1:22:54 what are you gonna do
1:22:57 all right let's see I I am I'm more than
1:23:00 99 messages behind I'm gonna jump to the
1:23:03 bottom and then scroll up a little bit
1:23:04 so I'm gonna skip a bunch of questions
1:23:05 right now because I'm too far behind so
1:23:07 I'm just gonna do it and uh the uh the
1:23:11 uh
1:23:12 the the heart the heart tappers are
1:23:15 tapping away thank you you're up to 33
1:23:18 400 so Kudos good job good gaming
1:23:23 um
1:23:25 if you're new here everyone my name is
1:23:28 Kyle Shannon it's the AI learning lab
1:23:31 and if you've not played with chat GPT
1:23:34 go there right now
1:23:36 chat.openai.com that's chatgpt
1:23:39 and if you don't know what to do there
1:23:41 then go to there prompts.chat so go
1:23:44 there and sign up for an account and
1:23:45 then open another tab or another window
1:23:47 and go to prompts.chat that will teach
1:23:50 you how to use those things
1:23:53 um and just trust me it it will not be a
1:23:57 waste of your time even if you end up
1:23:58 not using it
1:24:00 it is impossible to understand what
1:24:03 these things are without experiencing
1:24:05 them it ain't a Google search
1:24:09 Pate knows this
1:24:10 paint works for Google
1:24:13 's Google these things are different now
1:24:17 they're all based on technology that
1:24:18 Google invented there's the big irony in
1:24:20 it and you know I have a feeling
1:24:22 Google's gonna be
1:24:24 um
1:24:26 making some noise here shortly so
1:24:31 and and here's here's my deal like I'm
1:24:34 I'm technology agnostic I don't give a
1:24:36 [ __ ] who comes up with the cool [ __ ] I
1:24:38 just want more cool [ __ ]
1:24:40 the fact that I've got seven of these
1:24:43 chat like companies on a list that I
1:24:45 think are at least worth your time to
1:24:47 explore is mind-blowing to me
1:24:50 because for the better part of this year
1:24:53 there were only three on that list and
1:24:55 one of them was kind of
1:24:58 now there's seven
1:25:00 and I would think that there's probably
1:25:02 going to be
1:25:03 10 to 12 by the end of the year
1:25:06 and I'll I'll prune but but we're in a
1:25:09 pretty remarkable time right now so go
1:25:11 play go play go play Go Do It Go Do It
1:25:14 Go Do It
1:25:15 go go
1:25:18 don't just sit there
1:25:20 chat dot openai.com write me a country
1:25:24 song about a chubby Dog go
1:25:27 oh I didn't know I could do that
1:25:33 oh no it can't write a marketing plan
1:25:35 can it
1:25:37 make me a marketing plan
1:25:39 oh [ __ ]
1:25:44 go have that moment go have your Kevin
1:25:46 McAllister moment
1:25:49 that's when the fun really begins that's
1:25:52 when you realize oh [ __ ] I can't unsee
1:25:54 this
1:25:55 [Laughter]
1:25:57 there's some things in life you can't
1:25:59 unsee this is one of them
1:26:04 and it ain't getting uninvented
1:26:06 so everyone that's like oh but I heard
1:26:09 opening eyes gonna be bankrupt by the
1:26:11 end of 2024. yeah that's not gonna
1:26:13 happen
1:26:15 they might get sued out of existence
1:26:20 because Google cut all the deals with
1:26:22 all the publishing companies
1:26:25 that could happen but that's just
1:26:26 business
1:26:27 but they ain't gonna run out of money
1:26:31 because there's just too much at stake
1:26:33 I mean and it's not just that Microsoft
1:26:36 invested 10 billion dollars
1:26:41 it's that it's in everyone's interest to
1:26:44 make sure that that thing stays alive
1:26:48 for some amount of time
1:26:51 all right
1:26:54 all right
1:26:55 what's in the fridge why would I do that
1:26:57 she asked all right
1:26:59 Stacy trauma therapist is laughing good
1:27:02 I'm glad we make you made you laugh
1:27:06 um it's very easy to set up in the cloud
1:27:09 very cool
1:27:11 um paid people are dumb and will not use
1:27:13 correct permissions no matter what you
1:27:15 do there's a there's a man who's
1:27:17 frustrated that people aren't using
1:27:19 Force two-factor Authentication
1:27:22 by the way paint the the the one thing
1:27:24 that Google single sign-on is doing now
1:27:27 that I actually kind of like because the
1:27:29 the two-factor authentication with the
1:27:32 [ __ ] authenticator app I want to
1:27:33 throw my phone so hard at so many walls
1:27:36 multiple times a day but the new thing
1:27:39 where it's just like we sent you
1:27:41 something in your Google mail and then I
1:27:43 just look at my phone and it's like is
1:27:45 this you and you just click yes and you
1:27:47 and then you're done like that's good
1:27:49 like actual Improvement in two-factor
1:27:53 authentication yay now we just need to
1:27:55 get rid of the uh
1:27:57 the password altogether and here I'll
1:28:00 show this I'll freak some people out
1:28:01 with this this is my this is my my
1:28:04 little this is my little uh
1:28:07 little AI Flex
1:28:10 d
1:28:15 I am looking forward to when
1:28:19 I can log in with that
1:28:25 that is my world ID verified at orb
1:28:31 what is the orb you ask
1:28:34 well it is a horribly named thing from
1:28:37 Sam Altman's company
1:28:39 called World coin
1:28:41 and the orb is a thing that scans your
1:28:44 retina to get your biometric data to
1:28:47 verify that you are indeed uh proof of
1:28:52 personhood that you are a unique human
1:28:54 being verified on that system
1:28:58 um why did I do that I did that because
1:29:01 I actually think this kind of proof of
1:29:04 humanness is going to be increasingly
1:29:07 important I think it's actually going to
1:29:08 be important
1:29:09 sooner than I think any of us are
1:29:12 prepared for I think the 2024 election
1:29:14 is gonna it's gonna make it it's gonna
1:29:16 be crazy
1:29:18 and
1:29:19 um but if I have a an anonymous proof of
1:29:22 personhood and I'm the only one that can
1:29:24 get that well that could be my ID to get
1:29:26 into sites and I would never have to
1:29:28 remember a [ __ ] password ever again
1:29:29 that'd be nice
1:29:32 can we do that now
1:29:37 first it was have a password
1:29:39 then it was have a password with a
1:29:41 capital letter okay I can hit the shift
1:29:43 key then it was hit the have a password
1:29:45 more than six characters with a capital
1:29:48 letter and a number okay I can do that
1:29:51 then it was have a password that's more
1:29:53 than eight characters with a special
1:29:55 symbol a capital letter and a number
1:29:58 that and it it just doesn't [ __ ] end
1:30:01 and then it's like
1:30:02 um oh I'm sorry you haven't changed your
1:30:05 password in more than three hours you're
1:30:07 gonna have to go ahead and change that
1:30:08 out and then you change it and you're
1:30:09 like well you can't use something that
1:30:11 you've used in the past decade why would
1:30:13 you do that and you're like oh my God
1:30:15 and then and then it's like would you
1:30:17 like me to recommend a password you can
1:30:18 never ever possibly remember okay how do
1:30:22 I remember that you have to remember a
1:30:23 password to this password manager and if
1:30:26 you remember if you forget the password
1:30:27 manager you lose all your passwords but
1:30:29 you can't remember them anyway so don't
1:30:30 lose that one well what password do I do
1:30:32 there
1:30:33 see and and now then now I feel like
1:30:38 Howie Mandel isn't yeah and then now
1:30:43 um now we get two Factor where just
1:30:45 having one password is not enough you
1:30:46 have to have the double password and
1:30:48 then you have the gotcha the captcha
1:30:50 with the look at the bike and which
1:30:52 one's the bike and it's it's like
1:30:53 they're increasingly obscured because we
1:30:56 got Machine Vision they can see them now
1:30:57 so so now the bike is like this big in
1:31:00 16 different spots on the thing and and
1:31:02 you have to click 15 of the squares but
1:31:04 you only clicked 14 and so now you have
1:31:06 to do it again and then after the third
1:31:07 attempt to kick you out you have to
1:31:09 change your password again
1:31:11 this is every [ __ ] day of my life
1:31:17 congratulations you logged in now go to
1:31:19 your authenticator app which one I've
1:31:21 got seven
1:31:25 [ __ ] awful people
1:31:29 paint did you do this
1:31:31 is it you you have a PhD you've got the
1:31:35 brain to be able to do this
1:31:36 [Applause]
1:31:38 it might be your fault
1:31:40 we might have to gang up on pay it might
1:31:42 not be a good night for you pate
1:31:45 I've got password trauma password PTSD
1:31:49 it's not good I like turtles
1:31:53 am I the only one
1:31:58 I think I am
1:32:02 I do declare I feel a little flustered
1:32:07 is it warm in here is it me oh
1:32:13 all right I've been playing around with
1:32:15 training data sets this is from Rudy
1:32:17 with my llama 2 13B on my laptop but
1:32:20 speed is not my issue
1:32:22 what is your issue Rudy that's a very
1:32:25 intriguing little thing huh
1:32:29 uh
1:32:30 um you should use a TPU I hear they work
1:32:33 really well in a ridiculously cost
1:32:35 effective super cool again I love that
1:32:38 you guys are supporting one another
1:32:40 that's I love that uh did you explain
1:32:42 why you took chappie chat GPT off your
1:32:46 list I didn't it's the top one it's just
1:32:51 why they didn't all right I'm back
1:32:54 I just froze
1:32:56 okay like all right looks like the
1:32:59 hearts are stopping
1:33:00 can you hear me like tap a lot right now
1:33:03 tap the hearts a lot right now I know
1:33:05 you've been tapping a lot but I just
1:33:07 want to make sure that I can still be
1:33:08 heard or seen
1:33:10 because I froze solid on my side okay
1:33:14 I think we're good okay
1:33:16 did you explain why you took chat GPT
1:33:20 off your list I didn't chat.openai.com
1:33:22 is chatgpt I was going to say why they
1:33:25 didn't register chatgpt.com is beyond me
1:33:27 but I I again
1:33:30 they didn't think chat GPT was going to
1:33:33 be a thing they didn't think people were
1:33:35 going to use it
1:33:37 and then a hundred million people signed
1:33:39 up in three weeks or six weeks
1:33:43 but that's the that's the address
1:33:45 chat.openai.com it it it will be on my
1:33:48 list for a while I can't imagine chat
1:33:50 chat GPT dropping off the list
1:33:53 for at least
1:33:55 a year maybe more
1:33:58 because they're gonna they're gonna come
1:33:59 out with some [ __ ] now Google's gonna
1:34:01 come out with some [ __ ] like Google's
1:34:03 probably gonna slide up into this second
1:34:05 slot here
1:34:06 and I think I think Bing is going to
1:34:09 drop to this page
1:34:11 and then I think Poe's gonna probably
1:34:14 like pose already borderline because
1:34:16 it's what what Poe has that's really
1:34:18 cool is you can make your own bot and
1:34:20 you can play with different models in it
1:34:23 but it's a bit of a sort of preschool
1:34:25 kind of interface it's it's it's a good
1:34:27 app on your phone
1:34:30 um
1:34:31 but that one's gonna probably go because
1:34:33 I think
1:34:34 I think we're gonna see something else
1:34:35 here in the I think this is going to be
1:34:37 a busy fall
1:34:38 I think it's going to be a very busy
1:34:40 fall I like it Picasso
1:34:46 Waka Waka
1:34:50 um
1:34:51 oh it just jumped me to the bottom again
1:34:53 oh that Tick Tock that Tick Tock with
1:34:55 their little bug they got a little bug
1:34:57 you know I don't know if you know they
1:34:58 had a bug they got a bug I could tell
1:35:00 them to bug I could tell them right now
1:35:01 I could talk to the product people and
1:35:02 say I have documented a bug I can
1:35:04 recreate it regularly over and over
1:35:06 again I do it nightly it's a pain in my
1:35:08 ass could you fix it
1:35:10 no
1:35:13 but we've added 16 new AI filters that
1:35:15 can make you look like a lovely lady
1:35:18 perfect
1:35:21 can I get rid of my zit
1:35:27 hello
1:35:28 hello my name is Carl Shannon with the
1:35:31 AI learning lab
1:35:32 what do you mean I have no necks this is
1:35:34 my face this is the normal face why do
1:35:36 you ask
1:35:43 hey
1:35:48 okay
1:35:57 I really should have gotten that music
1:35:59 composition degree shouldn't I because
1:36:01 this is bad
1:36:02 it's like
1:36:03 my grandfather my grandfather when he
1:36:06 got nervous he used to whistle all the
1:36:08 time and he was completely out of aware
1:36:09 aware of it and it would just be like
1:36:11 what I just did
1:36:13 um but it would be just songs that no
1:36:15 one had heard of him
1:36:17 and his little cheeks would pop out
1:36:22 and then like mid conference like he'd
1:36:24 be in a conversation and in between
1:36:26 talking
1:36:28 completely just unaware that he's
1:36:31 whistling
1:36:33 oh my God we used to make fun of him it
1:36:35 was fun
1:36:36 it was fun
1:36:38 as as my uh as my son said dad you're
1:36:42 funny but you're mean
1:36:45 it's true it's true it's true I tell
1:36:48 y'all I don't I don't know what to tell
1:36:49 your son
1:36:52 my son's one of those guys that he reads
1:36:54 manuals it's like yeah
1:36:57 it's disappointing
1:36:59 sad it's sad it's
1:37:02 I'm like it's a hammer you don't need to
1:37:04 read the manual it does one thing it's
1:37:07 got two ends and a handle it's you don't
1:37:10 need to redo it yeah but but Dad
1:37:13 you can open a beer can with it okay
1:37:15 okay
1:37:20 oh
1:37:22 they they used to they used to get Lego
1:37:24 sets and actually build the thing that
1:37:28 was in the instructions like I have
1:37:30 never had that experience in my life I
1:37:32 have ADD I would get to like page four
1:37:35 and just be like [ __ ] it throw the
1:37:37 manual away
1:37:38 build a dragon
1:37:40 [Laughter]
1:37:44 no set of mine ever had complete pieces
1:37:47 in it from day one
1:37:49 wow man that was a fantastic hat
1:37:54 yeah rather we're on the Safari ah
1:37:56 welcome here
1:37:58 in England welcome to Galapagos Island
1:38:01 tonight we're going to be riding Turtles
1:38:03 for fun so uh
1:38:06 eat up
1:38:08 who wants another cigar
1:38:18 I handled the channel well I agree with
1:38:21 Emilio's wife that moderators should pin
1:38:23 comments I think that would be the
1:38:24 biggest thing and then you know the
1:38:26 moderators could also just people let
1:38:29 people know no this is how he is this is
1:38:31 just it this is yes it stays like this
1:38:36 no he doesn't appear to be on anything
1:38:38 we're pretty sure that's just how it is
1:38:45 oh good lord God lord good Lord good
1:38:49 Lord good Lord all right not a lot of
1:38:53 people with my name out there you can
1:38:55 probably find it
1:38:57 yeah it's a cool name Pete the owner is
1:39:00 pate short for something Peyton maybe
1:39:03 that's cool
1:39:05 um
1:39:06 click add comment and screenshot oh
1:39:09 fantastic shouting Beaver I'll tell you
1:39:11 you know the other night I was out
1:39:14 beaver hunting and uh we stumbled upon a
1:39:17 beaver
1:39:18 it it was the size of a of a Labrador
1:39:22 Retriever and heat and uh
1:39:26 it just stood there it's it it hiked up
1:39:29 on its back legs I guess it was sitting
1:39:30 on that big flat tail
1:39:32 and it looked at us like what the [ __ ]
1:39:34 are you doing here
1:39:36 and we looked at it like good Lord I've
1:39:38 never seen a beaver like that and I mean
1:39:40 I like a good beaver I mean who doesn't
1:39:43 like a good Beaver right
1:39:45 and I thought do I shoot this thing
1:39:47 Plumb between the eyes or do I high tail
1:39:50 out of here
1:39:51 so it stared at me and I stared at it
1:39:54 and the dog stared at me and I stared at
1:39:56 dog
1:39:57 and before you know it that Beaver let
1:40:00 out a screech like I've never heard
1:40:02 jumped on the other side of the day and
1:40:04 splashed Down Under the thing never saw
1:40:06 him again
1:40:07 craziest thing I ever did see
1:40:10 I like a good beaver
1:40:13 you know they're fuzzier than you think
1:40:16 you see him in the water they're all
1:40:18 slick to get out of the water they hike
1:40:20 up on their legs they're they're a fuzzy
1:40:23 Little Beaver
1:40:24 I like a good beaver
1:40:27 all right
1:40:28 carrying on
1:40:32 I like turtles
1:40:33 [Laughter]
1:40:37 oh what is the best for concept art
1:40:40 mid-journey
1:40:41 mid Journey no wait
1:40:43 wait
1:40:44 Milo Milo listen
1:40:48 listen
1:40:49 seriously Milo
1:40:52 Milo you and I need to have a
1:40:53 conversation
1:40:55 I like the name
1:40:57 Milo Dennison
1:40:59 very nice to meet you welcome to the
1:41:02 repository
1:41:04 um
1:41:05 we like to think we're a little special
1:41:06 around here anyway
1:41:09 you asked about concept art okay
1:41:11 fantastic listen mid-journey fantastic
1:41:14 imagery
1:41:16 type in some words out comes a picture
1:41:18 they're beautiful
1:41:19 photograph painting Styles up one side
1:41:22 down the other now
1:41:25 don't type in hand model
1:41:29 oh they're getting better but the hands
1:41:32 are a mess
1:41:32 arms are sometimes far too short
1:41:35 [Music]
1:41:36 um
1:41:37 doesn't quite seem to understand snails
1:41:39 for some reason anyway brilliant mid
1:41:41 Journey fantastic there's a new kid on
1:41:44 the Block we don't like the Nouveau
1:41:46 riche if you know what I mean anyway
1:41:51 ideogram ideogram wait
1:41:56 [Laughter]
1:42:01 there we go yes
1:42:04 Milo Dennison I met Milo Dennison at the
1:42:08 club 19 of 23 20 19 of 22 19 of 22.
1:42:15 fantastic lad fantastic lad a shame
1:42:18 about the bicycle accident anyway
1:42:20 ideogram
1:42:22 is
1:42:24 not nearly as good as mid-journey but it
1:42:28 can do text and none of these things
1:42:30 have been able to detect so
1:42:33 that image and that image and that image
1:42:37 and that image we're all done with
1:42:39 ideograms so if you're doing concept art
1:42:41 and you need wordy words in it and you
1:42:44 want them integrated into the image like
1:42:46 say that one with all the fancy [ __ ]
1:42:48 down here you're like well I could
1:42:49 Photoshop that because I'm pretty good
1:42:51 at Photoshop well yeah but if you just
1:42:53 wanted to type in
1:42:54 lobby with a big giant s AI learning lab
1:42:57 sign on it and it made that 30 seconds
1:42:59 later why not do that so
1:43:03 ideogram.ai
1:43:05 if you need text and mid-journey if you
1:43:08 want goodness
1:43:09 and then these other two are based on
1:43:11 stable diffusion if you're super geeky
1:43:13 and want to geek out go automatic 11 11
1:43:15 but basically the top two
1:43:18 all right
1:43:21 fantastic Milo thank you for playing
1:43:23 along
1:43:25 welcome welcome welcome welcome welcome
1:43:26 welcome welcome welcome
1:43:28 well well well well I'm not special yeah
1:43:33 well Milo Dennison well isn't he just
1:43:36 terrific
1:43:38 oh Milo Milo Milo he already get that
1:43:41 name where did he get that maybe in the
1:43:43 depths of Hell could it be Satan Satan
1:43:45 give you that name fantastic Milo all
1:43:48 right carry on go make your concept art
1:43:50 let us know how you do share it
1:43:53 my Raspberry Pi actuator is being used
1:43:57 by Tinder I really want to know what's
1:44:00 going on there that's fantastic do not
1:44:02 scroll out of life
1:44:03 Kyle where the prompts at which prompts
1:44:06 crypto Joe
1:44:09 which prompts you mean prompts.chat
1:44:12 what are you talking about
1:44:14 I don't know what you're talking about
1:44:16 crypto Joe
1:44:17 I do not know
1:44:21 um Dr J what's yours wait you're not
1:44:24 Rosie
1:44:25 oh that was a lovely troll I like it
1:44:29 what's up talk wait you're not Rosie
1:44:32 I think my favorite of those was I was
1:44:35 scrolling through Tick Tock and thought
1:44:36 I saw Alec Baldwin but fat
1:44:40 and stopped entered my life and took the
1:44:44 time to share that
1:44:48 that was a solid troll because you're
1:44:50 like oh Alec Baldwin butt fat
1:44:53 huh
1:44:55 it feels good warm and fuzzy like oh I
1:44:58 gotta go oh right now yep nope yep that
1:45:01 was the correct answer
1:45:02 all right we have to train AI to be nice
1:45:04 uh please is a good thing well like at
1:45:08 some point the AI is going to start
1:45:09 keeping track of us right when it has
1:45:12 memory and it's like if you're a
1:45:13 belligerent [ __ ] to AI it's just
1:45:15 gonna get worse and worse so it's gonna
1:45:17 be in our best interest to be nice to
1:45:19 the robots
1:45:24 uh
1:45:27 or build a recipe with what's in the
1:45:29 fridge why would I do that she asked we
1:45:31 got another oh
1:45:33 that is about your mama they're nice I
1:45:36 liked it you're talking about your mama
1:45:39 um oh Tick Tock just
1:45:44 I don't know sometimes it jumps me to
1:45:46 the bottom and then other times it jumps
1:45:47 me to just some weird random place
1:45:50 way up above I don't know what's going
1:45:53 on
1:45:54 I don't know who'd program this the
1:45:57 Chinese oh right yeah they did DHS just
1:46:00 Chinese the Chinese did it yeah well
1:46:03 there you have it they have it you see
1:46:06 what I'm saying
1:46:07 do you see what I'm saying
1:46:14 all right let me see if I can find where
1:46:15 I was in my questions
1:46:17 a bear the bear to pierce
1:46:20 the appears he was here earlier I was
1:46:24 here earlier
1:46:25 did you see my screenshot work around
1:46:30 hi there uh write a funny poem about
1:46:34 handling comments in life exactly it'll
1:46:39 it'll be good at that
1:46:41 and also just for posterity
1:46:44 I worked in AWS Alexa Ai and now Google
1:46:47 TPU oh wow AWS made me hate permissions
1:46:51 AWS permissions dude holy [ __ ]
1:46:56 what did I just do on AWS the other day
1:47:01 well I mean there's some staggering
1:47:03 things about AWS like like you can't
1:47:06 search it like and they're like well no
1:47:08 it's just a bucket you have to write
1:47:09 your own application layer but like I'm
1:47:11 trying to retrieve a [ __ ] file well
1:47:13 if you didn't write the application
1:47:14 layer that could retrieve files that's
1:47:16 on you I'm like [ __ ]
1:47:18 oh
1:47:20 it's got you know eight trillion
1:47:23 functions and not one of them is
1:47:25 retrieve it's like [ __ ]
1:47:31 yeah AWS is special but you know
1:47:34 powerful and all
1:47:37 anyone by the way the fact that Pate
1:47:40 worked at AWS Alexa Ai and now Google
1:47:43 TPU
1:47:44 he earned that PhD
1:47:48 um anyone that's using Claire at Denver
1:47:50 International
1:47:52 already doing that I am wait anyone is
1:47:56 using Clarity at Denver International
1:47:58 already
1:48:00 doing that I am I don't know what that
1:48:01 means One Love
1:48:04 um it's so much fun to click on stop
1:48:05 lights to prove we're human
1:48:08 that's good
1:48:11 um that kind of proof of humanness may
1:48:13 be good but it freaks me out yeah I know
1:48:15 Andrea it so here's here's
1:48:19 so so what Andrea's talking about or
1:48:22 Draya is
1:48:24 um I got my Iris scanned at Sam Altman's
1:48:28 one of his orbs the the world coin World
1:48:30 ID
1:48:32 um I did it for a couple of reasons I
1:48:33 actually think
1:48:36 a system like that it might not be that
1:48:38 one but a system like that is going to
1:48:40 be a very important thing
1:48:43 um so I wanted to be in on it early I
1:48:45 also think like I don't know that I
1:48:48 trust Sam Altman but I think at a
1:48:50 minimum Sam Altman is very thoughtful
1:48:52 and aware of the dangers of what he's
1:48:56 creating it ain't slowing him down
1:48:59 he's still creating the [ __ ] but in one
1:49:02 one of the interviews with Sam Altman he
1:49:04 said um the [ __ ] did he say he goes
1:49:08 oh there's some things we will never
1:49:11 release I'm sure pate's got some of
1:49:12 those at Google I'm sure there's some
1:49:14 [ __ ]
1:49:15 they will never release right
1:49:17 so so I feel like Sam is at least
1:49:19 conscious of the risks and pitfalls of
1:49:23 the [ __ ] he's creating
1:49:26 um
1:49:30 if
1:49:31 here here's how I see the risk
1:49:34 okay maybe
1:49:36 open AI lies and they're actually
1:49:38 keeping my biometric data
1:49:41 well
1:49:43 you know people send in their DNA to
1:49:45 ancestry which is owned by the Mormon
1:49:48 church right so so they've got your DNA
1:49:51 Google's got all of our data and our
1:49:53 Behavior Facebook's got all of our data
1:49:55 and behavior Apple's got all of our data
1:49:57 Behavior at least Apple's protecting it
1:49:59 like Apple's not selling it to the
1:50:01 highest bidder
1:50:02 um they're monetizing the [ __ ] out of it
1:50:04 on their own
1:50:05 um so like I feel like all my data's
1:50:07 already out there this is an anonymous
1:50:09 thing
1:50:11 um and
1:50:13 right now it's in the development stage
1:50:15 if you go to
1:50:17 um
1:50:17 worldcoin.org
1:50:19 uh it's just it's really just
1:50:21 documentation for developers to go build
1:50:23 on top of this infrastructure they're
1:50:25 looking to build and then Sam Altman's
1:50:27 talking about the the way
1:50:29 open AI is structured it's got a capped
1:50:31 profit structure so that any investor
1:50:34 puts in a million dollars can only get a
1:50:37 hundred X
1:50:38 back right so if I put in a million I
1:50:41 can earn up to 100 million and after
1:50:43 that it goes back into open ai's
1:50:46 non-profit
1:50:48 which Sam Altman's going to use for
1:50:49 Universal basic income what because what
1:50:52 Sam is is assuming is going to happen is
1:50:54 that AI is going to drive productivity
1:50:57 up and the need for human participation
1:50:59 down and so there's going to be this Gap
1:51:01 that's created where there's all this
1:51:03 Prosperity up here which if you just
1:51:05 left it to the PE firms the private
1:51:07 Equity firms they'd be like [ __ ] it more
1:51:10 money for the rest of us great fire all
1:51:12 the people take all the money what Sam's
1:51:15 trying to do is say I'm going to take
1:51:16 you know above a certain reasonable
1:51:19 profit 100x and I'm going to dump that
1:51:22 back to the people and so I think it's
1:51:24 an interesting experiment and so I just
1:51:26 want to be a part of it I want to see
1:51:27 does anyone actually develop on this
1:51:29 platform and and you know does he
1:51:32 actually keep his word and if he does
1:51:33 you know I want to be there and see it
1:51:36 and experience it and it was it was
1:51:37 painless right it was it was you log
1:51:40 into an app and you get it set up you
1:51:42 get your account set up and once that's
1:51:44 set up you
1:51:45 hold your
1:51:48 QR code up to the thing it syncs it to
1:51:51 that account you stick your iris on it
1:51:54 it makes a little circle like this and
1:51:57 then it does some calculation [ __ ] and
1:51:59 and the app said
1:52:01 um
1:52:02 determining your uniqueness which I
1:52:04 thought was kind of cool and then it
1:52:06 went pring
1:52:08 here's your ID
1:52:11 so yeah I know it I like I understand
1:52:13 the freak out of it it's weird it's
1:52:15 weird it's weird as [ __ ] 10 30 here how
1:52:17 much have I missed tonight I don't think
1:52:19 you miss much any night here's what
1:52:22 you've missed I've acted like an idiot
1:52:24 and occasionally done this
1:52:31 I hope that GPT is in another Netscape
1:52:33 groundbreaking beginning but eventually
1:52:35 disappeared it might be Marie I mean
1:52:38 you know there's there's seven I like
1:52:41 I've got seven on this list and that's
1:52:43 just I'm just some dude those are the
1:52:44 ones I think are worth putting your time
1:52:46 into right now
1:52:47 um
1:52:49 it always shakes out to three right
1:52:53 um Google effectively you know shook it
1:52:57 down to a single Monopoly or you could
1:52:59 say it was like Google Facebook and
1:53:01 Twitter were the three of recent years
1:53:04 um but for search you know sort of
1:53:05 Google locked it in as an effective
1:53:08 Monopoly for I don't know what a decade
1:53:10 and a half two decades
1:53:12 and then all of a sudden this shit's
1:53:14 shaking it up so so I think there's
1:53:16 going to be all sorts of permutations
1:53:18 there's going to be all sorts of
1:53:19 bubbling whether or not open AI survives
1:53:22 this I think they've got a decent chance
1:53:23 at it because they've sidled up with
1:53:26 Microsoft and Microsoft
1:53:28 it's Microsoft has so tied their future
1:53:31 product roadmap into open AI that I
1:53:35 think even if open AI fails there's
1:53:37 probably some poison pill clause in the
1:53:39 contract that Microsoft subsumes it
1:53:41 rather than it just dying Microsoft
1:53:43 probably gets the IP so
1:53:45 um
1:53:48 so I think though I I think it'll be
1:53:50 there for a while in in in some sense it
1:53:53 will be nice to see you know maybe open
1:53:55 ai's you know the new the new Big player
1:53:58 on the Block
1:54:00 um
1:54:01 but but we'll see we will say verified
1:54:04 Human by a machine exactly it's so true
1:54:09 preach I don't know what I was saying
1:54:10 but I was I was on a little bit of a I
1:54:12 was on a little bit of a uh sermon on a
1:54:16 mountain I was on the mount
1:54:19 doing it I was doing it wsn radio Hard
1:54:22 Rock and Hip-Hop welcome thank you for
1:54:25 the rose
1:54:26 arose by another name would smell is
1:54:29 sweet wsn radio
1:54:32 smooth jazz on the 88s
1:54:35 we'll have traffic on the nines
1:54:39 all right
1:54:41 always a daylight dollar short passwords
1:54:45 only keep me out of my accounts but not
1:54:47 this cameras exactly Amelia's wife
1:54:50 and please don't think I'm I'm an
1:54:52 authority on anything still relatively
1:54:54 new to machine learning compared to
1:54:55 others yeah I know I don't pay isn't it
1:54:58 wild it's like I've I've got I know a
1:55:00 lot of people that have been in machine
1:55:01 learning for you know some time and
1:55:04 they're like but this generative AI [ __ ]
1:55:06 it's just different it's just a [ __ ]
1:55:07 different thing
1:55:09 it's just it's I don't know [ __ ]
1:55:11 crazy man is he's crazy what time is it
1:55:14 when do I gotta go
1:55:16 what what
1:55:19 I got tomorrow I think I got an eight
1:55:21 o'clock meeting I can't be staying up
1:55:23 too late you people keep me up too late
1:55:24 you know you keep me up too late it's
1:55:27 not fair it's not fair 59 000 likes
1:55:30 that's very impressive
1:55:32 but 59 000 you're going to get to 60.
1:55:35 what are you doing why are you not
1:55:37 tapping
1:55:38 is my internet going poo poo
1:55:40 oh there's a there's a little hearty
1:55:42 heart oh they got tired you know what it
1:55:44 is they're older they're older the
1:55:46 tappers are older so they're good for
1:55:49 about an hour and then their their wrist
1:55:51 gets sore because they've been mousing
1:55:53 for decades
1:55:55 they're good at the mountain but they're
1:55:57 they've been doing it for decades and so
1:55:59 they got the carpal tunnel and so we're
1:56:00 tapping on the screen they're happy for
1:56:03 a while and then I think two things
1:56:04 happen one is I think the wrist you know
1:56:06 freezes up it gets a little numb and
1:56:08 then other ones I think they just I
1:56:10 think they just take a nap I think they
1:56:12 get a little sleepy in it
1:56:18 so but they're they're regulars what do
1:56:21 you want from them they're you know
1:56:22 they're just not they're not normal
1:56:23 they're irregular okay we love them we
1:56:27 love them they're they're us they're us
1:56:30 they're all of us okay
1:56:31 no hate no hate on the Irregulars
1:56:34 all right you're up to 60 000. my
1:56:37 shaming worked shaming strategy always
1:56:39 comes through
1:56:41 shame is my love language
1:56:43 all right now we have two passwords
1:56:44 written down posted notes exactly
1:56:47 exactly I got all my first Factor
1:56:50 passwords to the left and all my second
1:56:52 Factor passwords to the right oh my God
1:56:55 it's so true
1:56:57 I have the same password issues there's
1:57:00 no way I'll remember them that ship has
1:57:02 sailed exactly just [ __ ] be done with
1:57:04 it just make all of my accounts open
1:57:06 source just steal all my [ __ ] money I
1:57:08 don't care at this point so long as I
1:57:09 don't have to log in again
1:57:11 like everything right now is Google
1:57:13 single sign-on because I'm just like
1:57:15 [ __ ] it I don't care [ __ ] it let Google
1:57:18 have all of my accounts they can do
1:57:19 everything they can know everything I do
1:57:21 I don't care anymore
1:57:23 I had to shoot the most of my passwords
1:57:25 made me feel old we've got a few of
1:57:28 those around the household here
1:57:30 um I have nothing to do with passwords
1:57:33 oh pain you have to see paints even get
1:57:35 defensive he knows I'm right he knows
1:57:38 what a [ __ ] show it's become although I
1:57:40 did give Google a compliment with Kyle
1:57:42 my Kyle Turtle
1:57:45 oh that's awesome
1:57:47 like I guess I guess the first Turtle
1:57:50 reference of the evening I mean I said
1:57:52 it before but the first one coming from
1:57:54 the Irregulars
1:57:56 deserves
1:58:01 [Music]
1:58:07 back here live at the waterfront village
1:58:10 with my friend the zombie John Jonathan
1:58:12 you're looking good Jonathan just got an
1:58:14 awesome face paint job what do you think
1:58:16 I like turtles all right they're great
1:58:19 zombie good times here at the waterfront
1:58:25 all right you're a great zombie good
1:58:29 times here at the water park
1:58:33 I don't know how she held it together
1:58:35 she deserves an Emmy for that
1:58:40 oh every [ __ ] time it gets me
1:58:46 Clint hey Kyle what's up what's
1:58:48 happening Clint good to see you
1:58:51 email is risk neutral and encrypted I
1:58:53 emailed myself I emailed myself my
1:58:55 passwords yeah exactly I just like I'm
1:58:58 like yeah okay I I don't think I have
1:59:01 anything interesting enough apparently I
1:59:02 do have something interesting enough
1:59:04 because some [ __ ] scammer duped my
1:59:06 whole account here apparently this is
1:59:08 some sort of friggin Cash Cow for the
1:59:12 scammer
1:59:13 the
1:59:15 good Lord
1:59:17 sharks are older than trees 50 million
1:59:19 versus 400 million cool
1:59:21 not sure what that's in reference to but
1:59:23 I like it Uncle Kyle how are you tonight
1:59:26 trade Checker
1:59:27 um I'm good there little you little you
1:59:29 little whippersnapper I'm doing good I'm
1:59:32 doing good you know we're hanging out
1:59:34 we're talking about artificial
1:59:36 intelligence we talk about it nightly we
1:59:40 talk we you know what we were talking
1:59:41 about tonight Troy passwords I don't
1:59:44 know if you were here we were going off
1:59:47 on passwords see we we have we have a
1:59:50 little interloper from the Google
1:59:53 and he's here and he was a little cagey
1:59:56 and said something Big's coming it's
1:59:57 going to be exciting you're gonna love
1:59:59 it
2:00:00 and then that got me to thinking about
2:00:01 the last time I tried to sign into my
2:00:03 computer and I broke my computer because
2:00:06 I wanted to hurt things and I took it
2:00:08 out on the machine and and so we talked
2:00:11 about that for a while
2:00:13 and uh
2:00:15 oh you know what else we talked about
2:00:17 turtles you saw that you were just here
2:00:21 yeah yeah well welcome Troy good to have
2:00:25 you here it's really nice I'm at level
2:00:28 15 now Kyle
2:00:30 far out Clint by the way it is a
2:00:33 competition the Irregulars are trying to
2:00:36 out level up each other
2:00:38 um so the AI learning lab is much like
2:00:40 Scientology that to reach Enlightenment
2:00:43 you really do need to reach a higher
2:00:45 level
2:00:46 look at dragonfly coming in with the
2:00:48 gummy bear coming in hot with the gummy
2:00:50 bear oh they're not gonna [ __ ] me up man
2:00:51 I'm in here I'm gonna level up
2:00:56 come on man doesn't Tom Cruise work here
2:00:58 he used to he used to
2:01:00 see Tom was our spokes model
2:01:03 before he went to that other place right
2:01:06 and you know that's why we are who we
2:01:09 are you know it's all Tom it's all Tom
2:01:11 and Rosie all Tom and Rosie cousin Rosie
2:01:14 oh cousin Rosie she is a gas
2:01:18 all right
2:01:19 pie calls itself chatty bot and started
2:01:22 calling me chatty human that's hilarious
2:01:25 thank you Ted that was awesome and thank
2:01:27 you for the uh making it rain in here
2:01:29 that was nice I will I will get to the
2:01:32 pole dancing and the twerking you know
2:01:33 it's gonna happen because I'm lost
2:01:36 connection thanks CenturyLink or no it's
2:01:39 Xfinity Comcast uh thank you Matthias I
2:01:44 have done that I if you know another way
2:01:46 to report someone by the way uh
2:01:50 the the account's been duped this
2:01:53 account's been duped my my top pinned
2:01:56 post
2:01:57 um talks about it
2:02:00 so if you want to report it you if you
2:02:02 want to report the account on my behalf
2:02:04 you have to declare me a celebrity which
2:02:07 I will accept I will begrudgingly accept
2:02:10 that I am a celebrity so feel free to
2:02:13 report and just declare me a celebrity
2:02:15 and then I think everyone will be happy
2:02:18 [Laughter]
2:02:21 I got a degree in acting I gotta justify
2:02:23 it somehow
2:02:27 oh yeah it's fantastic it's fantastic
2:02:31 Bob yeah Bob tell him what he's one blob
2:02:33 yeah you want a new car
2:02:37 and we're back
2:02:40 um I'm at level 15. oh we were here okay
2:02:42 I found myself for sure I like totals
2:02:45 and I like zombies that like turtles
2:02:48 but but I also highly support beavers
2:02:51 beavers beavers are fantastic beavers
2:02:54 are fantastic Well Chad GPT be the
2:02:57 modern day Netscape game changing
2:02:58 beginning to disappear possible we
2:03:00 talked about that
2:03:01 on poke and you make Bots that is
2:03:04 similar to an agent no the Bots well I
2:03:09 suppose if you prompted it cleverly
2:03:11 enough you could
2:03:13 where where you you know how I don't
2:03:15 know if you've ever done this where
2:03:18 you can tell Chachi BT to give you
2:03:20 step-by-step instructions and it'll just
2:03:21 give you all the instructions but then
2:03:23 you can say no no
2:03:25 just give me step one and then don't
2:03:27 move on to step two until I say okay or
2:03:30 whatever the
2:03:31 whatever your secret code word is with
2:03:34 chat gbt you might be able to do
2:03:36 something with Poe like that where it's
2:03:38 sort of an interrogating bot
2:03:41 um but it but it doesn't have any sort
2:03:42 of autonomous anything I would say use
2:03:46 um zapier for that if you want to do
2:03:48 something where you're doing a more
2:03:49 sophisticated chain of events and you
2:03:51 don't want to do programming zapier with
2:03:54 its interfaces uh beta where you can
2:03:56 create your own chat bot and that you
2:03:58 could probably have a string of of
2:04:01 you know tools and logic and stuff like
2:04:04 that so
2:04:05 there are two cows in here are you
2:04:08 related I do not know I do not think so
2:04:11 glamor filter for the zip oh yeah I
2:04:13 should probably put on the glamor filter
2:04:15 oh well [ __ ] it
2:04:17 at this point it's like what I have 16
2:04:20 chins and a zit really
2:04:23 really is it that's where you're gonna
2:04:25 go I finally deal with the nose hairs
2:04:28 and and now this
2:04:29 what what next a hair out of the zit
2:04:32 probably it's probably why it's there
2:04:34 all right by the way I'll fix your Tick
2:04:36 Tock scrolling issues yeah exactly Vicky
2:04:41 someone's got to read the manuals I'm
2:04:43 proud of him oh Vicky you're so
2:04:45 disappointing
2:04:48 I tried it I tried to teach my kids the
2:04:52 best of everything
2:04:54 don't read manuals just live your life
2:04:58 gotta read just trick step one step one
2:05:02 a step one be no
2:05:04 it's here this is square and it goes in
2:05:06 the Square thing
2:05:09 look that might not be it and then 15
2:05:11 minutes later
2:05:12 square a into square whole B okay yep it
2:05:17 was this one I
2:05:24 cannot exist in this environment
2:05:28 reading manuals
2:05:35 I'm a manual reader from way back I'll
2:05:38 Silver Fox slow your roll Cowboy you all
2:05:40 you all with your oh this is
2:05:45 it was it it must be nice to have
2:05:47 executive function
2:05:50 block of Swiss cheese from like here
2:05:52 forward
2:05:54 in me not not you guys you have
2:05:57 functional frontal lobes me um no
2:06:01 I bought a car recently and it came with
2:06:03 three manuals 400 plus pages I can't
2:06:05 handle it well now now you could just
2:06:06 pop it into Claude or if Google comes
2:06:09 out with some large context model just
2:06:11 pop it in there
2:06:12 pop it in there and say I want to know
2:06:15 actually now that I think about that
2:06:18 when we start to have large language
2:06:20 models local on our phones or even
2:06:23 cloud-based I suppose and we can save
2:06:26 things
2:06:27 you could take a PDF of a car manual pop
2:06:30 it into your little
2:06:31 thing there turn it into a standalone
2:06:33 little application and have like your
2:06:36 card GPT
2:06:38 and you could not only take the official
2:06:39 manual but you could take like all the
2:06:41 you know the hack things like if you're
2:06:43 gonna if you're gonna chip mud your car
2:06:45 you can take the manual for that and so
2:06:48 you could have the chipmod stuff in
2:06:50 there plus the real stuff
2:06:51 how do I reset the door passcode
2:06:54 here you go that'd be pretty cool
2:06:58 to report okay we did that one frumple I
2:07:01 didn't get the illusion sorry
2:07:04 um doesn't appear to be on anything or
2:07:06 forgot to take his medication not sure
2:07:08 which
2:07:11 oh Amelia's life you're a card Monty
2:07:15 with an e is better snapping B
2:07:18 snapping Beaver you gotta be careful of
2:07:22 them snapping bathers who doesn't like a
2:07:24 good Baker he's getting spicy in here
2:07:27 I don't know what you're talking about I
2:07:30 was talking about a nature experience I
2:07:32 had out in the field with my lovely
2:07:34 doggy
2:07:37 you Saucy minx's every one of you
2:07:41 Wynonna Big Brown Beaver
2:07:46 she did she got it mounted too
2:07:51 and not the nerve spicy kind uh I
2:07:54 brought the wrong charging cords bets
2:07:56 counting down oh that sucks I may
2:07:58 disappear if I do it's been fun see you
2:08:00 Thelonious six percent no say anything
2:08:02 so
2:08:03 so happy I captured the beaver character
2:08:05 oh that's fantastic share bear that's
2:08:08 awesome he was pretty good I stuck with
2:08:11 it I was I was gonna I was gonna bail on
2:08:13 the beaver bit but I thought if if
2:08:15 you're gonna do a beaver bit you can't
2:08:17 half commit to it
2:08:19 because if you have commit to it it
2:08:21 makes everyone feel awkward you just
2:08:22 gotta go in [ __ ] full
2:08:26 so I'm glad you got that thank you
2:08:28 that's that's awesome I think that's
2:08:30 yeah it's probably going to be held
2:08:31 against me in a court of law but I do
2:08:33 appreciate it
2:08:36 uh
2:08:40 [Music]
2:08:43 claw peace how you doing
2:08:48 one of the irregulous one of the many
2:08:49 Irregulars here tonight by the way we're
2:08:51 at 67 000 of those tappy Taps how many
2:08:54 logging into AI salon tomorrow
2:08:57 I am
2:09:03 Milo is awesome 50k Hills yeah oh yeah
2:09:07 that was a while ago I'm behind some 25
2:09:10 messages down so I'm not too bad right
2:09:12 now
2:09:13 can someone assist me in verifying my
2:09:15 Claude AI account I need a us or UK
2:09:18 number to verify
2:09:22 I would use a VPN don't I would not
2:09:25 share a phone number with someone you
2:09:26 don't know
2:09:27 uh they they don't that don't that
2:09:30 doesn't sound good to me that do not
2:09:32 sound good to me that sounds like what
2:09:34 the scammers would be telling people I
2:09:36 don't like it I do not like it everybody
2:09:39 has to be on camera on Salon AI you
2:09:43 don't have to be but most of us are
2:09:45 because it's about the community like
2:09:47 the the AI Salon honestly is about the
2:09:50 people it's not about the technology
2:09:52 so like we have values like generosity
2:09:55 and empathy and curiosity and
2:09:59 exploration
2:10:02 because we're all [ __ ] trying to
2:10:04 figure it out
2:10:06 and no one's an expert right now no one
2:10:08 no one zero zero zero experts anyone
2:10:14 that tells you know what they know what
2:10:16 the [ __ ] is going on with this
2:10:18 generative AI stuff is lying to you or
2:10:23 to make money
2:10:25 they're one of these douchebags
2:10:29 you know how they're making money right
2:10:30 it's not on chat CPT
2:10:33 it's off of you
2:10:36 thinking someone knows more than you do
2:10:39 about
2:10:41 they don't
2:10:43 no one does all right we're almost a
2:10:45 year in we're getting close to some
2:10:47 people maybe know a little bit but but
2:10:50 in general not because the things are
2:10:51 changing too fast in fact Pate that's
2:10:53 here tonight you know said that there's
2:10:55 some [ __ ] coming
2:10:58 from yield Google
2:11:01 um these things are all changing so fast
2:11:04 that even if you know what you're doing
2:11:05 today you're not going to know what
2:11:06 you're doing even a month from now two
2:11:09 months from now just shit's moving too
2:11:11 fast so you can't rest on your laurels
2:11:14 and you also can't get too married to
2:11:15 the whatever brilliant genius thing you
2:11:18 created
2:11:20 um because it's just gonna be uh it's
2:11:22 just gonna be Irrelevant in a in a month
2:11:24 or two and that's okay like the the
2:11:27 value right now is in being in the game
2:11:29 learning being curious understanding
2:11:32 that these things are not what we've had
2:11:35 before that they're profoundly powerful
2:11:38 that we are lucky to be able to play
2:11:40 with them like [ __ ] lucky
2:11:44 like think about it
2:11:46 like when I was a kid
2:11:49 Star Trek what I knew in my soul
2:11:53 was that the [ __ ] in Star Trek wasn't
2:11:57 possible
2:12:00 it's almost all possible with the
2:12:04 exception of the teleporter right almost
2:12:07 everything else
2:12:09 language translators the Sim room that
2:12:12 you know that just [ __ ] makes the
2:12:14 places
2:12:16 Sim people simbots
2:12:19 crazy communication crazy tricorder
2:12:22 things that read your [ __ ] heart rate
2:12:24 and your
2:12:26 we're living in a post-sci-fi world
2:12:29 and and this is like the final
2:12:33 the final thing because all the [ __ ]
2:12:35 that hasn't been invented yet these
2:12:37 things are going to start accelerating
2:12:40 Innovation like they did with Alpha fold
2:12:42 where they [ __ ] just invent
2:12:45 four hundred thousand protein
2:12:47 predictions over a couple of days
2:12:51 and they're accurate to like some stupid
2:12:53 [ __ ] degree
2:12:56 so they're gonna start doing that for
2:12:58 all of the Undiscovered [ __ ] one after
2:13:01 the other after the other after the
2:13:02 other that's why I think we're in for a
2:13:04 great Renaissance where it's going to be
2:13:06 science it's going to be creativity it's
2:13:08 going to be literature it's going to be
2:13:09 [ __ ] all of it
2:13:11 it's just going to explode like we've
2:13:13 never seen before
2:13:15 I mean holy [ __ ] are we lucky we get to
2:13:17 be at the beginning of that
2:13:19 we are going to be people
2:13:23 who get to say
2:13:26 I remember what the world was like
2:13:28 before this [ __ ] showed up
2:13:33 just like you know
2:13:36 gen xers can say I remember what the
2:13:38 world was like before the internet
2:13:44 we get another opportunity
2:13:49 to live that Delta
2:13:52 the time before and the time after
2:13:54 November 30th 2022
2:13:59 that's where the flag went in the ground
2:14:03 everyone adopted it Google lost their
2:14:05 [ __ ]
2:14:06 Microsoft gloated
2:14:09 meta said we're never going to say the
2:14:11 word metaverse ever again it's all about
2:14:13 AI now everything [ __ ] changed on
2:14:16 that day
2:14:19 and because of that because it took off
2:14:22 so much this [ __ ] ain't going away
2:14:24 Meadow's like [ __ ] it we'll just we'll
2:14:26 open source our [ __ ] because we need to
2:14:27 stay relevant
2:14:29 Google's like well we invented this [ __ ]
2:14:31 oh I guess we're behind now they're
2:14:32 catching up
2:14:35 Apple's like yeah we spent 24 billion of
2:14:38 our research dollars on generative AI
2:14:40 we're gonna be good we'll we'll be good
2:14:42 you'll see
2:14:46 would any of those things happen no
2:14:49 but Sam Altman and his merry band of GP
2:14:53 tears
2:14:56 you know
2:14:58 release something that captured the
2:15:00 world's imagination and we're never
2:15:02 going back
2:15:04 not going back
2:15:06 never never never never never never
2:15:08 never never never never
2:15:10 discover in your eyes with clear at the
2:15:12 airport
2:15:14 all around so skip the line
2:15:16 oh disc in your eyes yeah yeah I haven't
2:15:19 done the clear thing
2:15:21 um
2:15:22 I don't know why
2:15:24 what's up how you been hey Chris Walker
2:15:25 been good
2:15:27 love my Google number had it forever
2:15:29 very cool I'm in South Africa I was
2:15:32 originally able to create an account
2:15:33 through a VPN and a mock local location
2:15:36 app there you go
2:15:38 they don't have my DNA
2:15:42 once Bill Bill Burr has that that thing
2:15:44 he said why would you send you let me
2:15:47 say why would you send your DNA to the
2:15:49 internet
2:15:53 about
2:15:54 Bill Burr is [ __ ] hilarious
2:15:57 [Laughter]
2:16:02 hilarious it's almost 10 p.m Kyle I know
2:16:05 it's 10 14 now all right I gotta I gotta
2:16:08 I gotta roll this any talks on Gemini
2:16:10 well Peyton here from Google he said he
2:16:13 can't tell us anything but I heard a
2:16:15 rumor that Gemini was going to launch in
2:16:17 the fall so possibly and also I kind of
2:16:21 lost the urge to keep playing how do you
2:16:24 keep it going
2:16:26 um
2:16:28 I go through waves I I mean listen I
2:16:31 started out when I started out I started
2:16:33 like five Ventures within a three-week
2:16:35 period this was one of them
2:16:38 I started an art project
2:16:40 this the AI Salon a newsletter
2:16:44 and all the [ __ ] I'm doing with my
2:16:46 company I did all of those in like a
2:16:48 three week period back in December
2:16:52 and like I left the the newsletter sort
2:16:55 of drift off to the side of my part and
2:16:56 I got a big powerful job and and he
2:16:58 didn't have time for it so that just
2:17:00 kind of let's just it's kind of
2:17:02 mothballed right now
2:17:04 um
2:17:06 I don't know like so for me
2:17:11 the reason I keep sticking with this
2:17:13 stuff is just having been through this
2:17:14 with the World Wide Web
2:17:17 um
2:17:19 what I know
2:17:22 is that you don't get to do this a
2:17:24 second time
2:17:26 like we're at a singular point in
2:17:28 history
2:17:30 like
2:17:32 like there there is the time before chat
2:17:36 GPT came out in the time after
2:17:38 and the the first three years of this
2:17:41 the formative years of this are really
2:17:44 important
2:17:46 and they're really exciting
2:17:48 and they're just a [ __ ] mess because
2:17:51 like all of this software is just
2:17:55 like just steaming piles of [ __ ] and not
2:17:58 for the fault of the developers like
2:17:59 they're fine they're just not
2:18:01 sophisticated software yet right
2:18:03 they're like they're like oh [ __ ]
2:18:06 everyone in the world wants our [ __ ]
2:18:09 what do we do I don't know code it quick
2:18:11 launch
2:18:14 foreign
2:18:18 and so
2:18:20 so they're like beautifully imperfect
2:18:22 right now because they're just so raw
2:18:24 and and like we get to like
2:18:27 type in them and break them and
2:18:29 jailbreak them and figure out the Dan
2:18:32 jailbreak and and I don't know it's just
2:18:35 like a beautiful time so while while I
2:18:38 get exhausted too
2:18:40 like like
2:18:42 the distance between being able to
2:18:44 prototype something oh thank you for
2:18:46 that share bear the distance between
2:18:49 being able to prototype something and
2:18:51 actually deliver it to a client is
2:18:53 [ __ ] massive like you can get to a
2:18:56 prototype really quick and you're like
2:18:57 this AI shit's the best and then you
2:19:00 actually try to build something for a
2:19:02 client you're like this [ __ ] doesn't
2:19:03 work
2:19:05 [ __ ] what do I tell the client
2:19:09 you know and you keep hammering and keep
2:19:11 hammering you keep hammering and three
2:19:13 years from now
2:19:15 you'll just have the button
2:19:17 make the client happy
2:19:21 clients happy
2:19:24 so there's something [ __ ] vital and
2:19:27 exciting about
2:19:29 engaging with these tools in this raw
2:19:31 State and and for me like trying to
2:19:34 Envision okay if it looks like this
2:19:37 today
2:19:39 and if I know like what it was two years
2:19:42 ago and then one year ago and now what
2:19:44 it looks like today I can see how far
2:19:45 it's come in two years
2:19:47 like just look at image generation if
2:19:49 you go back like a year and a half
2:19:51 mid-journey and stable diffusion the
2:19:53 images that they made were laughably bad
2:19:56 like hysterically just not even
2:19:58 hysterical they weren't even so bad
2:20:00 they're funny they were just bad
2:20:03 and you're like it kind of looks like a
2:20:05 bear
2:20:07 or a dog
2:20:09 it has fur
2:20:12 to a year and a half later
2:20:14 holy [ __ ] [ __ ] so these are gonna
2:20:17 take the same trajectory right
2:20:20 so a year and a half from now this
2:20:22 shit's gonna be unrecognizable and it's
2:20:25 gonna be [ __ ] brilliant
2:20:27 and so there's something about being in
2:20:31 it when that evolution is happening that
2:20:34 I know
2:20:35 we have one we have you have this you
2:20:38 have this Moment Like This Never exists
2:20:40 again
2:20:41 and if you're on the sidelines you'll be
2:20:43 like oh yeah I remember when that came
2:20:44 out yeah it was cool it lied a lot like
2:20:46 you know whatever you know you know
2:20:49 whatever
2:20:50 it wasn't great like I didn't use it
2:20:51 because it wasn't good I don't want to
2:20:53 be that guy I want to be like holy [ __ ]
2:20:55 and then it made me an application out
2:20:57 of python and I didn't know how to use
2:20:59 Python and it did a thing and then it
2:21:01 did this thing and then it did that
2:21:03 thing and then it started this thing and
2:21:04 then and then and that turned into this
2:21:06 thing and then
2:21:07 I don't know
2:21:10 it's like it's like
2:21:12 I don't know for me this is kind of like
2:21:14 this is I think this is why I exist
2:21:18 I've always been weird like this like
2:21:20 people are
2:21:22 people in my life they just they just
2:21:24 kind of roll their eyes and like up
2:21:26 there he goes again
2:21:28 because he thinks he found another one
2:21:33 because I was sort of like this with the
2:21:34 blockchain and nft [ __ ] for a while and
2:21:36 then the deeper I dug into that the less
2:21:38 reel it got and then this was the
2:21:40 opposite the deeper I dug the more real
2:21:42 it got I was like oh [ __ ] this is this
2:21:44 is it
2:21:46 anyway I gotta go it's 10 19. I'll do
2:21:49 another 10 minutes here yay sixty
2:21:51 thousand now we're at 77 000. all right
2:21:53 I'm gonna I'm gonna scroll to the bottom
2:21:55 and then I'm gonna
2:21:56 uh Stevo thinks 80k is doable yeah I
2:22:00 think so we're at 77. I'll I'll hang out
2:22:02 to you until you hit 80 000.
2:22:05 because it's all about the Irregulars
2:22:07 here
2:22:08 I hope you all know that this is not
2:22:09 about me it's about the Irregulars at
2:22:11 this point they now own the channel even
2:22:13 though douchey mcscamington thinks he
2:22:16 owns the channel it's really the
2:22:17 Irregulars
2:22:18 which by the way some of the Irregulars
2:22:20 are [ __ ] with him and like talking to
2:22:22 him and trying to suss out where he is I
2:22:25 don't think he's in the U.S it feels
2:22:27 very much like
2:22:29 um
2:22:30 offshore crypto energy to me
2:22:34 and I'm sure you're in here and if you
2:22:36 are
2:22:38 sorry you can't see that there you go
2:22:40 against the black shirt
2:22:48 I'm doing AC and away we go Clint all
2:22:52 right later Vicky we need an easy button
2:22:59 um I'm 58 with 80 D2 tap tap tap look a
2:23:03 bird welcome welcome Ted
2:23:07 yay 60.
2:23:09 um can you explain a seed number sure so
2:23:12 when you generate an image in
2:23:14 mid-journey or in stable diffusion
2:23:17 you can generate it with a seed number
2:23:20 you can put a seed number in there it
2:23:22 basically is just a serial number for I
2:23:24 don't quite what know what it's doing
2:23:26 mathematically but basically
2:23:29 um you're sort of picking
2:23:31 you're you're picking a point in latent
2:23:33 space that it is sort of clustering its
2:23:36 things around I don't really know what
2:23:38 the [ __ ] it's doing but you're doing
2:23:40 something where
2:23:43 you're using a consistent set of inputs
2:23:46 multiple times so generally the way
2:23:48 seeds are used
2:23:50 is you'll generate an image and then
2:23:52 what you'll probably do is generate like
2:23:54 20 images and then there there might be
2:23:56 one image that you really like and
2:23:57 you're like oh I want to do variations
2:23:59 of that one but you want to keep like
2:24:01 the the composition kind of the same and
2:24:03 the you like most of what's in it
2:24:07 excuse me
2:24:09 um you can then get that seed number and
2:24:11 you can figure you can just basically
2:24:13 Google this or or like you know go to
2:24:15 Bard and ask it and it'll it'll tell you
2:24:17 um for how to get the seed numbers for
2:24:19 stable diffusion or for mid-journey and
2:24:22 then
2:24:23 um when when you can modify your prompts
2:24:25 so do the same prompt but slightly alter
2:24:27 it but then include the seed number
2:24:29 it'll basically take that same base uh
2:24:33 image information and kind of modify it
2:24:36 you know sort of from us from the same
2:24:38 starting point so you have some
2:24:40 consistency from those things
2:24:43 now that mid-journey has in painting and
2:24:46 and they've got
2:24:47 um zoom out and and things like that
2:24:50 um and now that we've got generative
2:24:52 fill within Photoshop the seed number is
2:24:54 becoming kind of less necessary to get
2:24:57 some consistency it was basically a hack
2:24:59 to get some consistency across images
2:25:03 um but that's basically what it is it's
2:25:05 kind of like a serial number for that
2:25:06 generation
2:25:08 generations and uh and uh
2:25:13 uh you know keep some consistency with
2:25:15 them
2:25:16 um are the PO chat Bots the same as the
2:25:18 agents
2:25:20 um I don't think they call them agents
2:25:22 no if you mean autonomous agents no
2:25:24 they're not so autonomous agents are
2:25:27 like a series of tools strung together
2:25:29 where you give it a goal and then it
2:25:31 kind of is using the large language
2:25:33 model to prompt itself and assign itself
2:25:37 tasks so one of those tasks might be go
2:25:40 search
2:25:42 um Google one of those tasks might be
2:25:44 write something you know embed something
2:25:46 and write it to the vector database one
2:25:48 of those tasks might be send a prompt to
2:25:51 open AI with the information I just got
2:25:52 from the Google search
2:25:54 and and it it's a it's a long series of
2:25:56 strings where the autonomous part is
2:25:59 it's using tap GPT to prompt itself and
2:26:03 if you want to see it in action he is is
2:26:06 a bit of an autonomous agent
2:26:08 so
2:26:10 so I'm in chat gbt you get to cut
2:26:13 interpreter here if you haven't played
2:26:15 with it before this is some mind-blowing
2:26:17 [ __ ] too so you have this little plus
2:26:19 button here so you can upload you can
2:26:21 upload multiple files
2:26:23 and let me see if I can go find okay
2:26:26 here's an archive one dot zip I don't
2:26:30 know what's in here but that doesn't
2:26:32 matter it will help us figure it out
2:26:35 is it going to upload let's see maybe
2:26:39 that one was too big hang on let me get
2:26:41 something else
2:26:43 archive two is that
2:26:48 well we'll see
2:26:52 yeah there it is okay
2:26:54 um
2:26:58 please unzip this
2:27:01 and tell me what the contents are and if
2:27:04 it's a CSV
2:27:06 let me know what the data is and what
2:27:10 might be interesting about it
2:27:14 Okay so
2:27:18 all right let's start by unzipping the
2:27:20 file and checking its contents so you
2:27:22 see it it popped up that little
2:27:24 um that little plug-in looking thing if
2:27:26 I say show the work
2:27:27 what it's actually done so so what code
2:27:30 interpreter is doing is when you launch
2:27:33 a code interpreter chat it's spinning up
2:27:35 a virtual machine on the back end
2:27:38 that has a bunch of python libraries in
2:27:41 it or access to a bunch of python
2:27:42 libraries
2:27:44 and then when you ask it something it
2:27:46 starts writing and executing its own
2:27:49 python code and it starts analyzing data
2:27:52 so the first thing it does is it pulls
2:27:54 in a library to unzip the file
2:27:59 talking to itself
2:28:02 executing code so it says given the
2:28:05 vastness of the data and the numerous
2:28:07 CSV files available please let me know
2:28:08 if you have a specific questions or if
2:28:11 there are particular aspects of the data
2:28:12 you'd like me to dive deeper into
2:28:15 um
2:28:17 270 unique countries do you have U.S
2:28:21 data
2:28:27 so it's going to go back look into the
2:28:29 data
2:28:31 the other thing it does with data like
2:28:33 this is when you have it start actually
2:28:34 analyzing the data it creates a data
2:28:37 frame in its virtual computer memory and
2:28:40 it basically loads in all of your data
2:28:43 into its own internal data structure so
2:28:46 it can analyze it locally basically so
2:28:48 it can do it really fast and then you
2:28:51 can have it output
2:28:52 other CSV files so you can say let's say
2:28:55 you had 10 CSV files at all had
2:28:57 disparate data and it wasn't clean data
2:28:59 it'll pull all that in and then you can
2:29:01 say make me a Consolidated you know data
2:29:04 set in a CSV file or in a in an Excel
2:29:07 file with multiple worksheets and it
2:29:09 just will build it for you
2:29:11 the first CSV contains U.S data this
2:29:13 includes so I'm going to say
2:29:15 um
2:29:17 please identify
2:29:20 what you feel
2:29:23 um
2:29:26 are the most interesting trends
2:29:29 and visualize that data
2:29:33 in a visually stunning way
2:29:42 to identify interesting data
2:29:45 here's here's the steps I'm going to
2:29:47 give myself so this this again is that
2:29:49 idea of of an autonomous agent I said I
2:29:52 wanted to unders to give me interesting
2:29:54 Trends so it said I'm going to give
2:29:57 myself these
2:29:59 tasks I'm going to do some data cleaning
2:30:01 I'm going to do some Metric selections
2:30:03 and then I'm going to do the visualize
2:30:04 visualization so so for you manual
2:30:08 readers on on the call tonight uh that
2:30:11 read manuals this is doing its own
2:30:13 project management right this is doing
2:30:15 its own list making let's start with the
2:30:18 data cleaning so did that let's proceed
2:30:21 to this and with wait let's see
2:30:23 calculate the variation in each series
2:30:25 over the years repeaters in grade one of
2:30:28 primary education both sexes so wait
2:30:30 based on the variation across the years
2:30:33 the top series with the highest
2:30:35 variations are repeaters of grade one
2:30:39 both sexes male female gross enrollment
2:30:43 ratio
2:30:44 to visualize these Trends we'll need
2:30:46 line plots for each of these series
2:30:49 showing the changes over the years these
2:30:52 will provide visually appealing
2:30:54 representation of the data and so now
2:30:56 it's gonna there so there so it wrote
2:30:59 code to visualize the data so so like I
2:31:04 just this this [ __ ] doesn't it fail to
2:31:07 just blow my [ __ ] mind every [ __ ]
2:31:10 time
2:31:12 like I don't know what I'm doing with
2:31:14 data visualization at all obviously
2:31:18 but I don't have to now
2:31:20 right like if I knew even a little bit
2:31:23 about what this data set was if I was
2:31:24 running my business
2:31:26 and I had a data set of customer
2:31:28 Behavior or customer support or whatever
2:31:32 it might be I can now just upload it to
2:31:34 this thing
2:31:35 I don't need a data analyst anymore
2:31:39 like unless you're a real relatively
2:31:41 large company you can't afford them
2:31:42 anyway but what you what you do end up
2:31:44 hiring is consultants and you know
2:31:47 brainiacs that are data analysts
2:31:50 well you don't need to now
2:31:53 right
2:31:55 and I mean I mean what did this take me
2:31:57 three minutes four minutes and and what
2:32:00 did I do I've I typed twice the first
2:32:02 thing was what's in here then I asked is
2:32:05 US data in there and then I said tell me
2:32:08 what's interesting and visualize it I
2:32:10 wrote three things
2:32:12 maybe a total of 30 words
2:32:15 and it did
2:32:18 all of this
2:32:21 on its own
2:32:26 this ain't [ __ ] Google
2:32:30 the Santa search engine
2:32:35 everything's different but no one knows
2:32:37 it yet
2:32:39 that's what's happening that's what
2:32:41 we're in the middle of that's whoever
2:32:43 asked I forget who asked like
2:32:45 how do I maintain the sort of
2:32:47 inspiration to keep learning this [ __ ]
2:32:50 because every time I do that
2:32:53 it just reminds me like
2:32:55 [ __ ]
2:32:58 [ __ ]
2:33:00 this is insanity
2:33:03 foreign how lucky are we that we get to
2:33:07 play with this
2:33:09 you can go play with that right now you
2:33:11 can go just do what I did where did I
2:33:12 get that data
2:33:13 do you know do you know where to get
2:33:15 data sets because I didn't
2:33:17 but you know what I did I asked chat gbt
2:33:20 where do you get data sets oh kaggle.com
2:33:24 I don't know have you ever heard of
2:33:25 kaggle if you're a manual reader I bet
2:33:28 you have but as a non-manual reader I
2:33:31 didn't know what kaggle was you can go
2:33:33 get data sets what are they I don't like
2:33:36 people put [ __ ] in there about the NFL
2:33:37 and education and health and
2:33:40 I don't know toenail picking it's like
2:33:44 and then you get one of those pieces of
2:33:46 data and you can just go start playing
2:33:47 with it like like Play-Doh
2:33:52 you know like I mean
2:33:53 do you know how long it would take you
2:33:58 to a understand basic statistics
2:34:01 Concepts
2:34:03 and then
2:34:06 um
2:34:07 learn enough okay I found this on the
2:34:10 web series statistics really check it
2:34:13 out no
2:34:18 um and then you'd have to learn enough
2:34:19 python
2:34:21 and enough math
2:34:24 and enough data structure oh we can't we
2:34:27 can't do this because the data is not
2:34:28 clean because people didn't input their
2:34:30 address consistently across these six
2:34:32 different fields it doesn't care about
2:34:35 any of that it just fixes that [ __ ]
2:34:39 it's staggering
2:34:41 and like this is one teeny
2:34:45 little
2:34:46 sliver of what's happening in generative
2:34:49 AI it's one teeny little tool inside
2:34:52 this Grand
2:34:54 [Applause]
2:34:55 thing that is chat GPT
2:34:58 can we get zapier to tap the screen for
2:35:01 us I bet you can my pass at work has 16
2:35:04 characters with a bunch of other rules
2:35:06 of course
2:35:08 total totals
2:35:10 um it's 10 34 I said I was leaving at 10
2:35:12 30. what have you people done to me and
2:35:16 from 12 49 to 45.97 listen this channel
2:35:20 is not worth a 50 auto screen Tapper I I
2:35:23 I I you cannot buy 50 auto screen
2:35:27 tappers listen if you're gonna drop 15
2:35:29 bucks on an auto Tapper that's on you
2:35:34 oh my God that's hilarious all right I'm
2:35:37 jumping to the Bottom now this time for
2:35:38 sure for real for serious man I'm doing
2:35:42 it all right
2:35:44 all right yeah they're wimpy okay let's
2:35:46 see let's see let's see all right this
2:35:48 changed my entire business yep
2:35:52 Google is gonna tell us tomorrow cool
2:35:55 90k oh is tomorrow when their thing is
2:35:59 um still need someone to pay me to do
2:36:01 all these amazing things with chat gbt
2:36:02 mama needs a new laptop
2:36:04 [Laughter]
2:36:06 seriously lunar stick though as you're
2:36:09 learning about this stuff start talking
2:36:11 about it talk about it on Facebook talk
2:36:12 about it on LinkedIn come to the salon
2:36:14 talk about what you're learning there
2:36:17 there's there is
2:36:19 like the value right now in anyone who's
2:36:22 learning anything talking about it is
2:36:25 goes Way Beyond what it will serve you
2:36:27 it it will help other people but you
2:36:31 also don't know
2:36:32 right now because everyone's in kind of
2:36:35 you know sponge mode or as I said the
2:36:38 other night I'm in Swiffer mode I'm in
2:36:39 like super sponge mode
2:36:42 um everyone's trying to learn this stuff
2:36:44 so if you just put it out there that
2:36:46 you're learning this you're
2:36:47 experimenting you're playing you wrote A
2:36:49 Country Song you did this doesn't matter
2:36:50 just start talking about it here's my
2:36:52 opinion on this I think that's exciting
2:36:54 I think that's not doesn't matter
2:36:56 your friends will think you're a [ __ ]
2:36:58 weirdo but they already do so [ __ ] them
2:37:00 right
2:37:02 and then the rest of the people would be
2:37:04 like oh she knows what she's talking
2:37:05 about and at some point one of them's
2:37:07 gonna go you know I need to can I talk
2:37:09 to you and just ask you a few questions
2:37:11 and they might ask you really basic
2:37:13 stuff they might ask you stuff you don't
2:37:14 know answers to
2:37:15 but it will it will it will serve you so
2:37:21 um anyway
2:37:23 all right everybody
2:37:24 I'm gonna do this I'm going to take my
2:37:26 little pimple under my chin I'm going to
2:37:28 take my 17th chin I'm going to cover it
2:37:30 up like this humiliated I've got body
2:37:34 positivity issues now I don't
2:37:37 [Laughter]
2:37:40 um it's been a pleasure love hanging out
2:37:43 with you guys happy Monday night thank
2:37:45 you for hanging out uh oh the hotel oh I
2:37:47 should I should really hang out a bit
2:37:49 longer
2:37:50 you got 93 000 likes should I do it now
2:37:54 I'm gonna go I'm gonna go I gotta go I
2:37:56 gotta do this wait you need you need a
2:37:58 life I need a life congratulations you
2:38:02 almost got to 100K
2:38:04 uh Emilio's wife good night Tobias
2:38:07 lunish thick nugs for Life uh Zach zero
2:38:10 Pippins it's been nice you've been in
2:38:12 here A bunch of times nice seeing you in
2:38:13 here again brother 52 Winston thank you
2:38:17 Vicky good night good night uh Karen
2:38:19 let's see who else we got frumple Share
2:38:23 Bear night Irregulars Kurt all right
2:38:25 everyone it's uh it's Romper Room good
2:38:29 night everybody I appreciate your time