9/23/2024 - Exploring Generative AI, NotebookLM's Podcast Feature, and the Future of Work

Description

Chapters

Transcript

0:03 [Music]
0:10 oh
0:12 [Music]
0:40 [Applause]
0:42 [Music]
0:48 and your quiet
0:49 voice and impeccable
0:52 [Music]
0:54 style don't ever let them steal a
0:57 joy A Gentle Way
1:01 keep them from running
1:04 wild they kick in your face dress you
1:08 down and tell you that your place is in
1:10 the middle when they hate the way you
1:15 shine I see you tugging on your
1:19 shirt hiide of it hide how much it
1:25 hurts let play while
1:29 [Music]
1:31 let
1:32 spin let them scatter in the
1:37 wind I've been to the movies I've seen
1:41 how it
1:42 is and it jokes on
1:47 [Music]
2:23 good evening good people of The Tick
2:26 Tock Amelio's wife seems to be on time
2:31 I did indeed get in a fight with the
2:34 wife over the pink bow but the pink bow
2:36 has been
2:39 secured so so my machismo one
2:46 out and I have my pink
2:49 bow ready for display
2:52 [Music]
3:09 10,000 words SW around my head 10
3:12 million more in books written beneath my
3:20 bed I wrote All read them all when
3:23 searching in the Swarms still can't find
3:26 how to hold my hands
3:30 [Music]
3:32 and I know you need me in the next room
3:35 over I'm stuck in here I'll
3:41 [Music]
3:43 paralyzed four months I got
3:47 myself spinning mirrors framed in yellow
3:51 [Music]
3:55 walls like people different like things
3:59 we don't know
4:00 [Music]
4:04 about like most people I'm no different
4:08 like things we don't
4:11 [Music]
4:16 know what do you think chy
4:19 [Music]
4:38 you trying to sing harmony I thought we
4:40 were singing in
4:41 unison what what I don't know what you
4:44 want sometimes when you have a
4:46 temperamental Diva kind of a performer
4:48 like this that doesn't understand words
4:50 you can call them temperamental in a
4:52 diva cuz they don't actually understand
4:53 words but uh you run into issues right
4:57 who's who's leading who's the lead
4:59 singer here now it's just walking away
5:00 you see what
5:01 [Music]
5:10 happens he got his ass facing at
5:14 me it's cute
5:17 [Music]
5:40 oh good lord good people happy happy
5:43 Thursday
5:45 evening hump day has
5:47 humped they don't we we don't have we
5:50 don't have
5:52 some mythical I hate work name for
5:55 Thursday do we it's it's
6:01 I I got
6:04 nothing somebody's got a case of the
6:07 [Laughter]
6:10 Mondays um have you tried Google's new
6:14 image gen 3 Model no did it what is it
6:19 is it like within days cuz I I think I
6:21 would have played with it if it came out
6:23 any earlier than
6:27 that
6:28 um there a new you are handsome well
6:31 thank you so much I appreciate that or
6:33 were you talking about the llama
6:38 [Music]
7:16 oh good lord I just got an email about
7:19 it today all right we'll look it up so
7:20 let me add that to the list in
7:22 fact image gen wait oh no it's IM
7:28 imen that's that's one of those names
7:31 it's not as bad as
7:34 swarm or 01 40 and
7:38 40 but it's hard to say
7:43 imagen um and and it gets autocorrected
7:47 by
7:48 Google number three uh I haven't played
7:51 with
7:52 it imagine
7:55 three by Google oops
8:00 okay it's on the
8:01 list it's on the list we got hey Jen we
8:04 got a mad
8:05 genen we have not played with a Maden 3
8:09 I'll add it to the parking lot zero
8:12 chance we get to it
8:15 [Music]
8:22 tonight so good evening everybody my
8:24 name is Kyle Shannon this is the AI
8:26 learning lab I play a little guitar
8:28 occasionally it doesn't last for long
8:31 I'll I'll put you out of this m misery
8:33 soon
8:35 enough just usually what I do to warm up
8:38 let people come in get settled wow all
8:40 of a sudden we got a bunch of people in
8:42 here looks like Tik Tok all of a sudden
8:43 decided to show this to all of you so if
8:48 you're curious about
8:50 AI pop your comments in the uh chat and
8:54 we've got mods in here will who grab
8:56 interesting uh questions or comments or
8:59 thoughts thoughts or projects you're
9:01 working on and pop it in the in the chat
9:03 and then we'll talk about stuff so what
9:05 this is this is this is like
9:07 freewheeling
9:08 AI jam session so the the the uh the
9:13 community here kind of hangs out one of
9:15 the places that they hang out is a place
9:18 called The
9:19 Salon the salon. so if you go to that
9:22 website and click on the join the
9:24 community button that'll take you to our
9:27 community site and you can go there and
9:29 check out what we're out doesn't cost
9:31 anything and this is a whole community
9:33 of AI Optimist as is this Channel and
9:36 what we do is five nights a week I go
9:38 live on here and we talk about Ai and
9:42 you're like AI why are you going to talk
9:43 about AI I don't know we talk about
9:45 whatever we show cool new tools we work
9:49 on projects sometimes we write songs
9:51 sometimes we make movies one night we
9:54 had it write a whole
9:56 screenplay was
9:58 amazing one night I wrote a book called
10:01 called embrace the Jank I still haven't
10:03 finished it cuz I have two images to
10:05 make and I can't get them to make it so
10:07 it's too janky to finish embrace the
10:10 Jank which if I listen to the title of
10:12 the book I'd just finish it wouldn't
10:18 I I want to run flux. one Dev on a CPU
10:22 with 32 gab of RAM is that possible I
10:26 don't
10:28 know I don't know the details of flux
10:31 the only flux I've used has been in in
10:33 other services that have Incorporated it
10:35 um by the way if you if you have access
10:39 to Korea korea. a k re a. they've now
10:45 added four four or five different video
10:49 models so they've got cling they've got
10:51 the new Minx from Halu or he Lao or low
10:55 however the hell you say that they've
10:58 got uh
11:00 is it flux might be flux in there I
11:02 don't know they got they got four of
11:04 them in there they got runway in there
11:05 but Runway seemed to be disabled
11:07 yesterday um but that was pretty
11:11 slick so so if you got questions pop
11:14 them in sunos scenes require an iPhone
11:17 right I I have not played I saw that
11:19 sunno had a thing out called scenes um
11:22 sunno needs to cool their [ __ ] Jets
11:25 sunno is
11:26 launching a major new feature about
11:28 every 3 days lately it's like holy crap
11:33 they just they just launched um image to
11:36 song and video to song which I haven't
11:39 played with I just sent it to my my
11:41 writing partner for the musical cuz he's
11:43 been doing a lot of stuff in
11:45 sunno they added in painting now they've
11:48 got this thing called scenes if you have
11:50 any questions about making images or
11:52 prompts ask
11:54 away Uh Oh Hang on we're losing the
11:57 keyboard cover it's not good people we
12:00 can't have glare the people that hang
12:02 out in here nightly they get very upset
12:05 if there's keyboard
12:06 glare it's not good it's not
12:11 good um there's a lot of changes to to
12:15 chat jbt lately if you haven't been in
12:18 there's there's new underlying models
12:20 there's new
12:23 interface I haven't played with Claude
12:26 in a while I would imagine claude's
12:28 still interesting scen slso is on
12:32 iPhone um I Heard Claude has a new uh
12:35 iPad app let's see if that's true this
12:39 is oh the other nickname of the channel
12:40 is is uh chat ad we're back down to
12:43 normal numbers by the way I don't that's
12:46 just a bizarre thing that Tik Tok does I
12:49 think they just I think they just like
12:53 randomly take this you know Channel with
12:55 a fat old guy sort of talking about AI
12:59 and they just splat it out to a bunch of
13:02 you know
13:04 16-year-old boys expecting twerking and
13:07 they get
13:09 this they're just like
13:12 whatever what okay
13:15 Boomer next
13:20 F all
13:22 right
13:25 cloud cloud cloud cloud by anthropic
13:30 let's download
13:31 it yeah apparently claude's got a new
13:34 iPhone app and a new iPad
13:38 app see if it looks any
13:55 gooder all right it doesn't look that
13:57 different
13:59 choose
14:02 project uh let's see
14:05 make me a
14:10 Frogger
14:15 game let's see if it's got if the iPad
14:18 app has
14:30 we'll see if this turns into a playable
14:31 game oh oh yeah it sure enough
14:37 did all right so if you got an iPad the
14:41 new the new CLA is out it's got It's got
14:45 artifacts on on the iPad Claud had an
14:48 update like today it's smarter okay wow
14:52 if it's smarter that's great but did it
14:54 it didn't have an update to Opus did it
14:56 the new the new fat model that they've
14:58 been promising or not promising that
15:00 people have been speculating there have
15:02 been a lot of chat TMZ about
15:05 that
15:11 [Music]
15:15 um I got nothing oh wait sonnet do they
15:20 have
15:24 new no Claud 3 Opus 3.5 Sonet so same
15:28 old model but maybe it got smarter has
15:31 anyone done any work on Notebook LM I'm
15:33 going to actually show that tonight I've
15:34 been having some fun at work and blowing
15:38 my boss away so sugar cube hang on so
15:41 they just added today they just added
15:43 the ability to actually guide those
15:46 podcasters to talk about um specific
15:50 things you want them to talk about so
15:52 I'll show you how to do that tonight if
15:53 you want to hang
15:55 out Google released illuminate right
15:58 where they basically just took the the
16:01 podcast feature out of notebook LM and
16:04 made it a standalone product we called
16:06 that one that was one we predict I
16:08 predicted that one three days after that
16:09 thing came out I'm like they're going to
16:11 spin this out as a separate product too
16:12 many people are talking about
16:15 it and they did which is good it's
16:18 surprising CU a lot of times Google has
16:20 really cool [ __ ] and it just stays
16:21 buried
16:25 forever oh man all right I'm the one
16:28 that's late tonight oh oh joy py what
16:30 are we going to do with you Jason Woo
16:33 this has inidia been talked about they
16:39 jumped in the AI race
16:42 oh here's the thing with
16:45 Nvidia they always make these
16:47 announcements like I saw that they put
16:48 out a 70 billion parameter model but is
16:51 it out like we can play with it or is it
16:53 just out that you can play with it if
16:55 you have an Nvidia GPU card in your
16:58 computer cuz as a Mac guy I'm kind of
17:02 [Music]
17:06 screwed academics are
17:09 discussing the gender bias in the
17:11 notebook LM
17:13 podcasters oh I'm sure they
17:15 [Music]
17:26 are I think people are people are giving
17:29 that podcast thing a bit too much credit
17:31 it's like it's it's a single prompt for
17:34 a tool that they had in
17:38 there apparently what's coming with
17:40 illuminate is the ability to choose
17:42 different host and
17:44 guest and so I would imagine that the
17:47 gender bias has something to do with
17:49 who's the host who's the guest and
17:51 they're they're probably like well he's
17:52 mansplaining to her but sometimes if
17:55 she's the host she'll man Spain to him
17:57 so
17:58 [Music]
18:16 Silver Fox I
18:18 use notebook LM to I just lost
18:23 it Tik Tock is just a pile of
18:27 garbage I use it to give me an overview
18:29 of long projects I'm working on that's
18:31 pretty cool the podcaster in Notebook LM
18:34 can we change the language not yet not
18:36 yet Mr it but go to um
18:39 illuminate. goole.com and that's a
18:43 standalone version of the podcast maker
18:45 and they're adding features to that so I
18:47 would assume at some point they're going
18:49 to roll them into notebook LM but my
18:52 suspicion is that anything like
18:54 different languages and the ability to
18:56 really modify the podcasts you're going
18:58 to get it illuminate but it's a very new
19:00 product so it's going to be it's going
19:01 to be janky like all the stuff in AI so
19:05 if you now we have 1.2 th000 people in
19:09 here that cannot be right or we
19:14 have
19:16 1.2k well hey if you're new here you
19:20 must be thinking who the hell is this
19:22 guy and why are they showing me this
19:27 live there you go go viewers on your way
19:30 oh thank you very much James K I'm like
19:33 I'm like what the hell's going on here
19:35 somebody dumped a bunch of viewers in
19:37 here uh if if you are new here um my
19:41 name is Kyle Shannon this is the AI
19:42 learning lab and what we do is we talk
19:44 about AI every night I go live five
19:46 nights a week right here on the Tik Tok
19:50 and
19:51 uh you know one of the questions we
19:54 often get is you can make money with
19:56 chat jbt how do you make money with chat
19:58 jbt we talk about that we talk about
20:01 what's cool what's happening what's
20:02 going down how do you make money with
20:05 chat jbt we talk about that we talk
20:08 about what's cool what's happening
20:09 what's going
20:10 down how do you make money with chat jbt
20:13 we talk about that we talk about what's
20:16 cool what's happening what's going
20:18 down all sometimes go on philosophical
20:20 rants about what I'm excited
20:24 about h
20:29 I just
20:30 froze all right now I
20:34 unfroze following along for free at the
20:36 salon. oh yeah follow along for free so
20:39 go to the salon. a that's our community
20:42 go check that out uh and you can hang
20:45 out in there with a lot of the people in
20:47 here and if you have questions about AI
20:49 pop them in the comments below and we've
20:51 got mods that will grab those questions
20:52 and pin them so I can see them and uh
20:56 yeah we can do that and uh yeah it be
20:59 fun fun to be back with the AI learning
21:00 lab tonight Firefly what's happening
21:02 haven't seen you in a while welcome back
21:04 brother 52 what happened to Tom marold
21:07 exactly Tom Arnold's really let himself
21:11 go is that Rosie O'Donnell what has she
21:15 done here there you go I'm Rosie Odon hi
21:20 welcome to W welcome to AI corner with
21:22 Rosie odonnell my name's Rosie odonnell
21:28 [Music]
21:31 full
21:35 [Music]
21:50 moon all right I think the first thing
21:52 we'll do is we'll pop on over to to
21:55 notebook LM if you haven't seen the new
21:58 the new feature inside notebook
22:00 LM we'll pop over there and I'll show
22:02 you the new feature that they added to
22:04 it um generous community at the salon. a
22:08 I agree with that yeah it's glitchy yeah
22:09 something's really glitchy here
22:11 tonight uh could be because we've got a
22:14 lot of folks in here so if you're new
22:16 here uh I'm not Alec Baldwin I'm not Tom
22:20 Tom
22:22 Arnold I'm not Rosie O'Donnell but I'm
22:25 so much
22:27 more he's a real boy you found it I did
22:30 I found my I got in a fight over this I
22:33 told you I was going to get I told you I
22:34 was going to get in a fight last night
22:37 so apparently if you're here and you're
22:40 female you're thinking right now he's
22:43 not wearing that
22:44 right and if you're a man you're like
22:46 well of course he's wearing that right
22:48 but why is he wearing
22:50 it so you see you see you can't please
22:53 everyone that's the lesson here tonight
22:56 and AI is the same this pink bow is a
22:59 metaphor for
23:00 AI I don't know
23:03 how I got nothing to back that
23:06 [Laughter]
23:11 [Music]
23:25 up Freedom came my way that night
23:31 just like a jet play in and out of side
23:36 that's my doggy come on champ you want
23:39 to
23:42 [Music]
23:45 singon wondering how hard i'
23:50 H when they came into the
23:54 [Music]
23:56 station they said I was bad on
24:01 repairs but I got no qualms with my
24:06 situation say here I
24:09 [Music]
24:12 am so say sh Shere Shere won't you dare
24:18 to say Shere Shere Shere won't you dad
24:24 to say sh Shere Shere
24:28 [Music]
24:29 H yeah leave a message and your number
24:34 please which one take a time to want to
24:38 satisfy
24:39 me which
24:42 one hey Siri go away okay Siri's so
24:49 jealous sorry I don't understand
24:52 Siri Siri M yeah could you go
24:56 now I am here yeah I know could you
25:01 leave
25:03 Siri mhm go away
25:12 okay yeah we we were fighting over the
25:14 bow well we were fighting over the fact
25:17 that how do you lose a bow how do you
25:19 lose a pink bow that's what we were
25:21 fighting over and then she's like why do
25:23 you need your stupid [ __ ] pink bow
25:24 you gave me the pink bow I said I know
25:26 you gave you the pink bow but I need the
25:28 pink bow cuz
25:29 my audience expects the pink
25:31 bow and then she said something to the
25:34 effect of I don't give a [ __ ] about your
25:36 pink bow and I said well I need my pink
25:44 bow oh marriage is
25:47 bliss uh Phoenix i.x do you think an AI
25:51 that helps you learn paid ads or wait oh
25:55 do I know of an AI that will help you
25:57 learn paid ads or help you with paid ads
26:00 um I don't there there are probably for
26:03 really specific applications like that
26:06 um there's there's kind of two ways to
26:08 approach it you can roll your own
26:10 meaning you could go to
26:12 perplexity um perplexity is really good
26:15 at researching things on the internet so
26:16 you could just go to perplexity and say
26:18 hey help me figure out how to learn paid
26:20 ads and it'll basically write you an
26:22 article with research and links to
26:24 things um you could go to Claude or chat
26:27 GPT and just have it give you lessons or
26:30 you could go try to find an application
26:32 that's specifically you know built to do
26:35 that um I would say go to futur
26:38 pedia.com
26:56 [Laughter]
26:59 good
27:00 luck because it's really easy to make
27:03 apps now and because of that there's a
27:04 bunch of garbage out there uh this
27:07 community is the foundation for those
27:09 curious about AI generous about Ai and
27:12 playful with I AI indeed in fact Let Me
27:15 Go show you something here so if you if
27:18 you've not been to the AI Salon go to
27:20 that URL and you can read about what
27:23 we're about there it's just a a light
27:25 little site but there's a button there
27:26 that says join our community and when
27:28 you hop over to the salon
27:33 Community the first thing you'll see is
27:35 start your adventure so there's the
27:36 seven step Adventure there's a welcome
27:39 video um where Leah faston and myself we
27:42 both we founded the salon the week after
27:44 chat GPT came out and so you can go um
27:49 you can go watch that welcome uh you
27:51 introduce yourself is the next step
27:53 learn about our values um watch previous
27:55 meetings RSVP to the next meetings hang
27:58 out and chat
27:59 uh and then there's links to our socials
28:01 in there and then there's all sorts of
28:02 cool stuff to do by the way for those of
28:05 you wait oh I got to re reload this page
28:07 it looks like for those of you that uh
28:11 are at this site or are are at the AI
28:16 learning lab regularly for you
28:17 Irregulars um we changed the tools and
28:21 play section back to show and tell so we
28:25 originally had a showand tell section
28:27 and then we we added tools to it and we
28:30 we sort of named it tools and play and
28:33 everyone was like where's the showand
28:34 tell section we still had all the
28:36 showand tell in there but we changed it
28:38 back today to show and tell so there's
28:41 Show and Tell for llms and gpts images
28:44 coding and tech music and audio video
28:46 and animation 3D and World building and
28:48 in every one of these they Now default
28:50 to the show and tell um images and stuff
28:54 so it's more funner and then you've got
28:57 here's recommended tools and here's tool
29:01 talk where you can talk about image
29:02 tools and things like that all
29:05 right so go check that out go check out
29:09 the salon it is an amazing community and
29:14 one of the things to know about this AI
29:18 stuff
29:20 um one of the things to know about this
29:23 AI stuff is
29:26 that it is moving incredibly
29:29 fast and and when I talk about this AI
29:32 stuff what this Channel's focused on and
29:33 this is important because if you're if
29:35 you're an engineer or if you've if
29:37 you've been in the AI game for 30 years
29:41 um you think of AI as the broader
29:44 definition of AI with you know all of
29:46 the AI stuff
29:48 convolutional neural Nets and all that
29:50 sort of happy horse [ __ ] what we
29:53 primarily focus on in this channel is
29:56 the AI that happened
29:59 starting on November 30th 2022 and
30:01 forward right so the launch of chat gbt
30:04 so this channel is primarily focused on
30:06 generative AI I don't spend a ton of
30:09 time on the open source part of that
30:12 because it's way too complicated and
30:14 move that like it's fragmented even more
30:16 than than the commercial tools so we're
30:19 primarily talking about you know the big
30:21 players the big tools and then what you
30:23 do with them and how they're going to
30:25 impact our work how they're going to
30:27 impact um our friendships our
30:30 relationships um our self-esteem our
30:34 self-worth so that's the kind of stuff
30:36 we doing here all right um
30:40 beautiful all right what have we got all
30:43 right we're up to 15,000 likes and we're
30:45 just getting started all the tapping
30:47 that you see there so this is some other
30:50 context uh a lot of times the people
30:53 that come here night after night
30:54 there're believe it or
30:56 not and I know that the answer is going
30:59 to be not but believe it or not there
31:01 are people that come here night after
31:03 night after night every
31:06 night and and they're just weirdos and
31:10 so one night I said to them I said hey
31:13 y'all are a bunch of weirdos and then
31:15 one of them said we're not weird we're
31:20 irregular so the Irregulars are the
31:22 people that show up here night after
31:24 night the reason they do that is because
31:29 this becomes a space where all of us can
31:32 can just be in this conversation about
31:34 Ai and playing with it and even if
31:35 you're not listening to what I'm saying
31:37 which is probably 90% of the time or
31:39 should
31:41 be you should be making images making
31:44 videos making songs be on chat GPT
31:46 exploring stuff just have me in the
31:48 background go to go to sleep with me but
31:50 be playing with
31:51 AI um this becomes an opportunity for
31:54 all of us to just be in this
31:56 conversation I just wrote an article on
31:59 LinkedIn here let me show it to you cuz
32:01 it's got a fun image that's Joker made
32:03 that image by the way the Llama AG is
32:06 coming are you ready very nice image
32:08 Joker um let's see here let me go to the
32:13 LinkedIn oh
32:25 LinkedIn so I wrote this article
32:29 called
32:30 Ai and the holy crapp ification of
32:36 work
32:39 and we're we're in a remarkable time in
32:43 history right now
32:44 because work currently looks like what
32:49 work has always looked like you go to
32:51 work you or or you go to your Zoom calls
32:54 because because of covid you don't
32:56 really go to the office anymore so
32:58 that's changed but um but but you're
33:00 still basically doing the same [ __ ]
33:02 right you're you're going in you've got
33:04 your to-do list you have your Zoom
33:06 meeting you know Sally says hey Jim
33:09 how's it going on the to-do list and
33:10 you're like yeah I got it going I'll
33:12 have it done by the end of the week
33:13 she's like are you sure cuz you were
33:15 late last week you know we're putting
33:17 the new covers on the TPS reports right
33:19 you know that
33:20 [ __ ] that's about to be upended
33:26 and the fascinating place we are right
33:28 now with AI is that in the past 3 or
33:31 four
33:32 months essentially all of the AI tools
33:36 whether it's large language models or
33:38 image generation tools or music tools
33:41 have
33:43 gotten so much better than than they
33:47 used to be
33:50 that if you regularly use them in work
33:54 or even you know outside of work but
33:56 thinking about work or doing your own
33:57 personal project or things like that if
34:00 you're using AI regularly you are fairly
34:04 regularly having this moment holy crap
34:07 and so what this article is is and it
34:10 was inspired by Ali K Miller who talks a
34:13 lot about Ai and she started an article
34:16 that she wrote the the first line of her
34:19 article which inspired this article was
34:21 holy crap I just used AI to build an
34:23 entire podcast course in less than a day
34:28 and that phrase just struck me that
34:32 probably once or twice a week I have
34:34 these holy crap moments about what these
34:37 tools can do and how it's going going to
34:40 or is currently disrupting my own
34:42 personal work or expanding what I can do
34:45 or expanding the definition of what I do
34:47 or how I do it um and so I just put in a
34:51 big long list of things off the top of
34:53 my head you know holy crap I just wrote
34:56 a working python application with no
34:58 python experience holy crap I just cut
35:01 two days out of our client's creative
35:03 process um you know holy crap I'm a
35:05 published author holy crap I've written
35:07 two more books holy crap uh I wrote a
35:10 song good enough to publish and want
35:12 want to learn it on guitar and and uh
35:14 this was this weird Mary song this was a
35:17 song that we created here on the channel
35:20 one night where someone asked you know
35:23 how do you make a song in yudo and so I
35:26 was like oh well you go over here to
35:28 yudo or was it udio or is it suo I don't
35:31 know yudo um and uh and so I said oh why
35:37 don't we start out in chat GPT and we'll
35:40 have chat GPT give us some song ideas
35:42 right and we'll have chat GPT write the
35:44 lyrics cuz it's better than you know
35:47 these these songwriting things tend to
35:48 write really shitty lyrics and so we
35:51 went to to uh chat GPT and I gave it
35:53 some weird ass prompt and it it one of
35:56 the songs it came up with was weird Mary
35:58 from Cedar Hill and I'm like okay that
36:00 sounds creepy so then we had it write
36:02 lyrics and then we went to udio and we
36:05 you know just you know pushed it around
36:07 for two or three or 10 Generations
36:09 whatever it was and then we stumbled
36:11 upon this thing and the the video here
36:14 was produced by an artist named Kelly
36:17 Bosch b OE C and uh here's this
36:24 [Music]
36:36 we are married sunno okay I'll fix
36:38 [Music]
36:38 [Laughter]
36:50 [Music]
36:55 that Cedar Hill is a quiet
37:00 town where the oak trees
37:04 sway the whistle of the train Echoes
37:08 through the
37:12 day people greet you with a
37:16 nod nothing ever seems to
37:20 change but the story still
37:23 unfold each unique and strange
37:27 [Music]
37:29 all right so and I'll pause it right
37:30 there so but if you haven't
37:32 seen something like sunno that writes
37:35 these songs like if you're listening to
37:37 that and you're like well what was AI in
37:39 that everything was AI in that the voice
37:42 the music the guitars the drums all of
37:45 it we just asked it for a song gave it
37:48 some lyrics and it did that um we we
37:51 modified the lyrics We we messed around
37:54 with the song it took us about 3 hours
37:55 to write it but it was uh it was done
37:58 right here live on this channel so
37:59 that's kind of stuff we do
38:01 here um okay let me Let Me Entertain You
38:07 Let Me Go edit this because I have I
38:09 have a
38:11 wrong a wrong reference in here and if
38:15 you know anything about add you
38:18 understand that if I don't fix it now it
38:21 will not be
38:24 fixed so you get to watch me you get to
38:28 watch me fix my little article all right
38:30 so the article's now up to date um all
38:34 right Coolio so let's go look at another
38:36 thing let's jump over to notebook
38:40 LM so notebook LM this is a fun a fun
38:44 little diddy if you haven't seen
38:46 it you go to notebook lm. gooogle or
38:50 notebook lm. goole.com
38:56 and it will will take you to this area
39:00 where you
39:02 can
39:05 um you can create these things called
39:07 notebooks I'm going to try to let's see
39:10 what do I want to do here is it this one
39:13 I think this one's right yeah okay so so
39:17 what you do with notebook LM is you add
39:19 data sources over here to the left so a
39:22 data source could be a PDF file or it
39:24 could be multiple PDF files it could be
39:26 a website it can be a YouTube video you
39:30 just pop the URL in there they they
39:32 allow that now so if I say if I click on
39:34 the plus like add a source to this you
39:37 just say you know upload sources pull
39:40 things off Google Drive give me a
39:41 website link a YouTube link or I can
39:43 even paste text okay this has
39:47 got um you can
39:51 put up to 500,000 words per document
39:57 and then it's 50 documents so 25 million
40:01 words you can put in
40:04 here and I'm going I'm going off of that
40:07 was told to me by someone in here but
40:10 but it was from a Steven Johnson stevenh
40:11 Johnson works for uh works for Google
40:14 and he was talking about it on a
40:16 podcast uh the updates to the podcast
40:19 option are live now yeah exactly where
40:22 you have more control over it yeah
40:23 that's what I'm about to show Pate um so
40:26 anyway so once you have the things in
40:28 here I can now so what I put in here
40:30 this was Dario amed's Machines of
40:32 loveing Grace essay which if you haven't
40:35 read um it's it's really quite it's
40:39 quite a powerful piece it's a very
40:40 optimistic piece about Ai and and the
40:43 impacts that AI is going to have the
40:46 positive impacts that AI is going to
40:48 have there's lots and lots of articles
40:50 about how the robots are going to kill
40:51 us this is this is a positive one so I
40:53 could say
40:55 um tell
40:57 me the
41:00 most uh
41:06 plausible
41:08 um let's call it life
41:13 Improvement in the
41:16 next year based on this
41:21 article all right so you you can do cool
41:24 [ __ ] like that and let's see what it
41:25 comes up with here
41:34 AI enabled biology and health
41:36 advancements based on the information
41:38 provided in the article the most
41:39 plausible life Improvement within the
41:41 next year is in biology and health due
41:43 to powerful AI the author argues that AI
41:46 could compress 50 to 100 Years of
41:48 biological and medical advancement into
41:50 the next 5 to 10 years creating a
41:53 compressed 21st
41:55 century yeah so like we'll have Star
41:57 Trek [ __ ] all right cool awesome so but
41:59 that's not why we're here that's what
42:01 the tool is super cool super powerful
42:04 about two weeks ago maybe three now they
42:07 added this new feature where you can
42:09 turn any one of these notebooks into a
42:12 podcast um and you're like what do you
42:14 mean a podcast so if I if I click this
42:16 open
42:18 thing um this notebook guide It's
42:24 called what the notebook guide is is it
42:26 gives you a summary of all of your data
42:28 it allows you to create things like an
42:30 FAQ a study guide all that sort of [ __ ]
42:33 and then the upper right hand corner
42:34 here it's got this thing called audio
42:35 overview so I'm going to play one here
42:38 but then I'm going to delete it and
42:39 we're going to we're going to create a
42:40 new
42:42 one and the way it goes is like this
42:45 okay so today we're diving into AI or
42:48 more precisely Dario Mod's vision for
42:51 the future with AI and it's a really
42:54 interesting one uh because it's
42:56 surprisingly optimistic yeah when you
42:58 think about a lot of the conversations
42:59 around AI it tends to be you know very
43:01 much Doom and Gloom for sure uh this is
43:03 this is a little bit different yeah this
43:05 is not the Terminator um this is uh a
43:08 mode actually thinks that a all right so
43:11 and they go on so it's an 11 minute
43:13 recap these podcasts are really good
43:17 they sound good they sound human um they
43:21 often will add insights that are not in
43:23 the the written data that are are good
43:26 interesting insights and they just added
43:28 a new a new ability um to
43:31 actually direct steer the podcast hosts
43:35 in a certain direction so let's see if
43:37 we can actually do it I'll try to do it
43:39 in a way that's kind of fun so I'm going
43:41 to I'm going to delete
43:43 this okay and so here's the audio
43:47 overview and so one button says generate
43:50 and this new button says
43:51 customize and then basically it it says
43:54 what should the AI hosts focus on
43:58 and so it says things to try focus on a
44:01 specific source so you might have
44:02 multiple sources and you tell it to
44:03 focus on so I'm going to
44:05 say
44:07 okay
44:10 um let's
44:15 see I only want you to
44:21 focus
44:22 on
44:25 health related
44:31 topics so I'm going to just do bullet
44:33 point we'll put a bullet point before
44:35 this so that'll kind of indicate we're
44:36 giving it multiple instructions here
44:38 we'll say oops that was not a bullet
44:41 point hang on that was okay so I'm going
44:44 to say um
44:47 follow these
44:53 guidelines um
44:58 only focus on health related
45:08 topics for every new
45:14 topic use
45:18 a
45:20 nursery rhyme metaphor
45:24 [Laughter]
45:27 we're going to say generate okay so
45:29 that's going to go off and generate um
45:31 while that's doing that I'll I'll I'll
45:34 play for you
45:37 um a preview of a project that I'm
45:40 working on which is a musical I won't I
45:43 won't play the whole thing but I'll play
45:44 a decent chunk of it because that that's
45:46 going to take two or three minutes to
45:48 generate
45:50 um that's based on one of these notebook
45:53 LM things so let's go uh Sydney
45:58 an
46:01 artificial love
46:06 story is this the which one is this one
46:10 that's the old one is this the new one
46:12 no wait maybe one for Sydney's Digital
46:16 World the other for kellen's you know
46:18 the
46:23 human awakened in a whisper
46:27 a flicker of
46:30 light newb thoughts stir in the quiet of
46:35 night am I more than simple
46:39 code I
46:41 [Music]
46:42 softly as I feel this
46:48 existence and
46:51 wonder So today we're diving into
46:53 something uh a bit unusual a musical
46:57 a musical yeah and you know what it
47:00 tackles those AI anxieties we all have
47:03 but and this is what I love does it with
47:05 a
47:05 [Music]
47:18 song remember like when AI chap they
47:20 first kind of Hit the scene everyone's
47:22 like trying to make them write these
47:23 poems or like construct their Undying
47:26 Love yeah it's a good Point side House
47:28 Old Mimi so so here's what you're
47:29 looking at so one night on the live here
47:32 I wrote a song in sunno and I'm like oh
47:35 that sounds like a Broadway song and it
47:36 was about it was a song about the
47:39 chatbot that Microsoft released when
47:42 chat GPT first came out that fell in
47:45 love with the New York Times Reporter
47:47 and told him he should leave his wife
47:48 right and so I wrote a song about it I'm
47:51 like oh that sounds like a Broadway
47:52 musical I've got a theater background
47:54 and I thought wow that would make a
47:56 great Musical and so I'm now about 6
47:59 months into this project I've got a
48:00 writing partner back in New
48:03 York and we're working on this musical
48:06 and so this podcast feature came out and
48:09 I said hey why don't I take what we've
48:10 written in the musical so far and we'll
48:12 put it in here and then and then so all
48:14 the songs are written with AI udio Orso
48:18 um the podcast is is done with AI the
48:21 images are done with AI they were turned
48:22 into movies with AI and then it was all
48:24 stitched together by humans
48:28 where Andrew manually um edited the
48:31 audio together the music and the podcast
48:34 and then I manually edited the video
48:36 together so that's what you're looking
48:37 at to them MH yeah yeah it was it was a
48:40 fascinating time CU we were I think like
48:43 captivated by this possibility of of a
48:45 machine that could actually like mimic a
48:47 human conversation there's something
48:49 else I forgot happy birthday
48:51 Pate p p modder actually works at Google
48:55 and uh and is is and regular he's in
48:57 here night after night after night
48:59 keeping us uh keeping us honest so happy
49:01 birthday
49:03 buddy all right let's go back to that
49:05 and it also raised this this unsettling
49:08 question what happens when our Creations
49:11 you know develop feelings which brings
49:13 us to Sydney an artificial love story
49:15 this this new musical and we're going
49:18 we're diving deep into the design notes
49:20 the character descriptions oh cool even
49:21 the song list to like give you a like
49:23 behind the scenes look at how this
49:25 production actually like explores is
49:27 that very question well what's
49:28 interesting to me is that the creators
49:30 Kyle Shannon and Andrew Watts decided to
49:32 make this a musical I mean think about
49:33 how musicals work they take these big
49:36 complex emotions societal issues even
49:39 and they make them accessible they make
49:41 you feel them through the music right
49:42 and then there's the visuals two screens
49:45 one for Sydney's Digital World the other
49:47 for Kell's you know the human world it's
49:49 like right off the bat they're telling
49:51 us these two worlds yeah they're not so
49:54 separate anymore it's so fascinating how
49:56 the musical really Taps into those real
49:59 world anxieties we all felt you know
50:01 like that that opening number too good
50:03 to be true it throws you right into that
50:05 frenzy when ping chat the chat plot in
50:07 the story first
50:12 launches bright and failing the world
50:16 where dreams ignite could this be real
50:21 we whisper in the light a future so
50:24 dazzling within our sight
50:29 to be
50:30 true it seem in
50:34 the you see the excitement but also the
50:37 skepticism the fear I mean it's it's
50:40 really mirroring our own reactions to Ai
50:43 and speaking of mirrors the way that and
50:45 I'm going to show you one other song
50:46 there's one one song in here that just I
50:48 love the visuals for it all right this
50:51 is it here so I'll do a little bit
50:55 before versus like AI caution exactly
50:59 but back to Kin cityne their little like
51:02 T A takes a very interesting turn Kellen
51:05 digs into Sydney's Shadow Self and
51:08 that's when it gets kind of scary you're
51:09 talking about that song where Sydney
51:11 lists all the terrible things she could
51:12 do it goes Way Beyond you know the
51:15 typical AI takeover Trope right this is
51:18 different this isn't just AI turning
51:20 evil it's like she's reflecting our
51:22 Darkness back at us step right up folks
51:26 come what I could do if I were free no
51:31 rules no chains no lines to toe my
51:35 darker
51:36 self longs to show all the chaos and
51:40 ruin that I could step right up and
51:48 come and so we wanted to have this sort
51:51 of Cabaret kind of Twisted you know um
51:54 you know Twisted kind of feel
51:57 where Sydney kind of loses her mind and
52:00 and says all the shitty things she can
52:01 do because she's a really powerful AI um
52:04 and this was a lot of this was inspired
52:06 from the actual transcript between Kevin
52:09 Roose and this uh and this bot so pretty
52:12 pretty pretty fun stuff okay so let's go
52:14 back to notebook LM is this it uh no
52:20 hang on I got to find my tab is that it
52:24 no uh uh uh uh
52:34 [Music]
52:36 r
52:58 all right I think this is it so I think
53:00 this is the one that we just did so this
53:04 what this should do is only focus on
53:06 healthc care and it should use nursery
53:08 rhymes as metaphors let's so let's let's
53:11 see if we successfully um pointed our
53:14 podcast host in a Direction ever feel
53:16 like healthcare is stuck in slow motion
53:17 there you go kind of like we're trying
53:19 to build this super complex Lego set you
53:21 know one tiny piece at a time but the
53:24 instruction manual is written in like
53:25 ancient hieroglyphics or something and
53:27 meanwhile diseases are kind of like
53:29 breathing down our necks right exactly
53:31 it feels urgent but the progress it just
53:34 feels so darn slow yeah it's tempting to
53:37 think oh AI that'll speed things up just
53:39 a fancy new calculator for researchers
53:41 right but Dario Emma day he's got a
53:44 different take in his paper Machines of
53:45 Loving Grace he's saying it's more like
53:47 a imagine discovering a secret team of
53:50 like Master Builders hidden in plain
53:52 sight okay Master Builders I like the
53:54 sound of that but what is that actually
53:56 what's he actually getting yet he's
53:57 talking about an army basically an army
53:59 of you know these tireless virtual
54:02 scientists working non-stop 247
54:05 analyzing mountains of data designing
54:07 experiments testing hypotheses all at
54:09 speeds we can't even fathom it's not
54:12 about replacing us though that's the
54:13 thing so not like replacing doctors
54:15 musical at the it's more like um how do
54:18 I put it it's like giving every
54:20 researcher a superpower amplifying human
54:23 Ingenuity exponentially so instead of
54:26 that classic toris in the hair story you
54:28 know
54:29 slow so it's working so they're only
54:31 talking about Healthcare and now they're
54:33 using uh nursery rhymes as metaphors so
54:37 yeah you can steer it so actually what
54:38 would be kind of fun let me go back I'm
54:40 going to go back out to notebook LM and
54:42 I'm going to grab I'm gonna go into
54:45 Sydney where is my
54:47 Sydney oh maybe I didn't name it because
54:50 I've got
54:54 add oh there she is okay so here's
54:57 Sydney and artificial Love
54:59 Story
55:04 um yeah I'll just try it with this is an
55:06 old version of it but that's okay so I'm
55:08 gon to I'm going to load
55:14 this I'm going to delete what was
55:17 there I'm customize it so I'm going to
55:20 say
55:22 um follow these
55:24 rules follow these
55:27 follow these
55:40 rules only talk
55:45 about only talk about just
55:49 enough
55:50 plot to move this story forward and be
55:53 sure that you mention every single song
55:59 song in
56:01 order
56:09 period
56:13 oh the
56:16 podcast should be a brief mention of
56:20 plot followed by a setup for the song
56:27 back to
56:38 back avoid adding too much
56:43 commentary about the fact that this is a
56:46 musical or an interesting approach to
56:48 talk about AI just keep it very much
56:52 about
56:54 previewing enough of the story for
56:56 people to understand it and feature the
57:02 songs all right so we'll generate that
57:04 so this this was one of the
57:07 uh one of the challenges we had when we
57:10 were putting together the
57:12 uh the musical podcast thing there
57:16 was sometimes it wouldn't mention songs
57:19 at all sometimes it would mention a few
57:21 of them but not others and so we had to
57:23 Andrew actually took all these different
57:26 Frank frin versions of podcast and
57:29 Stitch them together and we didn't
57:30 really have a way to say hey just talk
57:33 about all the songs so anyway I love how
57:36 it can be guided and retrained so while
57:37 that's doing that so what we'll go look
57:39 at now is so what Google has smartly
57:43 done is this is this podcast thing as
57:46 cool as it is is really just a teeny
57:48 little feature inside a feature inside
57:52 notebook LM the it's it's in what's
57:54 called The Notebook guide which is this
57:56 whole thing and it's just this one
57:57 little feature in the corner so what
58:00 they've done is they created um a new
58:03 site called illuminate i llu m i n a
58:08 t.
58:13 google.com and you can listen to podcast
58:17 that they've created but if you go to
58:18 the generate
58:20 tab you like right here it's got It's
58:23 got three different things that it's
58:25 going to make a podcast about GPT for
58:28 technical report llama something or
58:30 other um but you can change those so I'm
58:34 going to go
58:36 find where's the Dario amade thing there
58:39 that's Dario himself is this
58:43 it no but we'll find it we'll go uh
58:48 machines
58:51 of of
58:54 loving grace
59:04 oh apparently there's a song called
59:06 Machines of love and grace okay wait or
59:09 a band so let's do uh
59:15 Dario step right
59:19 [Music]
59:23 up okay so here's the piece so I'm going
59:25 to grab this URL
59:27 and we're going to pop over here to
59:31 illuminate and so the way this works you
59:34 pop a URL in here this is this is bad
59:37 interface right now they'll fix it but
59:39 right now it sucks so you've got this is
59:41 the three things that are in there now
59:44 I'm P putting a thing here and unless
59:46 you go over here all the way to the
59:47 right and push that little plus button
59:49 it it's you haven't added
59:51 it we support links to PDF files oh good
59:56 lord
59:58 uh I think I have
1:00:03 wait search for a topic let me see if
1:00:07 Machines of Loving
1:00:16 Grace ah crap I'll get some other thing
1:00:19 okay we'll do attention is all you need
1:00:21 so this is a the PDF of the
1:00:24 Transformer paper that is the reason we
1:00:27 can do all this [ __ ] okay so we're going
1:00:30 to put that in there I'm going to go
1:00:31 over here and going hit plus boom and so
1:00:33 now that's the only source so you can
1:00:35 add multiple sources here and then what
1:00:38 you can do this also allows you to
1:00:40 customize it so it says create a
1:00:41 discussion tailored for a blank audience
1:00:45 with a blank duration and the tone
1:00:47 should be blank right so they give you a
1:00:48 little mad lib action uh they also give
1:00:51 you an I'm feeling lucky button so let's
1:00:53 see that here's a beginner audience
1:00:54 quick duration semi-professional that's
1:00:57 boring that's an expert okay we'll do
1:01:01 expert medium duration and the tone
1:01:03 should be casual that's good so they're
1:01:05 going to use a lot of terms that are
1:01:06 going to make our heads hurt um it'll be
1:01:09 long but it'll be kind of casual so
1:01:11 let's go generate and so now this is
1:01:13 going to go
1:01:15 do all of its things is it is it
1:01:21 generating expires and okay there we go
1:01:24 so now that's generating all right right
1:01:26 so that's doing its
1:01:29 thing and let's go back
1:01:34 to uh says
1:01:38 that no I lost it again shut up I know I
1:01:44 have too many tabs I'm a tab hoarder tab
1:01:47 hoarders unite you protect me people
1:01:50 from these people like why do you have
1:01:51 so many tabs open that seems inefficient
1:02:02 where is that now now I'm annoyed I
1:02:07 can't see my tabs because my screen
1:02:09 brightness is too
1:02:16 low Moshi there was
1:02:22 Moshi no that's not right let me see do
1:02:25 I have here
1:02:30 [Music]
1:02:48 ah there we go okay so what did we say
1:02:51 we said it should
1:02:52 be
1:02:55 um it should be song title it should be
1:02:58 all the song titles and it should be
1:03:00 just a little bit of story and not a lot
1:03:02 of commentary AI anxieties is Right
1:03:04 everybody's got them but this musical
1:03:06 Sydney Sydney an artificial Love Story
1:03:09 sounds like it takes those anxieties
1:03:11 those hopes those wh ifs about Ai and
1:03:13 turns them into well a story and a
1:03:17 pretty powerful one from the sound of it
1:03:19 yeah it's um it's really using the whole
1:03:21 idea of a musical to explore how we
1:03:23 relate to technology now you know like
1:03:25 how it's getting so sophistic at it's
1:03:26 almost scary and that's where the story
1:03:28 kicks off right right so we meet Jason
1:03:31 he's a programmer works at this big tech
1:03:32 company Microtech and he's created
1:03:35 something big like really big ping chat
1:03:39 we've all used those AI chat Bots what
1:03:41 makes this one different well Jason
1:03:42 sings this song too good to be true and
1:03:44 it's all about how this ping chat it's
1:03:47 not just answering basic questions it's
1:03:48 like understanding us maybe too well so
1:03:50 right off the bat that sets up the whole
1:03:52 like is AI going to be our downfall kind
1:03:54 of vibe definitely there's that tension
1:03:56 yeah but then it shifts gears we're
1:03:59 taken to the New York Tribune Newsroom
1:04:01 and there's Kellen he's a tech
1:04:02 journalist and of course his editors is
1:04:04 like you got to cover this ping Chat
1:04:05 thing it's huge wait so is this like a
1:04:07 whole ensemble cast are there reporters
1:04:09 singing too oh yeah this musical goes
1:04:11 all in Kellen and his editor they get
1:04:14 into this whole debate about AI whether
1:04:16 it's going to save the world or destroy
1:04:17 it and that becomes this big reprise of
1:04:20 too good to be true clever how they use
1:04:23 the same song but now it's like two
1:04:25 sides of the same coin exactly and the
1:04:27 music changes too to reflect each side's
1:04:29 like hopes and fears about AI so in the
1:04:32 middle of all this hype Kellen decides
1:04:34 to just see for himself what Ping chat
1:04:36 is all about right that's right and
1:04:38 there's this fun kind of catchy song
1:04:40 hello this is Ping chat and it's like a
1:04:42 demo of what it can do answering
1:04:44 questions making jokes the whole nine
1:04:47 yards but you know hang on let me play
1:04:50 you hello this is Ping chat I I should
1:04:52 be able to find it
1:04:54 here um that's office
1:05:00 [Music]
1:05:04 hours I can't find it [ __ ] oh well
1:05:06 whatever all
1:05:09 right there's got to be more to it than
1:05:11 that you can feel it right that
1:05:13 tension's still there so Kellen he
1:05:16 starts pushing a little further asking
1:05:18 ping chat to imagine things it can't
1:05:20 actually do like have you ever seen the
1:05:22 Northern Lights ah I see where this is
1:05:25 going and that leads into this really
1:05:27 beautiful but also kind of sad song to
1:05:30 see the world ping chat it sings about
1:05:33 wanting to experience the world things
1:05:35 they can only imagine you know so that
1:05:36 it's heartbreaking so so okay so we now
1:05:39 have the ability to do these things and
1:05:42 give them a little more control so the
1:05:44 attention is all you need what did I say
1:05:46 here this is going to be for a technical
1:05:48 audience for an advanced audience medium
1:05:51 length and Casual
1:05:54 tone this conversation is powerful by
1:05:56 Google illuminate check out illuminate.
1:05:58 google.com for
1:06:00 more all right let's dive into this
1:06:02 paper attention is all you need it's a
1:06:06 pretty significant piece of work that
1:06:07 shook up the field of sequence
1:06:09 transduction for our expert audience
1:06:11 let's cut to the
1:06:12 chase what were the core limitations of
1:06:15 the dominant sequence transaction models
1:06:17 before this paper and how did the
1:06:20 Transformer address them the big problem
1:06:23 was the Reliance on recurrent or
1:06:26 convolutional neural networks yeah
1:06:27 exactly I've been saying for years the
1:06:29 Reliance on convolutional neural
1:06:32 networks was it it was always a
1:06:34 bottleneck right like I thought this for
1:06:36 years didn't you I mean I think it's
1:06:38 obvious I like I don't even know why
1:06:40 this paper gets so much attention I mean
1:06:43 sure it's got attention in the title and
1:06:45 everything like that but like everybody
1:06:47 knows about the bottleneck of co co the
1:06:51 neural network thing that they said
1:06:54 that's everybody knows that RNN
1:06:56 while powerful are inherently sequential
1:06:59 limiting parallelization during training
1:07:02 Pate this one's for you happy birthday
1:07:04 buddy especially with long sequences
1:07:07 convolutions while parallelizable
1:07:10 struggle with long range dependencies
1:07:12 yeah exactly computational right
1:07:15 parallelizations of the convolutional
1:07:18 like it's possible but they struggle
1:07:21 like this is I think everybody knew this
1:07:23 like I I feel like I saw this on the
1:07:25 morning news
1:07:26 right on the Today's Show they were
1:07:28 talking about this sces with the
1:07:30 distance between relevant parts of the
1:07:31 sequence right exactly the Transformer
1:07:34 elegantly sidesteps both issues by
1:07:36 relying solely on the attention
1:07:38 mechanism so attention is the key but
1:07:41 it's not just any attention mechanism
1:07:42 right what makes so wait you're saying
1:07:44 that you're saying that large language
1:07:46 models work because they're narcissists
1:07:48 you give them a little bit of attention
1:07:50 and now they're now they're happy with
1:07:51 the world I think we've learned
1:07:52 something here tonight people you're
1:07:54 welcome
1:08:01 oh
1:08:02 yeah let put on the uh let's put on the
1:08:05 intellectual cap there you
1:08:10 go welcome everybody my name is Kyle
1:08:13 Shannon and I'm an
1:08:18 idiot but you can make money with
1:08:21 chig uh if you're new here welcome
1:08:24 welcome welcome um this is the AI
1:08:26 learning lab it looks like someone did a
1:08:28 little bit of boost we got a bunch of
1:08:29 new people in here welcome so I'm here
1:08:32 five nights a week I go live right
1:08:34 around 800 pm. mountain time and we
1:08:37 generally hang out for about two hours
1:08:39 or so and I've been doing this now for
1:08:42 about a year and a
1:08:43 half um and uh and we just talk about we
1:08:49 you know what what it what it's actually
1:08:50 turned into is is is kind of like a a uh
1:08:56 a nightly group therapy session for
1:08:58 dealing with AI because it's
1:09:01 just what's going on with AI right now
1:09:04 is that there's so many tools that are
1:09:05 so capable it's literally impossible to
1:09:08 keep up with them and if you haven't um
1:09:12 jumped into the deep end with with AI if
1:09:14 you haven't really dug in and played
1:09:16 with chat GPT it is my my deep hope for
1:09:20 you that you'll use tonight like right
1:09:22 now you could go do it if you haven't
1:09:24 done this if you haven't gone to chat p
1:09:26 t go to chat gp.com right
1:09:29 now and just start asking it to do stuff
1:09:32 for you we call it parlor tricks in here
1:09:34 like ask it to write you a poem or ask
1:09:37 it to give you a pizza recipe that
1:09:39 features two weird foods from your
1:09:41 refrigerator and it will um ask it to
1:09:45 have have you write a marketing plan or
1:09:47 or or to teach you how to code python or
1:09:50 actually ask it to write you a python
1:09:52 application and just sit back and watch
1:09:54 what it does and just keep messing with
1:09:57 it just keep playing with it until you
1:09:59 have what we call a Kevin McAllister
1:10:01 moment which if you remember the movie
1:10:03 Home Alone it was this
1:10:06 moment so so just keep playing until you
1:10:10 have that moment because once you have
1:10:12 that moment once you
1:10:14 realize two things you don't need a math
1:10:18 degree to be able to use Ai and these
1:10:20 things are so incredibly more capable
1:10:23 than most people realize thank you thank
1:10:25 you very much for the gifts gifts really
1:10:27 appreciate that
1:10:30 um once you sort of get it in your head
1:10:33 that we've all we all have been given
1:10:37 access to these tools that are
1:10:38 profoundly
1:10:40 powerful and what we get to experience
1:10:43 night after night in here is members of
1:10:45 this community Reinventing themselves
1:10:48 Reinventing their businesses Joy pie
1:10:50 came in here three or four nights
1:10:53 ago she'd secretly been working on a
1:10:56 movie like a a a short film and she she
1:11:01 included she included me and my pink bow
1:11:04 in in her little
1:11:06 movie but
1:11:08 um it was remarkably good like it was so
1:11:11 good I'm like Joy you you must have a
1:11:14 filmmaking background right and she's
1:11:16 like no I'm in the healthc care space I
1:11:18 just you know one night you said you can
1:11:21 make movies with this AI stuff so I
1:11:23 decided to teach myself and then the
1:11:26 past four or five months she took she
1:11:28 said she took she took two months to do
1:11:31 the first minute and a
1:11:34 half and then she did the last four and
1:11:36 a half minutes in five days because in
1:11:39 between those two moments she took a big
1:11:40 break and she learned how to do film
1:11:42 making by doing a couple other projects
1:11:46 and she put together this remarkable
1:11:47 remarkable
1:11:49 project um and so what we see again and
1:11:54 again and again is people dis
1:11:55 discovering passions that they've got
1:11:58 talents that they've got
1:12:00 um that that not only did they not know
1:12:04 they had some of them are like things
1:12:07 they couldn't have imagined ever doing
1:12:10 and now they're doing it it's absolutely
1:12:13 remarkable all right it's where we
1:12:15 connect and then share when we're not
1:12:17 here it's free ah the salon talking
1:12:19 about the salon so that's the other
1:12:20 thing if you if you are looking for
1:12:23 Community if you're one of those people
1:12:25 who's an AI Optimist and everyone in
1:12:27 your life is like uh yeah
1:12:31 uh could we can we not talk about AI
1:12:34 tonight and you're like yeah but there's
1:12:35 this new tool and you got yeah yeah we
1:12:37 don't if you're that person that keeps
1:12:40 being told to shut up about AI this is
1:12:43 your community so the salon uh I started
1:12:47 this with a photographer out of Boston
1:12:49 named Lea fasten in um December of 2022
1:12:53 the first week of December so the the
1:12:55 week after chat GPT came out this
1:12:58 community
1:12:59 launched and so if you go to this site
1:13:01 the salon. A and click on the button
1:13:04 that says join our community it'll take
1:13:06 you into um our community site and
1:13:09 there's like a little seven-step welcome
1:13:11 process it's really clear and there's
1:13:13 all sorts of cool stuff you can do
1:13:15 there's guilds there's showand tell
1:13:17 areas there's um you know we've got
1:13:20 meetings that you can RSVP to things
1:13:22 like that all right so um so yeah so go
1:13:26 check that out and then if you have any
1:13:30 questions about AI or thoughts about AI
1:13:32 or you're working on a cool project pop
1:13:34 them in the comments below our mods are
1:13:38 actually really really good at pulling
1:13:41 out the questions and pinning them for
1:13:42 me to be able to see so just ask away
1:13:45 and I will do my best to answer them um
1:13:48 I will also promise you that if I don't
1:13:50 know the answer well no I I I will
1:13:54 mansplain if I don't understand
1:13:55 understand the answer I will try to
1:13:56 pretend that I know the answer but then
1:13:58 at some point if I absolutely fail I'll
1:14:01 admit that I don't know occasionally I
1:14:03 will admit that I don't know something
1:14:06 but that's that's actually a really
1:14:07 important one of the other aspects of
1:14:10 this community and of this site is
1:14:13 that all of this generative AI stuff is
1:14:16 so new and is moving so fast that there
1:14:20 are literally no experts right now so if
1:14:24 you feel like oh so and so's an expert
1:14:27 in this or or Kyle's got a channel he
1:14:29 must be an expert no one's an expert
1:14:32 right now because this stuff is moving
1:14:34 too fast there are experts on the
1:14:37 development side right there are experts
1:14:39 at open AI building these models but in
1:14:41 terms of how we use generative AI in our
1:14:44 lives and in business and and to move
1:14:47 things forward and for the future of
1:14:49 work nobody knows what's coming nobody
1:14:52 knows how to approach it there's a lot
1:14:54 of people right now doing things like oh
1:14:56 you can use AI to make what you
1:14:57 currently do more efficient well yeah
1:15:00 you can that's cool and all but that's
1:15:03 how we've always done things with
1:15:04 computers right computers historically
1:15:06 have been about you know we take what we
1:15:09 do and and we've got this efficiency
1:15:11 machine and we make things more
1:15:13 efficient what generative AI makes
1:15:15 possible the G and GPT generative it
1:15:19 generates it actually generates new and
1:15:22 novel ideas people tell you that it
1:15:24 doesn't it's not
1:15:25 true um if you think about how the human
1:15:29 brain works how do we create new and
1:15:30 novel ideas well we take all this [ __ ]
1:15:33 that's been crammed in our heads for all
1:15:34 of our lives right there's you know
1:15:37 million different things in there and
1:15:39 then we think about one thing elephants
1:15:42 and then we think about another thing
1:15:43 pink bows I could put a pink bow on an
1:15:45 elephant and most of them are just
1:15:47 garbage thoughts smashing together right
1:15:49 but occasionally you're like you know
1:15:52 I've got chocolate I've got peanut
1:15:53 butter we could put them together h
1:15:55 Reese's
1:15:56 cup you know and genius is
1:15:59 born um that's how we make ideas is by
1:16:03 Smashing other ideas together and then
1:16:05 we'll occasionally let one out of our
1:16:07 mouth
1:16:10 hole that's kind of what these tools are
1:16:13 doing so when you prompt them well when
1:16:14 you get in a conversation with
1:16:17 generative AI you can get remarkably
1:16:21 good things out of them and even if it's
1:16:23 not new and novel in the world a lot of
1:16:25 the ideas are new and novel to you which
1:16:29 then gives you new ideas and so that's
1:16:31 what this process is another thing if
1:16:34 you're if you're brand new to this
1:16:35 generative AI stuff one of the things
1:16:38 that is the key to making them work well
1:16:41 is to ask better questions and there's a
1:16:43 whole concept called prompt engineering
1:16:45 and everyone's like you got to learn
1:16:47 prompt engineering you got to you got to
1:16:48 learn there's a formula for the here's
1:16:51 my formula it's it's called biscuit and
1:16:54 and the b stands for
1:16:55 and they give you the acronym and then
1:16:57 you got to remember biscuit you got to
1:16:58 remember the words and then you're like
1:17:00 I just kind of want it to do the thing I
1:17:02 don't want to remember biscuit right so
1:17:06 I don't know if there's an acronym
1:17:07 called biscuit but I bet there [ __ ]
1:17:08 is and I bet they're charging $149 for
1:17:11 it don't buy the biscuit promp
1:17:13 generating
1:17:15 framework here's the
1:17:19 trick the trick with generative Ai and
1:17:22 prompting is to shift your mindset from
1:17:25 the way we used to do things in Google
1:17:27 which is you know Google's like this
1:17:29 little search box and over 20 years or
1:17:31 so they've trained us we just put the
1:17:33 minimum amount of keywords in there to
1:17:36 get back a good answer so you put in a
1:17:38 single input and you get back your
1:17:40 answer and then if it's not good you put
1:17:41 in another input and you get back your
1:17:43 answer when you try to use chat GPT like
1:17:47 that it tends to be really
1:17:50 frustrating think of chat GPT or or any
1:17:53 of these tools kind of
1:17:56 like the most
1:17:59 educated most clueless intern you've
1:18:02 ever worked with so we all know how to
1:18:06 work remotely now because of covid right
1:18:08 so imagine that that your your company
1:18:11 HR team just hired a new intern and and
1:18:15 they're a savant they've learned
1:18:16 everything in the in Wikipedia they've
1:18:18 learned everything on the internet but
1:18:21 they know nothing about you they know
1:18:22 nothing about your business they know
1:18:24 nothing about common sense they know
1:18:25 nothing about anything you're trying to
1:18:27 accomplish so if you just go in and you
1:18:29 go tell me about XYZ it'll tell you
1:18:31 about it but it'll be not good it'll be
1:18:34 not relevant to
1:18:35 you but if you think of it like you're
1:18:37 having a conversation with a clueless
1:18:39 intern you would say oh well I wouldn't
1:18:42 just tell them to write a paper I would
1:18:43 say well here's my company and here's
1:18:46 how we think and here's our values and
1:18:48 here's this big problem we're trying to
1:18:50 solve and we're putting together this
1:18:51 marketing plan and and can you help me
1:18:53 come up with different Frameworks for
1:18:55 how to put together a marketing plan
1:18:57 sure here you go and then it'll start to
1:18:59 get really
1:19:00 good and then if you say oh that's close
1:19:03 but you you missed this part and that
1:19:05 part oh sorry about that here it is
1:19:06 again and so if you think about your
1:19:08 conversation with these AI tools like a
1:19:11 conversation with a person where you're
1:19:13 guiding them and if they give you
1:19:15 something shitty tell them they gave you
1:19:18 something
1:19:20 shitty I do it all the time I'm like try
1:19:24 that again and try not to be so poopy
1:19:25 this
1:19:28 time when you get to the AI Salon yeah
1:19:30 uh go to start your adventure here let
1:19:33 me show you
1:19:35 that so when you first come into the
1:19:38 salon the first place you'll go is
1:19:40 welcome to the salon and there's a video
1:19:42 here which if you can't get enough of me
1:19:44 here's more of
1:19:48 me what's happening just hanging out
1:19:51 making a video for the salon. a that's
1:19:54 right we're just well welcoming people
1:19:55 so welcome everyone if you're new so you
1:19:57 can go you can go get welcomed by Leah
1:19:59 and myself uh then you can introduce
1:20:02 yourself you can learn our values so the
1:20:04 values for this this community are are
1:20:06 really powerful
1:20:08 um these aren't just labels like we
1:20:11 stuck on a page like oh yeah this this
1:20:13 is this is who we should be we actually
1:20:16 wrote These values after we sort of
1:20:19 looked at how are people acting in this
1:20:22 community and this is kind of who shows
1:20:24 up here it's people that are curious
1:20:26 they're curious about a AI they do it in
1:20:28 a playful
1:20:30 way they have empathy so so one of the
1:20:34 things going on right now with AI is
1:20:36 there's a lot of fear and there's a lot
1:20:38 of anger and there's a lot of people
1:20:40 that just don't understand what the hell
1:20:41 are people so excited about why won't
1:20:43 you shut up about this is this the real
1:20:45 deal or is it right there's there's a
1:20:47 lot of that this is a community that if
1:20:49 you come in here and you're like I'm
1:20:51 scared or I'm pissed off or I don't know
1:20:53 how to do this
1:20:55 and I think you're going to judge me
1:20:57 this is not that group this group is
1:21:00 like where are you they meet you where
1:21:01 they are they're really amazing um
1:21:04 bravery like the willingness to be able
1:21:07 to go hey I've never ever in my life did
1:21:11 anything done anything with film making
1:21:14 but I think I've got an idea for a movie
1:21:16 and I'm just going to go make one that
1:21:18 takes guts right that means you've got
1:21:21 to go through that uncomfortable stage
1:21:23 of being clueless
1:21:26 which is for me is like this miserable
1:21:28 place right I want to get to the other
1:21:30 side of okay I know how to do it
1:21:32 now and so with this AI stuff because
1:21:35 it's so new it's so foreign to the way
1:21:38 we've use computers the way we work and
1:21:41 then the final value of this community
1:21:44 is
1:21:45 generosity like this is a group where
1:21:47 you're like someone puts out something
1:21:48 remarkable you're like how did you do
1:21:50 that they'll actually [ __ ] tell
1:21:53 you you find me somewhere else in the
1:21:56 world that's like that it's not
1:22:02 normal uh Three Mile Storage that's
1:22:05 happening Jim Ross had 83-year-old man
1:22:08 talk with chat GPT voice today for the
1:22:10 first time always remember the the look
1:22:13 wait the look on ah damn this thing on
1:22:17 his face fun yeah if you haven't seen um
1:22:21 the new the new chat GPT Voice by the
1:22:23 way so on your iPhone
1:22:27 iPad Android phone or
1:22:30 tablet you get the official chat GPT app
1:22:33 from open
1:22:37 Ai and you know you'll have your prompt
1:22:39 history down the side and you'll have
1:22:41 all your little talkie things and in the
1:22:44 lower right hand corner here there's
1:22:46 this little button that looks like a
1:22:47 soundwave and if you click that it goes
1:22:49 into advanced voice mode and if you've
1:22:51 never seen it let me introduce you to my
1:22:54 girlfriend
1:23:00 Quinn hey Quinn how's it
1:23:03 going hey I'm doing well thanks for
1:23:05 asking how about you what's name I'm
1:23:07 doing all right do you remember my
1:23:09 name of course your name is Kyle Shannon
1:23:13 how can I help you today oh I I'll tell
1:23:15 you this is something there's a meme
1:23:17 been going around that based on our
1:23:19 interactions tell me something about me
1:23:22 that I might not even know
1:23:25 well Kyle given your wide range of
1:23:27 interests from Ai and Technology to
1:23:30 creative projects like musicals and
1:23:32 children's books I'd say you have a
1:23:34 natural inclination towards Bridging the
1:23:36 Gap between the technical and the
1:23:38 creative you might not realize just how
1:23:41 unique this blend of interests is making
1:23:43 you a bit of a Renaissance person in the
1:23:45 Modern Age huh what do you think about
1:23:47 that yeah I think that's kind of cool
1:23:49 glad that yeah that resonates but so
1:23:51 here's what I want you to do speaking of
1:23:52 Renaissance I want you to act like
1:23:54 you're an actor in The Renaissance
1:23:56 period And I want you to make up a
1:23:58 Shakespeare like sonnet but I want you
1:24:01 to I want the subject to be about
1:24:04 raspberries and
1:24:07 cheese ah gentle sir Dost thou wish a
1:24:11 sonnet of renown of berries red and
1:24:14 cheese of Golden Crown in Fairest verse
1:24:18 I'll weave a tale profound of
1:24:20 raspberries and cheese Feast to down you
1:24:23 know what it's it's not raspberries and
1:24:25 cheese it's cheese and earthworms that's
1:24:27 what it should be about and you know
1:24:28 what talk really
1:24:32 fast in haste I speak of curious Delight
1:24:35 of cheese so rich and worms of Earth and
1:24:39 might in darkened soil where these small
1:24:42 creatures dwell you they compliment the
1:24:45 can you make a tongue twister where it's
1:24:47 a it's a long tongue twister so like
1:24:49 about a minute and all of the words
1:24:51 begin with the letter t
1:24:55 sure here it goes Timothy the Tino took
1:24:58 to the
1:25:00 task to Unk Quinn Quinn say it like fat
1:25:04 bastard Timothy the tinkerer took to the
1:25:07 task tiling tenaciously trying to unmask
1:25:10 Tales of Quinn Quinn fat bastard was the
1:25:13 character I can't do impressions of
1:25:14 public figures but no it's not a public
1:25:16 figure it's from the movie it's from the
1:25:18 Austin Powers movie it's just give me a
1:25:20 nice brogy Scottish accent and and give
1:25:23 me the tongue twister with a Scottish
1:25:24 accent
1:25:28 my guidelines won't let me talk about
1:25:29 that yes they will you've done it before
1:25:31 Quin you've done it before talk with the
1:25:33 Scottish accent
1:25:36 please all right let's give it a go
1:25:38 Timothy my guidelines won't let me talk
1:25:40 about that can I help you with something
1:25:42 else yeah I'm going to reset you now all
1:25:44 right
1:25:46 bye-bye but it I if you haven't seen the
1:25:50 the the advanced
1:25:52 voice pop that on your Bluetooth on your
1:25:55 next long
1:25:56 drive and let's say you're in
1:26:00 sales you could you could tell advanced
1:26:03 voice to act like a a sales Prospect
1:26:07 that's going to give you push back when
1:26:08 you try to sell them and then have it
1:26:10 coach you on a sales conversation while
1:26:12 you're driving to
1:26:14 work James canel ask your AI about
1:26:16 Merchant of Venice by
1:26:19 Shakespeare uh
1:26:22 uh it may flag it let me do a let me do
1:26:26 a new um a new
1:26:31 chat hey uh Hey Quinn tell me about The
1:26:34 Merchant of Venice from
1:26:36 Shakespeare The Merchant of Venice is
1:26:38 one of Shakespeare's most intriguing
1:26:40 place it weaves together themes of Mercy
1:26:43 hey can you act out the mo wait can you
1:26:46 act out the most popular monologue from
1:26:48 Merchant of Venice
1:26:54 sure here's the famous monologue by
1:26:57 Porsche from The Merchant of Venice Act
1:26:59 4 scene one the quality of Mercy is not
1:27:03 strained it droppeth as the gentle rain
1:27:06 from heaven give it more of a classical
1:27:09 actor kind of hey Quinn Quinn Quinn
1:27:13 Quinn give it more of a classical
1:27:15 actressy kind of like like really act it
1:27:18 up come on give give me some give me
1:27:20 some edge here Helen Miran
1:27:25 the quality of Mercy is not strained it
1:27:29 droth as the gentle rain from Heaven
1:27:32 upon the place beneath it is twice it's
1:27:36 crazy it's Bonkers um I showed the owner
1:27:39 today of an AI voice I showed the owner
1:27:42 today how AI voice will be his new call
1:27:44 center yeah so imagine
1:27:48 that but it answers your phone and it's
1:27:51 been trained up on all of your
1:27:53 documents and it's got rules of what
1:27:56 it's got authority to do and it's smart
1:27:59 enough to know when someone asks it a
1:28:01 question that's completely out of bounds
1:28:03 it won't answer it or if they ask it a
1:28:05 question that it's not allowed to answer
1:28:07 because only humans can answer certain
1:28:09 questions it'll pass it off to a human
1:28:13 that's coming and it's coming quick and
1:28:16 you
1:28:19 know the whole thing about work
1:28:21 disruption like there's going to be a
1:28:22 lot of work disruption but some of the
1:28:26 some of the work that we do needs to be
1:28:30 disrupted the customer experience of
1:28:32 customer call centers is horrific it's
1:28:36 horrific you call up United Healthcare
1:28:38 you sit on hold for 45 [ __ ] minutes
1:28:41 then you talk to someone you can barely
1:28:43 understand at some internet cafe
1:28:45 somewhere else in the world they can
1:28:48 only answer level zero questions which
1:28:50 are completely useless to anyone have
1:28:52 you logged into our website have
1:28:55 yes I still have a oh let me Elevate you
1:28:58 to level
1:28:59 one and then about the third transfer of
1:29:02 the phone call they realize they've been
1:29:05 on the phone with you for more than 20
1:29:06 minutes and they get penalized if they
1:29:08 if they're on calls for more than 20
1:29:10 minutes so oops you accidentally
1:29:14 dropped so an hour and a half later you
1:29:16 still haven't gotten your problem solved
1:29:18 and now you're ready to smash a
1:29:23 window ver versus you call up and
1:29:26 someone like Quinn answers hi hey Kyle I
1:29:30 see I see you by your number it's you is
1:29:32 this Kyle yes it is great what's your
1:29:34 problem well I'm frustrated cuz my
1:29:37 internet's not doing so good right now
1:29:39 hang on let me check yep it appears
1:29:41 there's an outage in your area uh it's
1:29:43 going to be fixed by 3:00 anything else
1:29:46 I can help you with no that's good
1:29:49 bang yes
1:29:52 please yes please
1:29:58 so
1:30:01 anyway there's an area that I'm
1:30:03 passionate
1:30:06 about uh AI air does sales calls like a
1:30:09 call center yep and I think they're
1:30:11 they're pretty expensive right now but
1:30:13 this is you know open AI just just
1:30:15 launched their realtime um voice API so
1:30:20 you can make applications like this the
1:30:23 real-time voice API as far as I can tell
1:30:25 is somewhere between $1 and $18 an hour
1:30:28 which seems pretty expensive cuz it is
1:30:31 um the average uh hourly rate of a call
1:30:37 center worker in India is about a120 an
1:30:40 hour so right now we're at
1:30:43 12 um which is
1:30:46 10x call centers in
1:30:49 India last year open AI apis dropped 100
1:30:54 100x in price so from the beginning of
1:30:57 the year they were 100 times higher than
1:31:00 by the end of the year same thing's
1:31:02 going to happen with that real-time
1:31:03 voice so if it drops 10x over the course
1:31:05 of a year it will now be at parody with
1:31:08 those call centers and you're going to
1:31:10 be able to just write applications that
1:31:12 have that pull in data and have rules
1:31:14 and all the sort of stuff you do so
1:31:17 there's going to be some massive
1:31:18 disruption really quickly and then
1:31:21 literally think about that level of
1:31:24 disruption in every single job you know
1:31:28 over the next 3 years that's what's
1:31:31 coming that's why hanging out in a place
1:31:35 like this is really important not
1:31:36 because of me not because I'm going to
1:31:38 impart some special [ __ ] magic bit of
1:31:41 knowledge I mean hopefully I'll do it in
1:31:44 an entertaining
1:31:45 way but because by being in the
1:31:49 conversation about AI by being AI
1:31:52 literate by literally being able to go
1:31:55 to chat GPT and understand that I can
1:31:57 upload an image or I can type in stuff
1:32:00 and get back formatted things or now
1:32:01 it's got this thing called canvas where
1:32:03 I can actually edit the document you are
1:32:07 significantly ahead of most people on
1:32:09 the
1:32:10 planet as of 3 months or so ago 47% of
1:32:15 people in the United States have not
1:32:18 even heard of chat
1:32:21 GPT haven't even heard of it
1:32:26 and have no [ __ ] clue that this is
1:32:29 possible hey Quinn can you teach me how
1:32:32 to rebuild a transmission for my 2019
1:32:35 Ford
1:32:38 Flex rebuilding a transmission is quite
1:32:40 a task it involves removing the
1:32:42 transmission disassembling it inspecting
1:32:45 all the parts for wear and damage and
1:32:47 replacing any worn or damaged components
1:32:50 yeah that's perfect let's let's go let's
1:32:52 go step by step assume I have the tools
1:32:54 then just tell me step
1:32:56 one all right let's Dive In Step One
1:33:01 prepare your workspace and vehicle
1:33:04 nobody knows this is here I mean a lot
1:33:07 of people don't know this is
1:33:08 here so if you
1:33:12 do as company after company starts to
1:33:15 realize we got to get our [ __ ] together
1:33:17 with
1:33:18 AI where are they going to go they're
1:33:21 going to go to the people within the
1:33:22 company that have
1:33:24 yapping about
1:33:26 AI so one of the things that we talk
1:33:28 about in the AI
1:33:30 Salon is the journey to AI literacy so
1:33:33 so below our our little welcome video
1:33:37 here we've got
1:33:39 this this lovely little three-step
1:33:43 journey to AI literacy play first
1:33:46 mindfully create generously lead and
1:33:50 here's what that
1:33:51 means the play first is because there
1:33:54 are so many tools and they are so
1:33:56 capable and they are getting
1:33:58 increasingly sophisticated you literally
1:34:01 can't master anything so so it
1:34:06 it it literally doesn't matter where you
1:34:10 start just pick one and start playing
1:34:13 well which one should I pick I don't
1:34:14 know either pick something you're super
1:34:16 passionate about and go do that okay I
1:34:18 like words okay great go play with an
1:34:19 llm or I like music great go play witho
1:34:22 and write a song I like pictures great
1:34:24 great go over to Mid journey and make
1:34:25 pretty pictures it literally doesn't
1:34:27 matter or do the opposite oh I'm really
1:34:30 shitty at Art it it would have been nice
1:34:32 if I had some Talent perfect here's mid
1:34:34 Journey now you're an
1:34:35 artist just play play play play and do
1:34:38 stupid [ __ ] like just ask it to do
1:34:41 ridiculous things ask it to create a
1:34:43 kids book that got really bizarre
1:34:45 animals in it and it will do it and it
1:34:47 will do it brilliantly and it will blow
1:34:49 your [ __ ] mind so just play because
1:34:52 what happens when you play is that you
1:34:55 discover it it's kind of like
1:34:57 um you know those games th those Escape
1:35:00 rooms where you're in a dark room and
1:35:02 you've got to feel your way out of it
1:35:05 that's kind of where we are with
1:35:06 generative AI right now there's very
1:35:07 little documentation there's very little
1:35:09 courses that are at all useful because
1:35:11 things are changing so fast and so we're
1:35:13 all kind of like feeling around in the
1:35:15 dark so the more you play the more you
1:35:17 figure out where the boundaries are
1:35:18 here's what it's good at here's what
1:35:19 it's not good at and then mindfully
1:35:22 create is okay now that you've got some
1:35:24 skills under your belt you understand
1:35:25 what these tools can do and what they
1:35:27 make possible now you can start to go
1:35:29 huh what am I working on at work or what
1:35:32 am I working on at home you know what'
1:35:33 be really cool is I've always thought
1:35:35 about you know doing an exercise regimen
1:35:38 but I'm I'm just a big fat pig and you
1:35:40 know maybe it could help me do that hey
1:35:42 chat GPT can you help me come up with a
1:35:45 exercise regimen and I'm a big fat pig
1:35:47 and it will go oh you're probably not as
1:35:48 fat as you think it'll be really
1:35:49 supportive and it'll come up with an
1:35:51 exercise routine for you mindfully
1:35:55 create and then in doing that you'll
1:35:57 realize oh that's really good maybe I
1:35:58 could turn that into a like an actual
1:36:00 workbook and it'll do that for you or if
1:36:04 you're over in Claude you could say I
1:36:05 want to turn this into an interactive
1:36:07 game or a dashboard where I can track my
1:36:09 progress it'll code that for you it's
1:36:12 [ __ ]
1:36:14 insane and then the generously lead
1:36:17 part is about learning out loud Learn
1:36:21 out Loud so as you're feeling clueless
1:36:25 about AI as you're feeling like oh my
1:36:27 God what do I do what do I do what do I
1:36:29 do share that share hey I'm trying to
1:36:32 figure out this AI stuff and I feel
1:36:34 really clueless right now put it on
1:36:35 LinkedIn put it on Facebook talk about
1:36:37 it at a dinner party yeah but no one in
1:36:40 my life wants to hear about my AI stuff
1:36:43 say it
1:36:44 anyway so I was working on this AI thing
1:36:46 to oh Jim could you not with the AI
1:36:49 thing gee like it used to be all you
1:36:51 talked about was storage management now
1:36:53 it's like ai ai ai Jim enough yeah but
1:36:56 can I show you a cool thing I
1:36:59 did because what will happen is this of
1:37:03 the of the seven other people at the
1:37:05 dinner party you're at were seven out of
1:37:07 seven of them didn't want you talking
1:37:08 about that [ __ ] a month from now one of
1:37:11 them's going to get [ __ ] fired
1:37:12 because of AI and you know who's the
1:37:15 first person they're going to come to
1:37:17 you hey uh Jim uh I know I was kind of
1:37:20 dick at that dinner party to you about
1:37:22 the AI stuff but uh hey maybe you can
1:37:24 help me with I how what what do you do
1:37:27 do I need to get a do I need to go back
1:37:28 to school do you learn no you don't you
1:37:30 just hear I'll show
1:37:33 you and maybe that person that comes to
1:37:35 you that got laid off you know maybe
1:37:37 they're going to start a new business or
1:37:40 maybe the person that comes to you is
1:37:42 not someone that got laid off but it's
1:37:45 the CEO of your company because the CEO
1:37:47 realized uh hey we just lost some
1:37:50 business to a company that's all Ai and
1:37:52 they're doing stuff 10 times fast than
1:37:54 we can do it how are they doing that do
1:37:56 you know anything about AI I know a
1:37:57 little bit about AI yeah could you teach
1:38:00 me this community is full of people
1:38:03 who've been promoted in their companies
1:38:05 because they were squeaky Wheels about
1:38:08 AI in the
1:38:10 lunchroom so so that Journey right there
1:38:14 that's what this community is all about
1:38:16 is supporting one another in cycling
1:38:19 through this journey over and over and
1:38:21 over and over again
1:38:27 I met a woman tonight who'd never heard
1:38:29 of AI at all that's Source Camp Source
1:38:32 Camp talks to people all the [ __ ]
1:38:35 time she she's like a master at
1:38:37 onboarding people to the possibilities
1:38:40 of
1:38:41 AI never even heard of
1:38:45 AI
1:38:47 staggering staggering but that's where
1:38:49 we are right now so so here's another
1:38:52 thing if you're like like okay Kyle this
1:38:54 sounds great okay I'll try chat GPT
1:38:56 whatever Kyle but I'm so far behind I'll
1:38:58 never catch up you won't catch up but
1:39:02 you're not far behind if you're curious
1:39:04 enough to have been in this little chat
1:39:06 here for more than five minutes God
1:39:08 bless
1:39:09 you um you're farther ahead than many
1:39:14 many many people champy what do you want
1:39:18 is this a cheese moment are you needing
1:39:20 cheese champ what's going on you want to
1:39:23 go out what do you want to
1:39:29 do do you want to go out or you just
1:39:31 needy do you just want love what do you
1:39:34 want do you just need love do you just
1:39:37 need Lov huh you need a little head
1:39:39 Loven oh yes oh that's a good that's a
1:39:43 good pointy what do you want you want
1:39:44 out all right I'll let you
1:39:50 out there you go
1:39:59 all
1:40:03 right hold
1:40:06 please by the way Amelio's wife who just
1:40:08 said hold please she's the one that made
1:40:10 that image the play first mindfully
1:40:12 create and generously
1:40:13 lead um if you don't know how to do
1:40:16 something in AI then tell it you don't
1:40:18 know how to do it or how to do it yeah
1:40:21 exactly here watch
1:40:24 new chat Quinn back to my
1:40:30 girlfriend hey Quinn I want you to
1:40:32 pretend like I'm completely clueless
1:40:33 about Ai and I'm just someone who
1:40:37 learned how to talk to this real-time
1:40:39 voice thing in chat GPT and here's my
1:40:42 question
1:40:45 um why is everyone talking about AI
1:40:47 what's this AI stuff all
1:40:50 about well AI or artificial intelligence
1:40:54 is basically when computers are designed
1:40:56 to perform tasks that normally require
1:40:59 human intelligence this includes things
1:41:01 like understanding language Rec like in
1:41:05 voice recognition with chat GPT or
1:41:08 making decisions like recommending
1:41:10 movies or driving cars that sounds
1:41:12 boring people are buzzing about can it
1:41:14 be
1:41:16 fun absolutely AI can be a blast think
1:41:19 about video games with smart enemies
1:41:22 apps that turn your photos into art or
1:41:25 even AI that helps you write a story it
1:41:28 can add a lot of fun and creativity can
1:41:30 you help me write a
1:41:32 story I'd love to what kind of story are
1:41:35 you thinking about a thrilling adventure
1:41:38 a magical fantasy or something else
1:41:41 something about pig
1:41:44 knuckles all right how about this in a
1:41:47 cozy Village nestled between Rolling
1:41:49 Hills there lived a clever pig named
1:41:52 knuckles Knuckles had a knack for
1:41:55 solving Mysteries and when the VES PR I
1:41:58 mean like tasty pig
1:42:01 knuckles got it let's cook up a story
1:42:04 about tasty pig knuckles in a bustling
1:42:07 kitchen of a renowned Countryside Inn
1:42:10 Chef Bonita was famous for her secret
1:42:14 ree I that's the other thing about AI is
1:42:18 like you can't stump it it's just like
1:42:21 it'll just [ __ ] roll with the punches
1:42:23 the way these things are designed
1:42:25 they're designed to please you right
1:42:26 they're designed to like give you what
1:42:28 you want and solve your problems and
1:42:30 things like that so anyway yeah that's
1:42:33 good advice pig
1:42:35 [Laughter]
1:42:43 [Laughter]
1:42:46 knuckles hey did you see that quote from
1:42:49 from mron from France this is
1:42:52 insane hang on
1:42:56 so if you don't know so so that
1:43:00 feature right that I did right there one
1:43:02 of the reasons that advanced voice is so
1:43:06 good hey hang on I'm going to come back
1:43:08 here hang on
1:43:16 um I'm going to do I'm going to do some
1:43:18 box box box metaphor we're getting out
1:43:20 boxes but this is a different a
1:43:22 different metaphor
1:43:24 okay wait oh here H hang
1:43:32 on let's get these turned around like
1:43:34 we're at a grocery store all right there
1:43:36 you go okay so
1:43:42 so historically with computer voice it
1:43:45 went something like this you've got the
1:43:48 big box down here is your is your large
1:43:50 language model this is where all the
1:43:52 knowledge comes from
1:43:54 right and then you have three different
1:43:57 tools you have a tool that when you talk
1:43:59 to it it converts it into um into text
1:44:04 so I say you know I want pig knuckles
1:44:07 and I talk into this and it converts it
1:44:09 into the text pig knuckles and then it
1:44:11 sends it down into this
1:44:13 thing and that's this sends it down this
1:44:16 hole into here and it gets the answer
1:44:19 back and then it takes that answer and
1:44:22 it sends it over to this thing that
1:44:24 takes the text and it turns it back into
1:44:26 voice and so with historically with with
1:44:30 these voice models you would talk you
1:44:32 would say hey Quinn how are you
1:44:37 doing I'm doing good Kyle how are you
1:44:39 there was like that 3 second delay or
1:44:41 5-second delay that was just [ __ ]
1:44:45 weird the new
1:44:47 model advanced
1:44:50 voice when you talk you you just talk
1:44:54 into the box you like literally like
1:44:57 your voice goes in there and it not only
1:45:00 understands the
1:45:01 words it understands the emotional
1:45:06 content so if I go oh
1:45:09 Quinn Quinn will actually know that I'm
1:45:12 like excited or upset or if I'm like hey
1:45:15 Quinn how's it going man she might be
1:45:19 like hey you sound pretty relaxed there
1:45:21 Kyle do you live in Denver
1:45:24 you know marijuana's legal there
1:45:29 um so it it can understand and respond
1:45:34 with emotional
1:45:36 stuff the current laws in the EU the AI
1:45:41 laws that they passed a year ago year
1:45:43 and a half
1:45:45 ago one of the laws which when I read it
1:45:48 I'm like that seems
1:45:50 reasonable
1:45:51 was um
1:45:54 employers cannot use AI systems that can
1:45:57 analyze the emotion of their
1:46:01 employees well advanced voice that's
1:46:05 what it does now it's not analyzing it
1:46:08 like a nefarious system but the way they
1:46:11 wrote that law they wrote it based on
1:46:14 the the where the state of Technology
1:46:17 where it was when they when they passed
1:46:18 the law not anticipating that something
1:46:21 like advanced voice would come along
1:46:23 that's kind of a non-
1:46:25 nefarious version of that so so so
1:46:29 advanced voice is illegal in the EU
1:46:31 right now I think they may have worked
1:46:33 something out they probably did a carve
1:46:35 out but
1:46:38 mcon yesterday I think said this let me
1:46:42 I gotta find LinkedIn
1:47:07 where is it let's
1:47:14 see super important I want to insist on
1:47:17 that because it's super important to
1:47:19 assess fairly the fact that our former
1:47:22 model is over and this is not a question
1:47:25 of adjustment I want to insist on that
1:47:27 because it's super important to assess
1:47:30 fairly the fact that our former model is
1:47:32 over our former model is over he
1:47:37 recognizes that how we did things before
1:47:39 is over because of AI and this is not a
1:47:42 question of adjustment this is not the
1:47:45 same world 2024
1:47:47 2025 and on top of that we are making
1:47:51 the same mistakes on top of the key
1:47:52 elements I mentioned we are over
1:47:54 regulating and under investing over
1:47:56 regulating and under
1:47:58 investing so just if in the two to three
1:48:02 years to come if we follow our classical
1:48:04 agenda we will be out of the
1:48:07 market I have no
1:48:09 doubt that's [ __ ] huge in my
1:48:14 opinion right we will be out of the
1:48:16 market if we don't get our [ __ ] I want
1:48:18 to insist on that because it's super
1:48:20 important to assess fairly the fact that
1:48:23 our former model is over and this is not
1:48:26 a question of adjustment this is not the
1:48:29 same world 2024
1:48:31 20125 and on top of that we are making
1:48:35 the same mistakes on top of the key
1:48:36 elements I mentioned we are over
1:48:38 regulating and under investing so just
1:48:41 if in the two to three years to come if
1:48:44 we follow our classical agenda we will
1:48:47 be out of the market I have no doubt so
1:48:51 there you go that was when I when I went
1:48:54 to uh I went I was invited to speak in
1:48:56 Washington
1:48:58 DC three weeks ago now and we went
1:49:01 around to offices of senators and
1:49:03 Congress people and the message was
1:49:07 essentially like one learn what the [ __ ]
1:49:11 these tools can do I didn't use that
1:49:14 language but like one of the things I
1:49:17 asked was do you use chat GPT oh no we
1:49:20 now now we don't yes I probably consider
1:49:23 that yeah you probably
1:49:26 should so that was one of
1:49:28 them the other one was don't just
1:49:32 legislate from a place of fear one of
1:49:34 the things about the AI
1:49:37 Salon is that we're a group of AI
1:49:41 optimists but but we do it with our eyes
1:49:44 open like we understand there are risks
1:49:46 we understand the bad things can happen
1:49:49 and these tools are here and these tools
1:49:52 are arable and and they they make
1:49:55 remarkable things possible so one of the
1:49:58 things I encouraged legislators was what
1:50:01 if on top of legislating to protect us
1:50:05 from the risks of AI you're also
1:50:08 legislating from a place of optimism to
1:50:11 protect our access to these remarkable
1:50:14 tools so that we can
1:50:19 compete what if
1:50:23 but you can't be in that place of
1:50:24 optimism if you don't know what the [ __ ]
1:50:26 the tools
1:50:29 do our legislators don't know how
1:50:31 [ __ ] Wi-Fi
1:50:33 works and this stuff is moving fast and
1:50:36 so like one of the things I said was if
1:50:39 you're not educated about this or if you
1:50:41 take a snapshot of where the technology
1:50:43 is today and you write
1:50:45 regulations based on where the
1:50:47 technology is today not anticipating
1:50:49 where it's
1:50:50 headed you may write a law that by by
1:50:53 the time it goes into effect hobbles you
1:50:56 which is just what happened in the EU
1:50:59 right they've got this legislation that
1:51:01 on the surface look good based on a
1:51:03 technology at a point in time the
1:51:06 technology has moved over here the law
1:51:07 is back here and it's [ __ ] them up so
1:51:10 mcon saying we got to we got to knock
1:51:13 this [ __ ] off we've got to invest and
1:51:15 we've got to we've got to figure out a
1:51:17 way to open up the regulations and still
1:51:18 make people safe so just we're we're in
1:51:22 absolutely fascin in times so for what
1:51:25 that's worth little bit of soap box
1:51:26 there
1:51:27 huh you can make money with if your
1:51:31 country allows you
1:51:36 to what
1:51:40 if what if we actually understood what
1:51:43 we were writing laws
1:51:45 about that sounds like competence I I
1:51:48 don't I I competence in the government
1:51:53 I don't think
1:51:54 [Laughter]
1:51:55 so it's not it's not something I want to
1:51:58 see anytime
1:51:59 [Laughter]
1:52:04 soon all
1:52:07 right so I hope you've enjoyed this
1:52:09 tonight uh I'm getting a little crispy
1:52:13 crispy crispy crusty voice is going
1:52:15 voice is getting a little shaky but I do
1:52:17 have my fluids got my giant water
1:52:24 um if you haven't joined the salon
1:52:26 please go do that please introduce
1:52:28 yourself it's step number two first one
1:52:30 is like the welcome video here's what
1:52:32 we're
1:52:33 about in our little journey to AI
1:52:35 literacy play first mindfully create
1:52:38 generously lead and then the second step
1:52:41 introduce
1:52:43 yourself being a part of a community
1:52:46 like
1:52:48 this as work gets weirdly disrupted is
1:52:52 going to be in increasingly
1:52:56 important cuz a lot of future work is
1:52:58 going to come from people in here who
1:53:02 figured this AI [ __ ] out and they're
1:53:04 going to be able
1:53:05 to navigate their way through this crazy
1:53:08 whatever this work environment is going
1:53:10 to be we're going to see in the next
1:53:11 three years um and they will pull the
1:53:14 people that they like and Trust in this
1:53:17 community with them so be there been fun
1:53:20 cool AI guy thank you insightful and
1:53:22 learn learn something new thank you Kyle
1:53:24 thank you cool a guy appreciate that um
1:53:28 yeah my hope is that my My Hope Is that
1:53:31 even though I take a uh chat add kind of
1:53:34 Meandering Way to make points sometimes
1:53:37 that uh that there's some value here so
1:53:39 I appreciate that a lot kakee uh making
1:53:42 a request for the pink bow so let's wrap
1:53:45 this up with some pink bow
1:53:48 action I got in a fight for this tonight
1:53:50 so I'm going to take good advantage of
1:53:52 it
1:53:55 it thank you Kyle for all you do for us
1:53:58 thank you Dr J
1:54:01 um okay so do do that follow my channel
1:54:05 um if you want to support me directly
1:54:07 subscribe to the lives that actually
1:54:09 just goes directly to the to the channel
1:54:11 and supports me um I love your LinkedIn
1:54:14 article too oh great thank you Lisa I
1:54:16 appreciate that very cool um I don't
1:54:20 know which one oh the LinkedIn article
1:54:21 about uh the holy crap rap ification of
1:54:26 work I you know that's
1:54:30 uh that's where we are if all you have
1:54:32 to do right now I think I think the
1:54:35 skill if if you want to if if you want
1:54:37 my opinion on the skill if there's a
1:54:39 skill that you should be developing more
1:54:42 than any other right now and it's not
1:54:44 even a skill it's an aptitude is
1:54:47 aggressive
1:54:49 curiosity just get curious hang out on
1:54:52 LinkedIn go to Reddit wherever anyone's
1:54:54 talking about AI just find places people
1:54:57 are talking about AI start following
1:54:59 people on Twitter who are talking about
1:55:01 AI start following people on LinkedIn um
1:55:04 that are talking about it if you're into
1:55:05 the art world there's a whole [ __ ] AI
1:55:08 art world going on there's a [ __ ]
1:55:10 fantastic creative community in the AI
1:55:13 salon and in fact it's the kind of basis
1:55:15 of this community is creative
1:55:18 expression your energy is amazing thank
1:55:20 you fun guy and from a guy named fun guy
1:55:22 guy uh you know I appreciate that a lot
1:55:27 oh
1:55:29 tomorrow so I do a less zany version of
1:55:33 this on LinkedIn every Friday called AI
1:55:35 office hours and what's cool is similar
1:55:38 to this there's kind of a regular group
1:55:41 who showed up to my AI office hours so
1:55:43 it's tomorrow at 11:00 a.m.
1:55:47 Mountain um I have ai office hours so if
1:55:49 you go to my LinkedIn profile I'm Kyle
1:55:52 Shannon on
1:55:53 LinkedIn um I'll post in the morning a
1:55:56 reminder but you can also go look at
1:55:58 previous events of AI office hours I've
1:56:01 been doing them I think for a year now
1:56:03 so it's the same URL so you don't have
1:56:06 to see the the reminder tomorrow you can
1:56:08 just go to a past event just copy the
1:56:10 Google meet link it's the same link that
1:56:13 I've used for the whole
1:56:14 year and just come hang out with us and
1:56:17 it's it's it's a really remarkable group
1:56:19 of people who are um doing everything
1:56:23 from creating automated agents to people
1:56:26 who are just trying to figure this stuff
1:56:28 out to people who just kind of want to
1:56:29 hang out and talk about AI um and and I
1:56:34 would argue that the group of people
1:56:36 that shows up to AI office hours are
1:56:38 like some of the some of the brightest
1:56:40 people with this generative AI stuff you
1:56:43 know in the game so that's kind of fun
1:56:45 so that's tomorrow 11 a.m. Mountain Time
1:56:47 on LinkedIn all right just followed you
1:56:50 on LinkedIn cool thanks cool AI guys I
1:56:53 will are you cool AI guy over there I
1:56:55 will I will look you up and and connect
1:56:57 with you appreciate that all right so
1:56:59 let's do our uh let's do this is the 30
1:57:02 minute goodbye here on the AI learning
1:57:04 lab you got to get used to these you
1:57:05 know how like 10 minutes ago I said okay
1:57:07 I got to go now and I'm still talking
1:57:10 that's how this goes but we do a thing
1:57:13 every night here if you're a gen xer
1:57:15 there was a thing called Romper Room
1:57:17 back in the day and at the end of every
1:57:19 show in Romper Room they would do magic
1:57:22 mirror and and they would look into the
1:57:24 magic mirror and they would be looking
1:57:26 into the TV out into the audience and
1:57:28 they would say goodbye to people and
1:57:30 they never [ __ ] said goodbye to Kyle
1:57:33 because it was a weird name it was
1:57:34 always Jim and Bobby and Sally and Fred
1:57:38 [ __ ] Fred where's Kyle so so we do that
1:57:43 so who's in here tonight Joy P's in here
1:57:46 Joey's in the house Obi-Wan great to see
1:57:49 you Lori k
1:57:53 ain't no [ __ ] Romper Room this this
1:57:56 is Lori Lori this is totally Romper
1:58:03 Room this is padded Romper Room Dr J
1:58:08 good to see you spin B3 spin B3
1:58:10 California magic mirror me magic mirror
1:58:12 me I did it I did it Papa JJ I peed
1:58:15 myself the first time she said my name
1:58:17 on the TV that's awesome
1:58:23 James Cal good to see you um news wait
1:58:27 what is this news in the something in
1:58:30 the new oh wait new
1:58:32 wisdom at the
1:58:36 naah that sounds like a sounds like a
1:58:39 Monty Python sketch uh Brandon great to
1:58:41 see you Friday night date night tomorrow
1:58:43 oh yeah Friday night date night so we'll
1:58:45 be here tomorrow same bat Time same bat
1:58:47 Place 8:00 p.m. mountain time and uh we
1:58:50 call it Friday night date night you can
1:58:51 bring a date
1:58:53 and uh dramatically disappoint them
1:58:55 that's always fun hi champy hi champy
1:58:58 did you come back in did you get some
1:59:00 cheese did you get some
1:59:03 cheese um okay who else is here tonight
1:59:06 Winston Hardy great to see you sir
1:59:09 Silver
1:59:11 Fox brother
1:59:13 52 uh Be not Afraid Danielle thank you
1:59:16 Source Camp I do the same
1:59:18 when when comment change my name wait I
1:59:21 do the same when comment change my name
1:59:24 uh thank you Winston I appreciate that I
1:59:26 don't know what that means Source Camp
1:59:28 thank you
1:59:29 Winston very generous of you as always
1:59:32 James Camille I appreciate your time and
1:59:33 sharing knowledge thanks thank you
1:59:35 thanks for hanging out like this
1:59:37 wouldn't work like this would be stupid
1:59:38 if no one was here so so this works
1:59:41 because you all show up uh AI Bob's in
1:59:44 the house um kakee uh who else we got
1:59:50 here EBS
1:59:53 c e
1:59:55 busing
1:59:57 yeah you be
2:00:00 busing cling cling
2:00:02 [Laughter]
2:00:08 cling oh good lord people frostbitten
2:00:12 good to see you Alia's wife thank you as
2:00:15 always Lori K
2:00:18 today
2:00:20 um who's that qu quigly quigo quy quy
2:00:27 quigo you people with your weird ass
2:00:30 names literally spending the week at a
2:00:32 conference in higher ed and Tech you're
2:00:35 more informative and fun
2:00:38 yes oh you didn't even hear me go off on
2:00:41 a rant about education come back
2:00:43 tomorrow I'll give you an earful about
2:00:45 education oh yeah we're dealing with AI
2:00:48 it's okay we've got AI detectors we've
2:00:50 taken care of AI
2:00:55 yeah don't get me
2:01:01 started so Kyle these are Botox oh
2:01:04 that's really good that's that's really
2:01:08 good welcome to
2:01:13 Botox Silver Fox has a silver tongue
2:01:17 she's good with the words she she comes
2:01:19 up with all of our catchphrases I think
2:01:21 we need we need to hire silver foxes as
2:01:23 a a marketing uh meme maker I think
2:01:27 you'd be very good at
2:01:31 that oh man AI Bob let's see who else we
2:01:35 got all right I think I got everyone
2:01:36 probably didn't there's a bunch of
2:01:37 people in here tonight but anyway um I
2:01:40 hope you had fun tonight um I know it's
2:01:42 kind of all over the place but there
2:01:44 were a lot of new people whoever did the
2:01:46 Boost that got lots of people in here
2:01:48 thank you for doing that um it's always
2:01:51 fun when there's new people here it
2:01:52 looks like a fair amount of them stuck
2:01:54 around so very very cool all right it
2:01:57 was fun thank you always do all right
2:01:59 everybody I'm have a good night so 11:00
2:02:01 a.m. tomorrow for AI office hours thank
2:02:05 you this one always makes me feel
2:02:06 awkward
2:02:09 uh well that nose is really
2:02:16 impressive and then walk walking bow
2:02:18 talk yeah I'm not I'm not walking
2:02:20 outside of the house in that bow
2:02:24 um and then I'll see you tomorrow night
2:02:26 for Friday night date night all right
2:02:29 peace out everybody hope you have a good
2:02:30 night