Feb 23, 2024

Description

Chapters

Transcript

0:01 you
0:02 ready here we go we're
0:06 going Oh wrong
0:08 [Music]
0:20 shirt this place I can R
0:26 [Music]
0:27 my ga my thoughts and sweet
0:32 [Music]
0:34 s isest place where the feelings
0:40 [Music]
0:41 on that's not
0:43 [Music]
0:59 too
1:00 [Music]
1:12 is this place I can rest my for
1:18 head GA my thoughts and S silence you
1:23 singing
1:24 [Music]
1:29 champy
1:31 [Music]
1:59 w
2:02 [Music]
2:15 hello everybody Welcome to the AI
2:17 learning lab I got to go change my shirt
2:19 although I got food on it I got to go
2:22 change my shirt I can't I can't be
2:24 expected to be professional in a shirt
2:26 like this schutz all over it you can't
2:28 read the text on it we got to go get
2:31 black
2:31 [Music]
2:59 t-shirts
3:00 [Music]
3:06 [Applause]
3:08 it's Friday night date
3:10 night it's Friday night date night
3:13 welcome to the AI learning lab if you've
3:14 never been here before you are in for a
3:17 treat good Lord we start out with
3:19 singing
3:20 dogs we get clothing change we get
3:22 costume
3:24 changes the last
3:27 week I've been
3:37 I've been going on rants I've been going
3:39 on tie
3:40 raids and over stupid
3:43 [ __ ] people are like how do you spell AI
3:45 I'm like how do you spell AI well you
3:48 first you learn the alphabet it's just
3:49 been like kids get off my
3:52 [Music]
3:58 lawn um
4:01 [Music]
4:04 it was funny after the live last night
4:07 we were watching I forget the name of
4:09 the show is some space show it's about
4:13 these astronauts that were up in the
4:15 space
4:16 station and they did it they did an
4:18 experiment and it's split time and now
4:21 there's two of them and it's called but
4:24 doesn't
4:25 matter but I'm I'm
4:28 uh I'm watching in it like you do with a
4:32 show you put it on the
4:34 TV you know for those of you that only
4:37 do Tik Tok TV is a there used to be this
4:41 thing called television and then they
4:43 would have three channels and you'd
4:44 watch shows on it and it's sort of
4:46 different it's sort of like that now but
4:48 totally different anyway I was watching
4:50 that thing much bigger than the one I
4:52 grew up with right cuz now it's like for
4:54 a nickel you get an 80-in television
4:58 so so much my nickel expensive 70in
5:04 television and I'm watching this show
5:06 and and there's this scene where she
5:09 comes back from
5:10 space and they go out in the desert and
5:12 they get her and there's Russian
5:14 helicopters and there's these army
5:19 trucks and oh it was before they went
5:22 get her it was like when they were when
5:23 when they realized she was coming into
5:25 space and they all they all scramble out
5:27 to the
5:28 tarmac and they climb in the helicopters
5:31 and they're strapping in and then they
5:33 cut to the scene where they take off and
5:37 it's like these three Russian
5:38 helicopters take off and the dust kicks
5:40 up and there's these 10 army trucks
5:42 driving down the tarmac and they're like
5:44 taking off to go save the
5:47 astronaut and what hit
5:49 me
5:52 was holy
5:54 [ __ ] like you're going to be able to
5:57 just ask for that scene right
6:00 three of these kind of helicopters
6:02 covered in dust on a Dusty tarmac at
6:06 this airport with these kind of support
6:10 vehicles and the three helicopters take
6:12 off as the trucks take off and drive out
6:15 out into the
6:18 distance and I was thinking
6:20 about that shot that one
6:23 shot probably took two months to
6:29 plan
6:30 they had to rent Russian helicopters
6:33 they had to find an Airfield they had to
6:36 get all the trucks they had to get
6:39 Pilots
6:41 drivers probably stunt doubles safety
6:44 people medical people craft people
6:48 food um accommodations for people it it
6:53 it was probably days to set
6:55 up
6:57 right like that one shot that 10 seconds
7:06 of
7:08 was I don't know $100,000 $200,000
7:12 300,000 like some large amount of
7:18 money and now you're just going to be
7:20 able to ask for it and then I wake up
7:23 this morning to The Hollywood Reporter
7:28 headline that says Tyler
7:32 Perry
7:34 cancels his planned $800 million Studio
7:38 ex
7:41 extension for his movie
7:43 [Music]
7:48 company cancels it and then the headline
7:51 says be after seeing open AI
7:54 Sora someone dm'd me and said it might
7:57 have been taxes too well it might have
7:58 been taxes but the headline in The
8:00 Hollywood Reporter was after seeing open
8:04 ai's video model Tyler Perry
8:07 cancels the 800 million 80 million I
8:11 don't know some large amount of
8:13 millions movie studio extension holy
8:19 [ __ ] so that happened that was wild it
8:23 it was wild sort of thinking about it
8:24 and then and then kind of waking up to
8:26 news of like oh yeah this is this is
8:30 this is what's going to happen all right
8:32 I'm going to go change my
8:34 shirt all right I'm going to go change
8:36 my shirt it won't even be five but
8:39 whatever right champy you
8:43 good oh you good you fat little dog you
8:48 chubby little tubby you tubby little
8:52 dog you bad boy are you bad dog are you
8:57 a good dog are you a bad boy bad
9:03 [Music]
9:13 we'll go Brooklyn Brooklyn in
9:27 house all right
9:34 boom Brooklyn in the
9:38 house how's the hair how's the hair
9:40 looking is it pretty oh my goodness I
9:43 can't go live if my hair doesn't look
9:48 pretty all right so so here's what we do
9:52 here if you're curious about artificial
9:55 intelligence if you're like my goodness
9:57 everyone's talking about AI what do I do
9:59 where do I start Kyle can you help
10:02 probably not
10:05 but I will encourage you to give it a
10:11 shot it is my
10:13 contention until further notice that
10:16 there are no
10:18 experts there are no experts but Kyle
10:22 someone gave a seminar I
10:25 know I
10:28 know dud
10:32 apparent so Google or not Google
10:35 Microsoft co-pilot is get has gotten a
10:39 little bit of a
10:41 Polish I saw someone someone said on
10:44 Twitter that uh Coop pilot got you can
10:47 do horizontal images on it now blandon's
10:50 coffee uh no this is just normal coffee
10:53 problem with the blandon's coffee is
10:54 I've got to um grind it
10:58 and being with ADD and everything you
11:01 know sometimes you just forget that [ __ ]
11:04 so but I'll I'll do up a little a little
11:06 tub of it so I can do it but the
11:08 blandon's coffee is good it's real
11:13 smooth oh yes oh no this is just tea I'm
11:16 just having tea
11:18 tonight black
11:21 tea all right so here's what we do so if
11:24 you have questions comments about um AI
11:29 pop them in the comments
11:31 below
11:33 um before we get started why don't you
11:36 share the live so if you don't know how
11:38 to share the live it's super easy you
11:41 ask someone younger than you hi hey hey
11:45 Sally Sally listen you know what I was
11:48 wondering how do you share the live on
11:50 Tik
11:51 Tock you what you you just click on the
11:54 share
11:56 button how
11:58 what where's the Sally Sally Sally
12:03 Sally kids Sally listen where's the
12:06 share butt find it yourself
12:10 I so just find just ask Sally ask
12:13 someone younger than you they'll
12:16 know and then you push that button and
12:18 and here's then you but you're like but
12:20 then you're like okay Kyle okay I found
12:22 the share button who do I share this
12:24 live with here's who you want to share
12:27 it with this is really important
12:31 if you want your life to be
12:35 interesting you share it with someone
12:37 you know it'll piss
12:42 off thanks to you I'm more interested in
12:45 AI please continue with the great work
12:47 in educating us oh thank you very much
12:50 pavon yeah you know and and so pavon you
12:53 know thanks to you I'm more interested
12:54 in AI That's like that's if if if I have
12:58 a hope or dream for this channel it's
13:00 not much more than that it's that I hope
13:02 people like are willing to give it a
13:04 shot I'm more interested in it I'm
13:07 willing to learn what it is I'm I'm I'm
13:10 willing to play with
13:11 it I might even play with it enough that
13:14 I have my first Kevin McAllister
13:20 moment salty is my dog's name she was no
13:24 help oh Sally Sally is my dog's name she
13:27 was no help Sally can't tell you how to
13:29 share your Tik Tock
13:34 lives oh the other thing you can do if
13:37 you want to get carpal tunnel you can
13:38 tap the screen and sometimes the
13:41 Irregulars they like when I get boring
13:43 when I just sort of go off and they're
13:44 like oh good here he goes again and they
13:47 either do one of two things they'll go
13:48 to the AI salon and share images and
13:50 make jokes and do all that sort of [ __ ]
13:52 or they'll get to tapy taping and get
13:54 give themselves carpal
13:56 tunnel they have little challenges
13:58 amongst themselves you can do
14:00 that tapp's free except for the you know
14:07 surgery all
14:09 right I'm focused now on creating an AI
14:14 School based on something an AI School
14:17 based on something that's super cool Mr
14:20 Studio dhu that's super
14:28 cool no that laugh that laugh did I
14:33 laugh I don't know if I laughed did I
14:35 laugh I I'm not a very humorous person
14:38 you know I don't think I laughed Cory
14:41 Sandler i is it possible you were um
14:45 what do they call it in AI
14:48 hallucinating perhaps you put a
14:50 different kind of mushrooms in the
14:51 strogen off this
14:54 evening I don't think I was laughing I
14:56 wouldn't laugh at matter so serious as
14:58 artificial
15:02 intelligence good Lord this is very
15:05 serious very serious excuse
15:08 me what are my
15:11 qualifications W I'll tell
15:14 you my
15:16 qualifications it must be some Advanced
15:21 degree pinkies out
15:26 everyone we're classy
15:30 all right all right I tried grock today
15:32 world's fastest AI isn't that crazy
15:34 Kieran yeah grock is crazy grock's
15:38 nuts um let me go back to
15:44 my my list from yesterday so I added one
15:47 today it looks like Sora is getting Sora
15:49 is getting better so I added that
15:53 one where's my
15:56 image where's my image that I had up I
15:59 lost it oh here's the little
16:02 area there we go
16:06 um you
16:08 got open AI
16:12 Sora which is their new video thing if
16:14 you haven't seen their new video thing
16:15 we can talk about that we got 11
16:19 Labs that has a a coming text to sound
16:24 effects
16:26 model and they did a they did a version
16:29 of the Sora videos with sound in them
16:31 that's really cool they also 11 Labs
16:33 also got accepted into the Disney
16:35 accelerator which is a big
16:37 deal
16:40 um Google pro 1.5 sorry Gemini Pro
16:47 1.5 it like it PE people it's
16:51 funny Sora came out right at the same
16:55 time Gemini 1.5 Pro came out and
16:58 everyone was talking about Sora nobody
17:00 was talking about Gemini because it was
17:03 weird Gemini came out right after Gemini
17:05 Ultra launched and then that was kind of
17:07 a fizzle and then they just sort of
17:08 quietly released the 1.5
17:12 thing A lot of people are talking about
17:14 I'm seeing a lot of people talking
17:18 about Gemini 1.5 the million million
17:22 token context window is a big big
17:27 deal so that was fast
17:32 fascinating Deco here you can upload a
17:34 reference photo mid Journey's apparently
17:37 in talks with x.com to to be to have mid
17:41 journey be the IM image generation in
17:44 engine
17:46 inside
17:48 grock Elon musk's version of chat GPT
17:52 that would be
17:53 wild Nvidia had their fourth quarter or
17:56 their whatever
17:58 quarter yeah fourth quarter cuz we're in
18:00 the first quarter fourth quarter
18:02 earnings call blew it out of the
18:05 water they have a market cap higher than
18:07 most Count's G gdps at this
18:12 point
18:14 um people are still talking about the
18:16 Vision
18:17 Pro gro.com not the same grock as Elon
18:21 musk's grock but grock the chip making
18:23 company spelled with a Q instead of a k
18:27 gq.com
18:29 go there right now gq.com and type in a
18:32 prompt watch how fast that thing blasts
18:34 out answers it's it's so weird it's so
18:39 weird because you tell it something that
18:42 you know it's got to make up and then
18:44 it's just like there it is there it is
18:47 there it is there it is it's very weird
18:51 um reddits filing for an IPO and then
18:53 co-pilots so like that's that was my
18:55 short little list from yesterday so we
18:57 can talk about that stuff thank you for
19:00 the all the updates throughout the day
19:02 you're welcome yeah I put a few few
19:04 updates today on the
19:07 LinkedIn can I send a PowerPoint slide
19:10 to chat GPT and have it refine
19:14 it I don't know what you mean by refine
19:17 it but let's experiment with that I
19:20 think that's actually a really
19:21 interesting
19:22 experiment so we'll go over
19:28 to
19:34 that's a good
19:37 one oh wait oh I know what I can do hang
19:39 on
19:41 new
19:54 whatever okay so there is a slide I'm
19:57 going to export this as
20:00 PowerPoint cuz I'm in keynote cuz I'm on
20:02 a Mac because the Mac is a superior
20:06 platform and I absolutely judge all of
20:10 you people on PCS you like but we got
20:12 gpus and I'm like that's all
20:15 right you don't have a Mac all right and
20:19 then we got Apple silicon now the M3 you
20:22 know what you know what Apple's doing
20:23 with all that Apple silicon right now
20:25 they're they're they're they're
20:27 optimizing all of their AI [ __ ] to run
20:29 locally on the M M1 m2s and M3s you wait
20:33 you're going to be like I wish I had a
20:35 Mac you wait you just wait all right
20:40 we're going to call this untight so this
20:42 is um a file
20:47 um um
20:49 file hygiene best
20:52 practices when you save your file always
20:56 name it Untitled and save it to the
20:58 desktop
21:00 okay and what's nice is the operating
21:03 system will add the one two or three
21:05 after it so they don't you know you're
21:07 not going to overwrite
21:10 [Music]
21:17 anything all right so let's go over to
21:20 the chat
21:23 je here we are at the chat
21:28 Je that's how you're supposed to
21:29 pronounce it by the way I don't know if
21:31 you knew
21:32 that you can make money
21:39 with all right so I'm in chat J
21:43 4 and I'm gonna upload that PowerPoint
21:46 file if it lets me
21:49 and does it looks like it
21:53 does and then you asked can I upload a
21:56 PowerPoint file and have have it refine
21:58 it let's let's just
22:00 say
22:01 what's on this
22:10 slide reading documents that's
22:16 promising all right it pulled the text
22:19 off it so great so I'm going to
22:21 say can you change the
22:27 background of the slide
22:30 to
22:33 Orange and the
22:35 text to dark
22:39 blue and
22:50 add a row to the
22:54 table and
22:57 include three let's
23:02 see
23:06 um
23:09 idog dot what is it
23:15 doio and
23:18 then give me a
23:22 downloadable
23:24 version
23:26 of the slide
23:31 we will see so I'm currently unable to
23:33 directly modify PowerPoint slides the
23:36 answer is
23:42 no it can read it it can't do nothing
23:47 with it I don't think that'll last that
23:49 long and quite frankly um co-pilot
23:53 within Office 365 can probably do that
23:56 right now id. thank you all
24:00 right it's only going to create an image
24:02 if it does anything yeah yeah exactly
24:05 you were right Joker you were right mhm
24:09 yeah I'll tell you what I'll tell you
24:11 what no I'll tell you what what are you
24:14 going to tell me what are you going to
24:16 tell
24:18 me see always act like this this is
24:21 weird I thought this was some sort of
24:23 learning labit life tough tougher when
24:26 you're
24:27 stupid I I thought this was some sort of
24:30 nerding
24:31 lab he don't look like a professor and
24:35 he's acting like a dum dumb why does he
24:38 act like a dum
24:42 dum ask what it can output okay what can
24:48 you
24:50 [Music]
24:57 output it doesn't no it didn't I I
25:00 wonder if I
25:01 say can
25:04 can can you modify the slide
25:09 using Code
25:15 interpreter analyzing this is
25:18 promising hi Kyle and Irregulars and the
25:21 rest what's up what's up what's up talk
25:24 what's up Dylan what's
25:27 shaking well Shak
25:30 [Music]
25:35 in Winston Welch it seems I encountered
25:39 an issue the python. PPX Library does
25:43 not directly support adding a new row to
25:47 an existing table the way I
25:51 attempted I think it's
25:55 doing I've made the modifications to
25:58 change the slide background to Orange
26:00 the text to Blue however I encountered a
26:02 limitation adding a new row
26:05 programmatically so this
26:09 should it
26:11 should
26:13 people oh I see it
26:16 look it did
26:18 it yes you
26:22 can just tell it it's got to use code
26:25 interpreter and it'll write the file out
26:27 so that's that's a power Power Point
26:31 file it you know it wasn't smart enough
26:33 to know that to change the background
26:35 color of the table
26:37 also but it did change the background to
26:40 Orange and it changed the text to Blue
26:42 so uh yes you
26:44 can and that's actually I'm I'm really
26:47 glad you asked that question that's a
26:48 really good um okay all right so listen
26:54 listen
26:56 people listen
27:04 play
27:06 play play
27:09 play
27:11 play
27:13 play what's that other word
27:17 play start playing with these tools
27:20 don't feel like you got to learn them
27:22 and you got to learn and do your work on
27:25 them and make them more efficient and do
27:26 automations if you want to do all that
27:28 [ __ ] cool if you got [ __ ] you want to do
27:29 go do
27:31 it but rather than just sitting
27:34 around twirling your titly
27:38 bits play with AI do [ __ ] like I just
27:42 did someone goes He Can it can it update
27:46 a PowerPoint slide I don't know
27:54 maybe maybe it
27:57 can let's try another experiment I've
27:59 never tried this
28:03 before
28:10 um we're going to go over
28:21 here does he always thing like that it's
28:23 very annoying does he want to chase
28:25 people away from this channel it seems
28:27 like he wants to chase people people
28:28 away from this
28:30 channel it's AI learning lab You' you'd
28:33 think he'd want students to stay in here
28:35 to learn but then he sings the songs and
28:37 that that singing he's not a very good
28:39 singer and then is he singing any song
28:41 that actually exists or is it just like
28:43 a tick he doesn't seem
28:49 stable what am I doing I am looking
28:56 for okay so it's got it's got okay so
29:00 how are we going to do
29:03 this
29:04 [Music]
29:11 um yeah I guess I'll just use that
29:14 okay so let me export this as a
29:18 PowerPoint export to
29:21 PowerPoint fantastic Bob tell him what
29:25 is won uh you haven't won anything you
29:28 just exported PowerPoint there's nothing
29:29 to win there don't forget to save your
29:32 file name as Untitled to the
29:36 desktop yes replace it oh
29:39 [Laughter]
29:42 no okay back over to chat
29:46 je and we're going to
29:49 upload that PowerPoint we just
29:52 created to Yan
29:55 desktop but I don't see it where is it
29:58 desktop oh I guess I did it to
30:01 iCloud there it is is that
30:04 it yes
30:16 okay I'd like to come up with a
30:27 presentation
30:29 that you use
30:34 the
30:42 template in the
30:46 attached
30:48 PPT
30:51 for
30:56 um but first can you tell me what text
31:03 is in the file right
31:06 now all right so what I'm doing right
31:08 now is I'm just uploading the PowerPoint
31:10 document and I'm getting chat GPT to
31:13 recognize that there's something there
31:15 and have it go look at it when it says
31:17 analyzing that means it's using Code
31:19 interpreter where it's actually writing
31:22 an executing its own python code looks
31:24 like there was an error oh it's a
31:26 keynote file hang on on I thought I
31:29 exported it as PowerPoint didn't I was
31:33 that dum dum dang
31:35 it oh that's PDF dang it don't be a dumb
31:39 dumb don't be a dumb dumb like the AI
31:41 learning lab so here's what you should
31:43 do with AI you should get curious about
31:46 it you should play and you should not be
31:49 a dumb
31:50 dumb being a dumb
31:54 Dum does not help
31:57 anybody do desktop replace okay what
32:02 happened did we do it yes some changes
32:06 were made
32:08 fine let's
32:11 upload Untitled there we go it still
32:14 says keynote oh because I was in the
32:16 wrong folder let me get rid of that see
32:18 that was dumb Dum let's not be a dum dum
32:21 this
32:22 time here's the real problem I have to
32:25 turn my brightness down so much so that
32:27 you can see the screen
32:29 and then because of that I can't see the
32:31 screen so okay here is the PPT
32:39 okay analyzing beautiful oh so what I
32:41 was saying was if you open this box that
32:44 says analyzing you see all this code
32:46 well what's that code code interpreter
32:49 will it's got access to a bunch of
32:51 python libraries and it can write and
32:54 execute its own python code
32:57 so chat jpt doesn't natively read
33:00 PowerPoint but someone at some point in
33:02 history wrote a library for python that
33:05 says I can read a PowerPoint
33:08 document um so it's doing that okay so
33:11 it says on the first slide it says AI
33:13 open AI prompt book blah blah blah
33:15 second slide it appears to be
33:16 placeholders perfect great
33:20 okay I want you
33:23 to create a
33:27 10
33:31 slide
33:33 presentation
33:35 about
33:38 why gummy
33:40 bears are better than
33:46 Sushi let's
33:50 just start with the
33:54 text of the title and the
34:00 10
34:03 slides that each have a title
34:11 subhead
34:13 and body
34:16 copy
34:18 okay
34:19 so chat je is going to show us its sweet
34:23 sweet copyrighting
34:27 skills
34:34 the sweet debate Gummy Bears versus
34:37 Sushi exploring taste beyond the
34:41 traditional in this delightful
34:43 exploration we dive into the colorful
34:45 world of gummy bears and how they stand
34:48 out against the revered Sushi join us on
34:51 this journey of taste texture and fun
34:55 what about gummy Sushi though oh that's
34:57 a good idea
34:59 all right
35:04 conclusion
35:06 let's
35:08 change 9 and
35:12 10
35:14 to
35:15 present the Innovative
35:19 idea
35:21 of gummy
35:27 sushi
35:33 oh that's good I like it all right
35:37 Innovative Fusion The Best of Both
35:39 Worlds gummy Sushi imagine the joy and
35:42 versatility of gummy bears merged with
35:44 the Artful presentation of sushi gummy
35:47 Sushi brings a playful twist do you know
35:50 that we could probably start this
35:52 business and get it funded the gummy
35:54 sushi shop are you kidding me you tell
35:56 me who wouldn't go there every everyone
35:58 would go
36:01 there the gummy sushi shop come on
36:04 people we're innovating here at the AI
36:07 learning if you think this was just
36:09 about oh I come in here to learn AI this
36:11 I'm a prompt engineer and I go to the AI
36:13 learning lab to learn my prompt
36:14 engineering no you come here for billion
36:16 dooll
36:17 ideas you know it they know it the
36:19 Irregulars know
36:21 it they know why they're
36:23 here you can make money
36:27 with
36:29 who is that dragonfly Alchemy I will
36:32 absolutely give you credit for that
36:34 however all the cash is mine I'm opening
36:38 the world's first gummy sushi shop in
36:41 Toledo Ohio in 2020 five summer of and
36:47 just there you
36:50 go okay anyway what were we trying to do
36:53 oh okay so here's what I'm going to
36:55 here's what I'm going to try to do well
36:57 I'm not going to try to do it here's
36:58 what I'm going to get chat GPT to try to
37:01 do so I uploaded a TW
37:05 slide presentation in PowerPoint format
37:10 and I'm going to see if I can get it to
37:14 populate this into
37:17 the into a
37:20 PowerPoint should I do it should I add
37:22 it to that existing yeah we'll just add
37:24 it to that existing one um okay please
37:28 please use code
37:33 interpreter
37:36 to um add slides to the
37:41 PPT using
37:44 the included
37:48 template and
37:53 populate the
37:56 slides
37:58 with
38:00 this
38:03 lovely with this
38:06 lovely
38:11 gastronomic
38:16 exploration Play I'm telling you play
38:19 analyzing that's promising all right
38:21 let's see what it's doing it should be
38:23 loading in a
38:25 PowerPoint yeah the PowerPoint util
38:32 title is sweet history so it's adding
38:36 slides I think this may be
38:40 working it's slow it's going
38:48 slow oh it [ __ ] up it oh slide 3 to
38:52 eight oh did it do it I guess it did
38:56 it
38:59 holy dang it might have done it all
39:02 right let's see let's see
39:07 people the presentation why gummy bears
39:10 are better than Sushi download the
39:12 presentation here all right let's go
39:16 look at this bad
39:21 boy that was our that
39:24 slide holy
39:26 crap
39:32 it did
39:36 [Laughter]
39:47 it see remember when I said earlier
39:50 there are no experts and you were like
39:52 but KY you're the professor of the AI
39:56 learning lab it's a lab you clearly a
40:00 scientist and I was like oh but cont I
40:04 have no
40:05 qualifications and you were like we
40:06 don't believe you I didn't know it could
40:09 do that I didn't know it could do
40:13 that I knew it could write
40:16 files I knew it could write PowerPoint
40:20 copy I had no idea you could upload a
40:23 thing with a template in
40:26 it and and just have it do whatever you
40:29 do in chat
40:31 J and say use this template as the guide
40:34 and make me a new thing and let me
40:36 download it gummy Sushi store and show
40:39 it
40:45 to gummy Sushi store brought to life the
40:48 AI learning
40:50 lab gummy
40:52 [Laughter]
40:54 Sushi wouldn't that be a cool shop where
40:57 You've Got Sushi chefs and you walk in
40:59 and they've got all their little bags of
41:01 candy and they they sort of neatly do
41:03 them and they they just craft them and
41:05 stick them all together and plate them
41:07 and put them out for you I'm telling you
41:10 man you could charge you could charge
41:14 like $12 for a little Sushi plate that
41:18 would cost you like a nickel and gummy
41:25 bears this is the world's most
41:28 profitable candy
41:34 shop oh that's so
41:38 good oh mushroom soup you say kids get
41:42 off my lawn that's what that's this what
41:44 this week's been like for me I have just
41:46 been an orary little
41:49 Snapper oh all right so while we're here
41:53 at the AI Salon listen listen people
41:56 let's go through some things this is
41:59 really important
42:01 okay this is the
42:03 a how do you get there
42:06 well you
42:10 go my
42:18 [Music]
42:22 eyes see us heading off on that
42:25 adventure
42:26 together the AI Salon is a community
42:29 that I started the week after chat GPT
42:31 came out so December 7th
42:36 2022 and you're like but how why did you
42:39 start it how did you have such Insight
42:41 such foresight
42:43 because I'm
42:45 old
42:47 and in
42:49 1994 I stumbled upon this thing called
42:52 the worldwide
42:53 web Tim burners Lee had created this
42:56 protocol and
42:58 by '94 thing people were starting to
43:00 figure this [ __ ] out and people were
43:01 starting to talk about it and the buses
43:03 in New York City had like Mondo 2,000
43:06 ads and Wired Magazine ads and there was
43:08 starting to be a bit of a buzz about
43:09 this internet thing and I played with
43:12 the internet for a couple of years good
43:15 even good evening good evening to you
43:17 thank you for the kind gift the gifts
43:19 are
43:21 appreciated
43:23 um and so I stumbled on this thing
43:26 called the worldwide web and
43:28 and what I knew about the internet is
43:31 that's where the scientists and Geeks
43:33 and researchers and academics hang out
43:36 and I wasn't one of those I'm a
43:38 theatrical oh you never would have
43:40 guessed with all the stupid voices he
43:41 makes
43:45 exactly and so as a theatrical you know
43:48 got self-esteem issues as you do people
43:52 pleaser type you know you've seen them
43:54 before little insecure lot of crazy
43:58 and I thought well certainly I can't do
44:00 this worldwide web
44:03 thing and so I started to try to learn
44:06 it and it seemed like I think I could
44:08 learn this like I think I can you know I
44:10 some computer
44:13 skills and then I and then because I'm
44:16 smart I had a brilliant idea I said I
44:18 know I'm going to start a club so I
44:21 started this group called the worldwide
44:22 web artist
44:24 Consortium and it was a Shameless
44:26 attempt to surround myself with people
44:29 that knew more about the worldwide web
44:31 than I
44:34 did and guess what I found
44:37 out nobody knew [ __ ] this is New York
44:41 City it's like a big [ __ ] city right
44:44 millions of people like surely there
44:46 must be people here that know more than
44:48 I do what I found were people that were
44:51 really curious about it really
44:53 interested in it that were like yeah
44:56 you're trying to figure it out too what
44:57 did you learn I was like well I learned
44:59 this they're like oh [ __ ] I didn't know
45:00 it could do that and I was like what did
45:02 you learn they're like well you could do
45:04 this they like oh [ __ ] and how do you do
45:08 that and so it turned it turned into
45:10 this thing called whack the worldwide
45:12 web AR Consortium it ended up grown to
45:14 you know like 2,000
45:16 people we met every other week in the
45:19 Sony building in Midtown Manhattan Sony
45:22 gave us for
45:24 free a conference room that sat through
45:27 300 people and we filled that [ __ ]
45:29 thing for years every other week it was
45:34 amazing so when
45:37 I I've been [ __ ] around with stable
45:40 diffusion and I've been [ __ ] around
45:41 with gpt3 and the open AI playground and
45:44 you know I was like oh this AI stuff's
45:46 kind of interesting kind of cool this
45:48 these pretty pictures with staed Fusion
45:50 are kind of cool and then cat GPT came
45:54 out and I used it and I lost my [ __ ]
45:58 mind and I realized holy [ __ ] this is
46:03 just like the worldwide web
46:05 only way more
46:08 powerful way more
46:11 profound and the first thing that popped
46:14 in my head was start a community that's
46:17 that
46:19 community and it's [ __ ] cool
46:23 now
46:26 rabbit
46:28 um cancel
46:31 where's there we
46:34 go so who are we what do we do so my
46:37 first request if if you're interested in
46:39 this is go to the salon. click on the
46:42 top link on the link tree so the link
46:43 tree looks like this the top link right
46:46 here AI Salon Community will take you
46:49 here and then I want you to read who we
46:52 are and why we do what we do here I'll
46:54 help I'll read the first paragraph
46:55 they're short I have ADD I don't like
46:58 reading long [ __ ] but this is actually
47:00 important AI Salon is a ragtag group of
47:03 Misfits Naro Wells and miscreant who
47:05 have been gathered to to have gathered
47:07 to explore the Wilderness of generative
47:09 AI where everyone from entrepreneurs to
47:11 Educators from healthcare workers to
47:13 creatives and everyone in
47:16 between chat PT represents an
47:21 opening for the
47:24 non-engineers to experience the power of
47:27 AI
47:28 that's that's why this is
47:30 different that's why whatever this new
47:32 thing is that we're in the middle of is
47:34 a year old even though Ai and machine
47:36 learning are 30 years old 50 years old
47:39 70 years old depending on where you
47:41 start the start
47:44 line the generative AI movement the
47:47 thing that's going to completely blow up
47:50 everything I think in a good way started
47:54 just over a year ago so this is a group
47:57 to explore that now we have these values
47:59 here my request is that you check out
48:01 the values and you really take them in
48:02 you know read about them curiosity we
48:04 want people that are curious we want
48:06 people that are generous like when you
48:08 learn something you share it if someone
48:11 says oh how did you do that you don't
48:12 get all proprietary we're too new the
48:15 the industry is too new to be
48:16 proprietary everyone's trying to figure
48:18 it out so share put it out there because
48:21 if you put it out there that'll
48:22 encourage someone else to put it out
48:24 there and then you'll learn something it
48:26 and the thing about these values we
48:28 didn't just make this [ __ ]
48:30 up we had this amazing group of people
48:33 and as this community started to grow
48:34 we're like let's make sure that is this
48:36 thing grows we don't lose what we have
48:38 what is it we have and so we articulated
48:41 these values so these came out of who
48:43 this community already is exploration
48:47 people here are adventurous like yeah
48:48 I'm willing to do that oh there's a new
48:50 thing out let's go try it willing to say
48:53 I don't know have that beginner's
48:55 mindset of like I don't know
48:57 but let's
48:58 go that takes
49:01 guts collaboration being willing to say
49:04 hey let's work on this thing together
49:05 maybe if we did it together we could get
49:08 farther than we could if we did it alone
49:09 there's a lot of that going on in here
49:11 inclusion people are coming from all
49:13 different walks of life all different
49:15 backgrounds all different technical
49:17 experiences all different AI depth of
49:21 Knowledge
49:22 from Pate who runs the mechanics Guild
49:26 is op optimizing
49:29 tensor tpus at
49:32 Google to people who are like uh someone
49:35 told me I should learn Ai and they told
49:37 me to come here what do I what do I do
49:41 where do I
49:45 start and sort of tied to that one is
49:49 empathy understand that people are in
49:50 different places with this some people
49:52 are excited some people are scared some
49:54 people are pissed
49:55 off Tyler he's like well I guess I'm not
49:58 opening a new
49:59 studio he's probably got some feels
50:02 right
50:04 now if he's smart enough to know
50:07 that opening that investing that much
50:10 money in that studio now is not a good
50:12 idea I Betty's also smart enough bet
50:14 he's
50:15 getting he getting up to speed on this
50:17 AI stuff pretty [ __ ] quick right now
50:20 and then the last value is
50:23 gratitude like active actively
50:26 acknowledge people in the community that
50:28 give you something or that inspire you
50:30 to do something or or that teach you
50:34 something so those are the values that's
50:37 who this group already is and so here's
50:39 how values work if in reading through
50:42 those values they give you the warm and
50:45 fuzzies then these are going to be your
50:47 people if in reading through those
50:49 values you're like oh god oh that's
50:53 gross that's as gross as a gummy Sho
50:55 Sushi store then these are not going to
50:57 be your people and that's okay but just
51:00 know that like that's why that's why
51:01 they're there so so that's my first
51:03 request second
51:05 request is if you want to join it
51:07 doesn't oh it doesn't cost anything by
51:09 the way I know that's always the first
51:10 thing oh I say oh I say he's in here
51:13 acting all funny and then he's just
51:14 trying to Hawk us as Community what is
51:16 it $495 a year no it's $495 a month what
51:22 do you think we're cheap
51:24 no no it's free
51:29 it is community
51:32 supported what's that mean it means some
51:34 people in the community sponsor months
51:37 to cover our costs or when you go to the
51:40 the link tree at the salon. if you buy
51:42 the merch that
51:44 supports our infrastructure cost here's
51:47 the other cost though the first cost is
51:49 I want you to check out the values the
51:51 second cost
51:53 is I want you to introduce
51:55 yourself and but Kyle John Leland I just
51:59 had a really good conversation with him
52:02 today he just joined he introduced
52:04 himself nice nice generous introduction
52:07 look at that really let us know who he
52:09 is what's he's up what he's up
52:11 to so my request is if you're interested
52:14 in joining introduce yourself and if if
52:16 you're if you're all a scared if you're
52:19 all like but I'm so scared everyone's
52:23 going to judge me that may be true but
52:26 but gener generosity you know what
52:29 generous letting people know who you
52:32 are because you might be just the person
52:35 they need to meet and they might be just
52:38 the person you need to meet so just say
52:40 here's who I am here's what I'm up to
52:43 here's where I am with AI I'm totally
52:45 scared of it I'm totally pissed off
52:47 about it I think they're [ __ ]
52:48 stealing from all the artists what
52:50 wherever you are on the
52:52 journey and just jump
52:54 in and what you'll find
52:57 because those values are not just Hollow
53:00 successories posters is that people in
53:03 this community will kind of like seek
53:05 you out and Shuffle you around and go oh
53:07 you should go check out this little
53:09 section check out Anne Murphy's GPT jams
53:12 you should go to her event on Saturday
53:13 which by the way you should go to an
53:16 Murphy's GPD Jam tomorrow tomorrow
53:19 Saturday noon
53:23 Pacific so after you introduce yourself
53:26 scroll on down to the bottom and there's
53:28 a little Club
53:30 area and in that is this little club
53:33 called weekend GPT jams I think you have
53:36 to ask to be led into it but ask and
53:39 lets everyone in and then you just you
53:42 know be be
53:46 nice and
53:48 then check out the
53:51 information
53:52 for the event
53:55 tomorrow and it's an innovation lab
53:58 where a bunch of people are going to
53:59 talk about including uh lunatick Daisy
54:01 Thomas is going to talk about the 105
54:04 GPT she's made so if you're curious what
54:06 a GPT is if you go to the GPT Jam
54:10 tomorrow you get to hear from someone
54:12 who's made 105 of them holy [ __ ] and she
54:17 made really cool ones too like one of
54:19 them is she made the digital twin of
54:23 Alexander
54:25 Hamilton and then you know why she made
54:27 it cuz she had lost hope in
54:34 politics and for the future of this
54:37 country and so she created a GPT with
54:40 all of his writings and things that had
54:42 been written about him and had a
54:44 conversation with
54:48 him and it brought her to
54:54 tears Winston Welsh Safe Community where
54:56 people shouldn't be afraid to ask
54:58 questions nice yeah
55:00 exactly in fact in fact it's kind of
55:03 like it's it's it's almost the opposite
55:05 it's like people in here are like
55:07 excited to ask questions they're excited
55:09 to go I don't know what I'm
55:11 doing is it me or these tools janky
55:14 pieces of [ __ ] oh no they're janky
55:16 pieces of [ __ ] it's not you come on in
55:19 on my way to sign up great educational
55:22 Kind of Perfect there's an educ oh so
55:24 that's the other thing okay so there's
55:25 clubs the other club
55:27 so you should go check out uh the
55:29 weekend GPT jams that's from an Murphy
55:32 there's also in here a l Irregulars what
55:36 is that that is our club so we have our
55:40 own private space so if you want to jam
55:43 and like make jokes and make fun of me
55:45 and make sushi here we go the gummy bear
55:48 sushi shop if you want to make the gummy
55:50 bear sushi shop images drop them in our
55:53 regulars and we can have our inside
55:54 jokes and you can chat with people that
55:56 are in the community so like the chat is
55:59 going on and you can hop over there and
56:01 do that you can make
56:03 images you know jam with one
56:07 another and then the let me see what
56:10 else is there to check out here um
56:13 there's water coolers just an open
56:15 discussion showand tell is kind of an
56:16 open Show and Tell um there's a news
56:18 channel there's AI tools and resources
56:21 so AI tools and resources there's a big
56:24 deal if you're new if you're like but C
56:28 where do I start this stuff's so
56:31 complicated where do I start I talked my
56:34 buddy Dave mper so some of you know I
56:36 gave a talk back in I don't know April
56:39 or
56:40 May to creative mornings here in
56:44 Denver and Dave mper he he just stopped
56:47 stopped uh running it he was the he ran
56:50 creative mornings Denver for like 10
56:53 years um and he just he just gave it up
56:55 he just got married and had a had a a
56:56 kid and and you know he turned over the
56:58 reign of that thing but I I gave the
57:00 talk uh during during his tenure there
57:04 and so he and I just talked today and he
57:06 was like he was like he was like I you
57:08 know I've told him about AI a couple of
57:10 times and he's he's super into it but
57:12 he's busy and so he called me up today
57:13 he goes okay Kyle I think I'm ready
57:15 where do I start what do we
57:16 do and so I walked him through some
57:18 tools and frankly what I did was I just
57:21 told him the [ __ ] that's on this list
57:24 and and not even a lot of it like two or
57:25 three things on this list so what this
57:27 list is is this is a hand curated list I
57:29 curate this list
57:31 personally and this is like what's this
57:33 Mighty networks yeah it's mighty
57:34 networks
57:37 yep and mighty networks is just a
57:39 Community Management tool um it's pretty
57:42 good it's think of Mighty networks is
57:44 kind of somewhere between Discord and
57:46 Facebook something like that it's a
57:48 customized you know space for us we get
57:50 to go play there
57:52 connect um so so this list is is where
57:57 to start so the the first place you
57:58 should should start if you're just new
58:00 to this journey is chat GPT just chat
58:02 GPT that's
58:04 it and then if you can afford the 20
58:07 bucks a month pay for chat GPT plus just
58:11 do it because these three bullet points
58:15 are what you get with chat GPT free
58:18 which is still pretty [ __ ] powerful
58:21 which is why I have it
58:22 there and then for 20 bucks a month you
58:25 get all that plus all this and so what
58:29 what you should do is just go learn all
58:31 that and play with it and play play play
58:33 play play and then the rest of this list
58:35 is here's other large language models
58:37 like Claude and Gemini and perplexity
58:40 and co-pilot and pi and X or
58:44 grock llama 2
58:47 po here's the image generation tools
58:50 that list is this list is getting longer
58:52 here's the video creation tools I've got
58:54 Sora on here but Sora is not live yet
58:56 but I say that but this this is G this
58:59 has changed everything this caused Tyler
59:02 Perry to not start his $800 million film
59:08 studio so that's pretty powerful there's
59:11 music and audio there's 3D worlds and
59:13 objects and here's people you should
59:15 follow go follow Ethan mollik at oneus
59:19 thing. org and go look and and follow
59:22 him on Twitter and Linkedin just go look
59:24 at his posts over the past week it's all
59:27 about Gemini
59:29 1.5 and what the implications of a
59:32 million token context limit are and if
59:34 you're like Kyle I don't know what a
59:36 million what what's million tokens what
59:39 are tokens hey Kyle what why is there a
59:42 million is a million a good thing or is
59:43 a million a bad thing is that expensive
59:45 does it mean it cost you a million
59:46 dollars to make a thing what are you
59:48 talking about Cal million token context
59:50 window what are you talking
59:52 about just means you can put a lot of
59:54 [ __ ] in it
59:57 here's here's a good way to think about
59:59 chat gbt chat gbt is like an incredibly
1:00:03 well-educated
1:00:04 very inexperienced
1:00:08 intern and it can
1:00:11 keep like a decent 10-minute
1:00:14 conversation in its head well a million
1:00:17 token context window it can keep
1:00:20 like seven or eight major research
1:00:23 papers in its mind at the same time and
1:00:27 answer questions about all of them and
1:00:30 and draw conclusions across papers in a
1:00:35 single chat
1:00:36 session that's what Ethan mullock's been
1:00:38 talking about for the past week it's
1:00:40 it's
1:00:41 mindboggling and and he's a guy that he
1:00:44 just he talks again and again and again
1:00:46 about these tools and that you should
1:00:47 get up to speed on them and that you
1:00:49 should lean in and that if you're in
1:00:52 education stop using [ __ ] AI
1:00:54 detectors and start using in critical
1:00:57 thinking and engaging your students
1:01:00 brains stop pretending that the world's
1:01:03 most powerful knowledge tool in the
1:01:06 history of humanity is going to go
1:01:09 away if you're an education what's your
1:01:12 job to
1:01:14 to Foster and help
1:01:19 flourish um biological large language
1:01:23 models right that's your your job and
1:01:28 Along Comes This Thing Called a large
1:01:29 language model and what's the response
1:01:32 of the education Community burn the
1:01:34 witches burn the
1:01:37 witches burn the witches burn the
1:01:43 witches if you're a
1:01:47 writer I'm not going to use chat GPT
1:01:49 because it steals from
1:01:51 people you're not going to use the most
1:01:53 powerful writing
1:01:55 tool ever to be created you're not going
1:01:58 to do that
1:02:00 really if you're a writer you know words
1:02:03 you know Linguistics you know
1:02:06 context wouldn't it be curious to see if
1:02:09 this thing's any
1:02:10 good well
1:02:15 no so anyway that's that so you should
1:02:18 check that out then the other thing that
1:02:20 we got here in the in
1:02:21 AI by the way if anyone ever asks you
1:02:24 well how do you pronounce it
1:02:27 you say well duh you always lead off
1:02:30 with well duh and then you say the AI
1:02:35 fellow and if they [ __ ] their head funny
1:02:37 you go it's
1:02:39 French
1:02:42 okay um the other thing we have here are
1:02:46 guilds like there's a 101 Guild if
1:02:48 you're just getting started there's art
1:02:50 guilds and business guilds and
1:02:52 Technology guilds and social impact
1:02:55 guilds Healthcare and and education
1:02:57 there's video there's I mean um Vicky
1:03:01 Baptist and and joy py started the
1:03:04 ethics Guild like a two a week and a
1:03:05 half two weeks ago they're like we're
1:03:07 passionate about ethics can we talk
1:03:10 about these
1:03:11 issues Stacy and Mr K Matt felt the same
1:03:15 way they're like we're Educators we want
1:03:17 to start an education
1:03:20 Guild and so these are subcommunities
1:03:22 where you can find people that are into
1:03:24 the stuff you're into
1:03:26 and you can connect with them and
1:03:28 everyone's like on this same [ __ ]
1:03:30 journey and like all of these Sub sub
1:03:33 communities overlap in these kind of
1:03:34 beautiful interesting weird [ __ ] up
1:03:37 ways so anyway that's the salon. here's
1:03:41 the URL if you missed it that's not it
1:03:44 that's
1:03:49 it I mean your biggest question right
1:03:52 now should be I'm sorry how do you
1:03:53 pronounce it again the
1:03:57 [Laughter]
1:04:10 the AI learning lab I I have discovered
1:04:12 I I discovered this today on my AI
1:04:15 office hours on LinkedIn is that what
1:04:18 this channel is really about is me
1:04:20 entertaining
1:04:21 myself in the presence of
1:04:25 others
1:04:28 anyway um this is a pretty bitching
1:04:31 community so you should you should
1:04:32 [ __ ] join it the other thing you
1:04:34 should
1:04:35 do you should support this channel if
1:04:39 you can so there's a couple ways you can
1:04:40 do it follow the channel follow me
1:04:42 follow me follow me remember Oddworld
1:04:45 follow
1:04:50 me um you can share this live right now
1:04:53 I see there's a lot of carpal tunnel
1:04:55 tapping going on so
1:04:56 good on youate by the way the update
1:04:59 there is
1:05:00 23,000 23,000 love units I'm feeling
1:05:04 very loved 23,000 love
1:05:07 units so
1:05:09 far you should share this live you
1:05:11 should share it with people who are not
1:05:13 into AI but Kyle that would be
1:05:15 uncomfortable it's not going to be as
1:05:17 uncomfortable as when they lose their
1:05:18 job because they ignored this
1:05:21 [ __ ] they call you up uh listen Tad um
1:05:27 yeah you know how I uh I told you never
1:05:30 to talk about AI again or I'd punch you
1:05:32 in the face yeah uh hey could you teach
1:05:36 me about
1:05:37 that so that's going to happen to you if
1:05:41 you're into this stuff so invite those
1:05:43 people over and we'll we'll correct them
1:05:45 now so you don't have to have that
1:05:46 stupid
1:05:48 conversation um what else can you do you
1:05:51 can subscribe to the
1:05:53 live um you can do gifts you can in the
1:05:57 bottom corner one of those Corners is a
1:05:59 video series called underhyped where
1:06:02 it's a series of 18 videos I did talking
1:06:05 about how how large language models work
1:06:07 how diffusion models work why 2023 was
1:06:11 the new 1995 like how it compares to the
1:06:13 worldwide web so there's all sorts of
1:06:15 ways to support this the biggest way you
1:06:17 can support this channel though is keep
1:06:20 showing
1:06:21 up I don't know if you know this fun
1:06:24 fact some people here know this
1:06:27 um I come here every night and you're
1:06:30 like wow five nights a week I'm like no
1:06:33 seven nights a
1:06:35 week right well don't you have a life
1:06:38 don't you have a family yeah and you do
1:06:41 this why do you do this because it's
1:06:45 important so show up here well what do
1:06:48 you teach
1:06:55 nothing
1:06:59 oh how long time does it take for
1:07:02 someone that's new to this to catch up
1:07:04 to the latest and greatest and un
1:07:08 understand it to handle the different
1:07:10 tools your resonance good day your
1:07:13 resonance your resonance good day good
1:07:16 day to you sir carry on Cheerio pip pip
1:07:19 what is the frequency uh your resonance
1:07:22 carry on uh okay
1:07:29 the
1:07:30 answer for how long to catch
1:07:34 up is
1:07:37 zero and this is going to sound like I'm
1:07:40 just being a
1:07:43 smartass everyone in here including
1:07:46 myself especially
1:07:48 myself all the Irregulars all the mods
1:07:51 all the Irregulars all the people that
1:07:53 are that are in here who've been doing
1:07:55 this for a while people with phds in
1:07:59 this stuff machine
1:08:01 learning everyone right
1:08:05 now is trying to figure this stuff out
1:08:08 so as far behind as you
1:08:11 feel the fact that you're here asking
1:08:13 the question means that you are like
1:08:17 miles
1:08:19 ahead
1:08:21 of 85% of the the country and probably
1:08:26 more than that
1:08:30 globally this thing's barely taken off
1:08:33 yet it it's barely
1:08:38 started that
1:08:41 said if you really want to get up to
1:08:44 speed
1:08:46 spend on a given
1:08:50 day 24
1:08:54 hours without eating eating or drinking
1:08:57 doing nothing but Ai and do that
1:09:00 for 7 days in a
1:09:03 week no breaks No Sleep there's no
1:09:06 sleeping in Ai and then do that for six
1:09:11 months and if you do
1:09:13 that then you'll feel
1:09:19 clueless like it really it seriously it
1:09:21 doesn't matter just be in the
1:09:24 conversation
1:09:26 and you're like where do I start start
1:09:27 with chat
1:09:28 GPT start with chat GPT get GPT plus
1:09:33 understand what the different
1:09:34 capabilities
1:09:35 are understand what custom instructions
1:09:38 are understand that gp4 is connected to
1:09:41 the internet and and interacting with a
1:09:44 large language model that's connected to
1:09:46 the internet is different than one
1:09:47 that's
1:09:48 not understand that it can see images
1:09:51 it's got Vision you can upload a picture
1:09:53 of a whiteboard and have it described
1:09:56 what was on the Whiteboard text and
1:09:59 images and you could start a whole
1:10:01 creative conversation with something
1:10:02 that was on a
1:10:04 whiteboard or a whole bunch of Post-it
1:10:07 notes you know those brainstorming
1:10:10 sessions I've facilitated more of these
1:10:12 [ __ ] things in my life than you can
1:10:14 imagine where where you start out the
1:10:17 morning you're like everybody knows
1:10:18 nothing and we're going to innovate here
1:10:20 and blah blah blah blah blah and then
1:10:22 you're like okay now that we have
1:10:24 nothing mind uh here's the challenge
1:10:26 we're going to try to do and we want
1:10:27 everyone to write [ __ ] on Post-it notes
1:10:29 and then you stick them all up on walls
1:10:31 and you do like 400 of
1:10:33 those and then you organize them and
1:10:37 then they're like okay off to lunch you
1:10:39 know what happens when off to lunch
1:10:41 happens some poor
1:10:45 poor interns or new
1:10:48 employees have have to sit there with
1:10:51 laptops and type in all of those [ __ ]
1:10:55 Post-it notes
1:10:56 you can now just take a picture of it
1:10:57 upload it to chat GPT and go what were
1:10:59 these Post-it notes organize them
1:11:02 prioritize them all that sort of [ __ ]
1:11:04 just bang instantly
1:11:06 done it can make
1:11:09 images it's got a oh if you download the
1:11:13 chat GPT the official open AI app it can
1:11:16 talk and you can talk to
1:11:19 it did you know
1:11:23 that so here's chat jpt this is the app
1:11:25 this is my
1:11:27 iPad and then in the little lower corner
1:11:29 here is a little microphone you whack
1:11:34 that uh
1:11:39 wait what hang on hold
1:11:43 please they changed the app on
1:11:48 me oh it's in the upper right hand
1:11:50 corner sorry the microphone's just
1:11:52 dictation in the upper right hand corner
1:11:56 they got these little
1:12:02 headphones if I wanted to open a gummy
1:12:05 sushi restaurant how would I do that and
1:12:08 what would the recipes be for the
1:12:14 sushi opening a gummy sushi restaurant
1:12:18 sounds like a fun and creative business
1:12:20 idea here's how you could start and I'll
1:12:22 suggest some ideas for the gummy sushi
1:12:24 recipes too steps to open your gummy
1:12:26 sushi restaurant
1:12:29 so you can talk to it and it can talk
1:12:32 back oh by the way tell it you're going
1:12:35 to speak in English and you want it to
1:12:37 respond in German or Japanese or
1:12:40 Portuguese or
1:12:43 whatever it can
1:12:46 translate it's got this thing in it
1:12:48 called code
1:12:50 interpreter you're like well that sounds
1:12:52 really intimidating it kind of is but
1:12:55 it's kind of or
1:12:58 not cuz what code interpreter does is
1:13:01 what's called in the trade Advanced Data
1:13:07 analysis I'm not good with math I'm not
1:13:09 going to be doing Advanced Data analysis
1:13:11 I have a degree in acting okay I mean
1:13:14 sure I know you use computers and stuff
1:13:16 like that but I don't know what Advanced
1:13:17 Data analysis is I can tell you how to
1:13:20 break down a
1:13:22 sonnet a Shakespearean sonnet I I can
1:13:26 tell you I can tell you how to Ramic
1:13:29 pentameter from the perspective of an
1:13:31 actor I don't know [ __ ] about Advanced
1:13:34 Data analysis but what code interpreter
1:13:36 does you can upload a spreadsheet to
1:13:39 chat
1:13:40 GPT and say tell me what this is and
1:13:44 what's interesting about it and it will
1:13:46 write and execute its own python code to
1:13:49 look inside that Excel file read all the
1:13:52 data and tell you what's going on
1:13:56 and then you're like well are there any
1:13:58 Trends sure there's these Trends what
1:14:00 would you like to know uh is is the
1:14:04 revenue for this company going up or
1:14:06 going down well it went up for a while
1:14:08 and then covid happened and then it went
1:14:09 down oh could you turn that into a chart
1:14:12 and graph and then it will write python
1:14:14 code to make you a chart and graph from
1:14:16 the data that you didn't know anything
1:14:18 about it does
1:14:22 that it also I'm still on chat
1:14:27 GPT this is all in one
1:14:31 tool you can now make these things
1:14:34 called
1:14:36 gpts that are like your own little
1:14:38 personalized version of chat GPT let's
1:14:41 say you wrote a book and you wrote a
1:14:44 book on I don't know um growing almonds
1:14:46 in frigid climates I like whatever it
1:14:49 doesn't matter it doesn't
1:14:51 matter you can make a GPT say I want to
1:14:53 I want to have a GPT that people can ask
1:14:56 all the questions they want about
1:14:57 growing almonds and frigid clients and
1:14:59 I'm going to upload my book as knowledge
1:15:01 for this GPT and then you can give it a
1:15:03 little clever little name like the
1:15:04 knowledge whisper or the Almond
1:15:07 Whisperer and it'll make you a little
1:15:09 picture for your GPT of like a little
1:15:11 dancing almond in the
1:15:15 snow and you can share that with the
1:15:18 world the very first one I did I did on
1:15:21 this channel it's called Sin City
1:15:23 sipper if you if you go to chat GPT and
1:15:26 click explore gpts and go search for Sin
1:15:29 City sipper that's
1:15:32 mine it'll create an original cocktail
1:15:35 for you like the whole idea of it was
1:15:37 take a classic cocktail recipe and give
1:15:39 it a Vegas twist
1:15:43 Zing I had someone um last week Super
1:15:47 Bowl right after the Super Bowl this is
1:15:49 the coolest thing so they took advantage
1:15:52 of GPT Vision they took a picture of
1:15:55 their booze cabinet right they put all
1:15:57 their booze out they took a picture of
1:16:00 all their booze they uploaded it to chat
1:16:02 jpt and they said or to to Sin City
1:16:05 sipper my little
1:16:06 GPT and they said um look at all the
1:16:10 booze I have and come up with 50 Super
1:16:13 Bowl cocktails based on the booze I
1:16:17 have
1:16:19 what and it did it and so they took them
1:16:22 and they printed they printed out little
1:16:24 menus for for the Super Bowl guests and
1:16:28 then the guests would come up and they
1:16:30 would order whatever the thing was and
1:16:32 then behind the bar they had all the
1:16:33 recipes from this thing with all the
1:16:35 pictures and they would give them their
1:16:36 drink and they would give them a picture
1:16:38 of their
1:16:39 drink holy
1:16:41 [ __ ] that little thing I did that was a
1:16:43 brain fart in here like made someone's
1:16:47 Super Bowl party those are gpts those
1:16:50 are in
1:16:53 there there's a thing in chat CBT called
1:16:56 custom instructions where you can give
1:16:57 it a personality if you don't like its
1:17:00 personality you think it's too friendly
1:17:01 it's a little too
1:17:02 frr tell it to be a jerk and only answer
1:17:06 in three-word bullet points and it
1:17:10 will so all of that [ __ ] that I just
1:17:12 said all of that stuff I was just
1:17:15 rambling
1:17:16 there that's all just within chat
1:17:22 GPT
1:17:24 so
1:17:26 the answer to the question of you're
1:17:27 totally behind how do you catch up
1:17:29 you'll never catch up so you're you're
1:17:32 exactly where you need to
1:17:35 be if what you did is spent the next
1:17:38 month just playing
1:17:41 with those elements of chat GPT
1:17:44 understanding that it can write
1:17:46 understanding that it can analyze
1:17:47 documents understanding that it can code
1:17:51 did you know that chat GPT can code not
1:17:54 like it can show you how to
1:17:57 code it can
1:18:00 code my very not my very first probably
1:18:04 my 10th experience with chat
1:18:06 GPT my wife was watching some crappy
1:18:09 show rather than getting a fight over
1:18:12 the remote I thought I know I'll ignore
1:18:15 the TV show my good husband and I got on
1:18:19 my laptop and I was playing on chat
1:18:23 je and I remembered someone going oh it
1:18:26 can code and and so I said write me a
1:18:28 line of you know write me a some python
1:18:31 code write me a python script I forget
1:18:33 what I
1:18:34 said and it wrote the code and I was
1:18:36 like ah it's very impressive and then I
1:18:39 realized uh I don't know how to run
1:18:41 python
1:18:43 code so normally if you're in a
1:18:46 situation where you're like oh I'm all
1:18:47 excited about learning technology I'm
1:18:50 going to learn to program Python and
1:18:51 then you look at the python code and
1:18:53 you're like I'm not going to learn to
1:18:54 program py python I'm just I'm good I'm
1:18:57 going to I'm going to go have a
1:19:00 Snickers so I had that feeling and
1:19:02 rather than going to get the
1:19:04 Snickers I said hey Chad GPT I don't
1:19:06 know how to run python code it say OD
1:19:08 not a problem you just run your terminal
1:19:10 on your Mac and you put in this command
1:19:12 line and and before it could finish I
1:19:13 stopped it and I'm like H I don't want
1:19:15 to do command line I'm a Mac guy I want
1:19:17 I want just make it easy can you make is
1:19:19 there a website I can run this python
1:19:21 code sure here's repet and there it gave
1:19:23 me four other sites but I started with
1:19:24 repet which is the big
1:19:26 one and so I copied and pasted my python
1:19:29 code into repet and I hit play I'm like
1:19:31 here we go error it's like
1:19:36 [ __ ] so I copied the error and I went
1:19:39 dear chaty PT your stupid code didn't
1:19:42 work pasted in the error I didn't know
1:19:44 what the error said and it said um oh
1:19:48 well you don't have this Library
1:19:49 installed and then you know again I'm
1:19:51 like okay if I don't know how to run
1:19:53 python code you think I know how to
1:19:54 install a library you big dumb Dum so I
1:19:57 said I don't know how to do that so it
1:19:59 said in the lower leftand corner of the
1:20:01 screen click on this button so I go over
1:20:03 to repet there's that button I click on
1:20:05 it it said search for this library in
1:20:07 the upper right hand side of the screen
1:20:08 there's a little search bar I search for
1:20:10 it and it said hit install that library
1:20:13 and I did that and I hit the Run button
1:20:15 and it ran I was like holy
1:20:17 [ __ ] it
1:20:19 worked within 90 minutes I had written a
1:20:22 fully functional python application that
1:20:25 was making API calls to open AI to
1:20:28 generate original social media content
1:20:31 based on some
1:20:33 inputs in 90
1:20:35 minutes I've never written a line of
1:20:37 python in my life I still
1:20:39 haven't a fully functional python
1:20:45 application from nothing not knowing
1:20:50 python chat GPT is not just Fancy
1:20:54 Google it is a profoundly different
1:20:58 tool so just
1:21:01 play play do anything with it make up
1:21:05 menus for
1:21:07 the the what what's coming up I don't
1:21:10 know
1:21:12 hockey the
1:21:15 playoffs
1:21:18 um have it explain all the vitamins that
1:21:21 you should be
1:21:23 taking what
1:21:26 anything just play with it and just play
1:21:29 play play play
1:21:33 play and then here's my other hack you
1:21:35 want to know my other
1:21:37 hack do your regulars know this one
1:21:39 thank you very much Tracy that was
1:21:41 really
1:21:43 cool here's my other
1:21:48 hack tell people about
1:21:51 it but Kyle I'm cluess I don't know
1:21:53 anything tell people about it hey I just
1:21:57 went to chat GPT for my very first time
1:21:59 and I typed in make me a poem about
1:22:02 gummy Sushi and here's my poem about
1:22:04 gummy
1:22:06 Sushi well where do I put that put it on
1:22:09 [ __ ] LinkedIn but it's a stupid poem
1:22:12 about gummy Sushi I put you want to see
1:22:15 what I put on
1:22:22 LinkedIn you you you want to you want to
1:22:25 lead with dum dum you follow me I will
1:22:27 take you to the promised land Linked In
1:22:31 let's see
1:22:33 Linked
1:22:36 In let me see if I can find it now this
1:22:38 is going to be fascinating let's see
1:22:41 view
1:22:42 [Laughter]
1:22:42 [Music]
1:22:49 profile show all
1:22:52 posts just going to be so good you're
1:22:54 going to love it
1:22:55 you're going to love love love love love
1:22:57 love love it love love love love love
1:23:00 love love love love love love love love
1:23:03 can I search me is there a little search
1:23:06 no why would there be that why would you
1:23:09 do that Microsoft why why have it be
1:23:12 functional
1:23:13 why why have it be
1:23:16 functional why have this be useful to
1:23:19 anyone at all here
1:23:23 this
1:23:25 I'm like I'm here for the game they're
1:23:27 like so I come down to the field to play
1:23:29 the game and they're like thank you very
1:23:30 much Robert Rossy what are you doing let
1:23:32 me start this over I'm like I'm here for
1:23:33 the game they're like Frank down to the
1:23:35 field to play the game and they're like
1:23:37 hey Frank Frank what are you doing I'm
1:23:40 like I'm here for the game they're like
1:23:41 Frank it's Thursday there's no game
1:23:43 today I'm like you got to be [ __ ]
1:23:45 kidding me there's no game today I'm
1:23:48 ready for the game they're like Frank
1:23:49 Frank you look [ __ ] delicious I'm
1:23:52 like hey hey hey back up hey whoa whoa
1:23:54 whoa whoaa fella get your hands off my
1:23:57 bun oh my goodness so so I leave I go
1:24:01 down to the the Piggly Wiggly there I
1:24:02 get myself a soda I come back they're
1:24:04 like they're like they're like Frank uh
1:24:06 we we got an issue like what's that it's
1:24:08 like Tommy's here oh what do you mean
1:24:11 Tommy's here yeah Tommy's here what's
1:24:14 going on with Tommy Tommy's hungry oh no
1:24:17 no no no no Frank it's going to be okay
1:24:22 it's not going to be okay you know T
1:24:24 doesn't stop [ __ ] eaten oh my oh
1:24:26 Tommy Tommy no don't oh God oh
1:24:31 God I put that on LinkedIn I put that on
1:24:35 LinkedIn you're like well that's the
1:24:37 dumbest thing I've ever seen exactly now
1:24:39 why did I put that on
1:24:48 LinkedIn one of the reasons I put that
1:24:50 on LinkedIn is because I'm not the
1:24:52 brightest bulb on the tree
1:24:56 but the other reason
1:25:04 is so many
1:25:06 people do
1:25:08 not know what AI is they don't know what
1:25:12 chat GPT is they think that
1:25:17 um you have to have an engineering
1:25:20 degree to do it they think
1:25:23 that the robots are going to kill us
1:25:26 they think that it's going to take their
1:25:27 jobs they think
1:25:29 that we have no power we have no
1:25:32 agency that it's that it's fatalistic to
1:25:35 even try or it's going to be a lot of
1:25:37 work or it's going to be boring or like
1:25:40 I don't want to go back to school I've
1:25:42 I've already had a career for 30 years
1:25:44 and I'm kind of tired my hairs are gray
1:25:46 and you know they do oh love tough tough
1:25:49 and
1:25:51 stupid and so I put up stupid [ __ ] like
1:25:53 that because maybe someone will go
1:25:55 that's funny oh is that
1:25:57 AI oh I didn't I didn't know you could
1:26:00 make animations is that like how they do
1:26:02 s South Park no I just talked and it did
1:26:05 that oh how how'd you do
1:26:10 that and you know how I learned how to
1:26:12 do that someone in here said hey Kyle
1:26:16 did you know you could do a
1:26:19 thing and so I made a stupid pencil and
1:26:23 then I made Frank and I put them both on
1:26:26 LinkedIn so as you're learning this
1:26:28 stuff even if you're just like fig just
1:26:30 figuring it out share it and you don't
1:26:33 have to share good stuff you can go hey
1:26:35 I went to do this thing and it got it
1:26:37 all wrong I'm going to share that with
1:26:39 you because now I'm
1:26:40 confused I thought this was supposed to
1:26:42 be the most powerful tool in the history
1:26:44 of humanity uh the professor at the AI
1:26:46 learning lab told us that but now it got
1:26:49 all this stuff wrong so chat jpt is a
1:26:52 dumb
1:26:53 Dum and then someone out there will go
1:26:55 actually that's called hallucinations
1:26:57 and that's just part of the deal right
1:26:59 now and here's how you work around that
1:27:00 and you use your critical thinking and
1:27:02 you check your work and you don't just
1:27:04 be lazy oh
1:27:07 okay and then if you use this tool over
1:27:10 here called perplexity it'll actually
1:27:13 provide a citation for where it found
1:27:15 the answer for you and you can go check
1:27:16 the work really easily oh that's cool
1:27:19 and you start to piece part all these
1:27:21 little it's like there's like lots of
1:27:23 little scraps of paper you know all over
1:27:26 a football
1:27:27 field and our job is sort of piece
1:27:30 together single 8 by 10 8 and 1 half by
1:27:33 11 pages and we're just and and like
1:27:38 like most of the people are in the
1:27:40 stands but there's like 40 of us on the
1:27:44 field that that's that actually this is
1:27:46 a really good metaphor this is a
1:27:49 generative AI right now is this we're
1:27:52 we're in a a football stadium that can
1:27:55 seat 60,000 people there's 40 of us on
1:27:58 the field the field is littered in
1:28:01 little quarter inch pieces of torn up
1:28:05 paper and our job is to assemble an 8
1:28:08 and 1 12 by 11 piece of paper and we're
1:28:11 running around and we're comparing our
1:28:13 does yours have an n on it is Oh that's
1:28:16 oh mine's got an R oh oh no wait it's an
1:28:18 e wait what okay it's an e mine's an e
1:28:21 right we're all kind of piecing them
1:28:23 together and and then someone someone's
1:28:25 walking around with I got some tape do
1:28:27 you want do you want to tape those to oh
1:28:29 you got tape oh that's awesome I've just
1:28:31 been holding them here in my palm try
1:28:32 not to have the wind blow them
1:28:34 away that's it right now that's the
1:28:38 experience of being AI literate right
1:28:41 now is you're on the field trying
1:28:44 desperately to find enough pieces to put
1:28:46 together a piece of
1:28:51 paper and sitting up in the stands going
1:28:56 yeah yeah robots are going to kill us
1:28:59 I'm not interested in that that doesn't
1:29:01 that doesn't look like fun you know
1:29:04 didn't didn't I used to play football
1:29:05 here why are we here what what is is
1:29:08 this a game is this
1:29:12 competitive and then and then some of
1:29:14 the fans are like you know all that
1:29:18 paper is evil paper that was stolen from
1:29:23 people they they're up in the stand
1:29:27 shouting AI bad AI bad people on field
1:29:32 bad me
1:29:37 good
1:29:40 meanwhile the advances keep on coming
1:29:44 the advances keep on
1:29:46 coming Tyler Perry
1:29:48 sees what open ai's new video model does
1:29:52 and decides not to open an 800 million
1:29:54 Film Studio because he recognizes we're
1:29:58 not going to need
1:30:01 to now there's a depressing side of that
1:30:04 which is oh [ __ ] all that economic
1:30:07 growth is
1:30:08 lost there's a really exciting side of
1:30:10 that someone's going to figure out how
1:30:13 to get good enough with AI to make some
1:30:15 [ __ ] for a Tyler Perry
1:30:22 movie and it might not be a filmmaker
1:30:26 it might be someone in here it might be
1:30:31 you but you got to be on the field right
1:30:33 now cuz when the people in the stands
1:30:37 start
1:30:39 coming the the people that are on the
1:30:41 field massive
1:30:44 opportunity but then then the game
1:30:46 changes it's not as much fun I can tell
1:30:48 you that from the 90s from 1994
1:30:51 until mid 96 was a [ __ ] Blast 2 and
1:30:56 1/2 years pure pristine people trying to
1:31:00 figure the [ __ ] out how are you doing it
1:31:02 I don't know how are you doing it oh
1:31:03 that's cool oh you got that big client
1:31:05 oh we got this big client oh that's so
1:31:07 cool yeah yeah two and a half years then
1:31:09 the money showed up the technology
1:31:11 started to
1:31:13 stabilize and then it turned into
1:31:15 business
1:31:17 right and then you have just like this
1:31:19 frothy Insanity until the dot bubble and
1:31:23 everyone goes oh the internet's dead and
1:31:25 and any of us who were in it earlier
1:31:27 like no it's not dead at
1:31:29 all just the people that weren't on the
1:31:32 field are saying it's dead but if you
1:31:34 were playing the game you knew that it
1:31:36 wasn't going anywhere it didn't go
1:31:37 anywhere it's got bigger it's got bigger
1:31:39 and bigger and bigger and
1:31:41 bigger a bunch of companies that should
1:31:43 have died
1:31:46 died and the internet carried on that's
1:31:48 going to be
1:31:50 AI we're so early so early see what
1:31:55 happens when you ask a simple question
1:31:56 in here you get an hour of me [ __ ]
1:32:03 rambling thank you
1:32:05 K thank you for the gift I appreciate
1:32:08 that okay
1:32:09 so you want to support this channel
1:32:12 follow it follow follow my channel
1:32:15 follow follow follow me subscribe to the
1:32:19 lives buy the video series in the
1:32:23 corner show up here whenever you
1:32:27 can some
1:32:30 people I call them weirdos they call
1:32:34 themselves Irregulars they show up here
1:32:36 every [ __ ]
1:32:38 night they're like if you're going to
1:32:41 we're going to I'm like
1:32:45 okay thank you Sherry D Sherry D is one
1:32:48 of
1:32:51 them Sherry d
1:32:54 been in the nursing profession for three
1:32:58 decades you know what she's doing now
1:33:02 she's making music
1:33:06 videos how'd she do that did did she she
1:33:09 went back to film school
1:33:12 no she just figured some [ __ ] out she's
1:33:14 out on the field picking up little teeny
1:33:16 tiny piece thank you Silver Fox teeny
1:33:19 tin and pieces of paper she goes oh if I
1:33:20 make a picture with this one and then
1:33:22 put it in this thing it kind of makes it
1:33:23 twirl and huh if I stick a few of those
1:33:26 together and then oh so someone someone
1:33:29 saw me posting on LinkedIn or Facebook
1:33:32 that I make pretty pictures and [ __ ]
1:33:35 that I used to go to high school with
1:33:38 and they've got a band and they've got a
1:33:40 song and they want to turn that song
1:33:41 into a video hey Sherry could you make
1:33:44 us a video and she
1:33:47 went oh
1:33:50 [ __ ] I'm not an expert I'm not really
1:33:53 good at this
1:33:55 okay and she made this killer [ __ ]
1:33:59 video it's
1:34:01 brilliant and now she's made a bunch she
1:34:04 she did one for the the Gen AI um the
1:34:07 runway ml gen AI competition that was
1:34:11 just
1:34:14 awesome she's out on the field
1:34:17 discovering a whole new source of joy
1:34:21 and
1:34:23 self-expression that a year
1:34:27 ago probably wasn't even in Her
1:34:31 Imagination right let let alone her
1:34:34 dreams
1:34:36 right you don't just like all of a
1:34:38 sudden go I'm just going to make music
1:34:40 videos now and then like a week later
1:34:42 you've done
1:34:45 it
1:34:47 except now you
1:34:50 can it is my contention that 20 23 was
1:34:55 the last normal
1:34:59 year and it was a little freaky but 2024
1:35:03 things are going to start accelerating
1:35:05 and it's going to get weirder and
1:35:06 weirder and
1:35:08 weirder and if you're in the game if
1:35:11 you're in the conversation as these
1:35:13 tools level
1:35:15 up you'll know ooh there's a bigger
1:35:19 piece of paper on the field as the tools
1:35:21 get better the clues get bigger
1:35:25 oh I can do this with it now o I can do
1:35:27 that with it now ah that thing I thought
1:35:29 I had to learn this prompt engineering
1:35:31 thing I actually if I use these tools
1:35:33 over here they do the prompting for me
1:35:36 and I can just prompt with images and
1:35:40 audio I'm good with pictures and I can
1:35:42 talk this is
1:35:48 great so the time to get in the
1:35:51 conversation is
1:35:55 now 2024 Year of the Dragon Let's go ask
1:35:59 chat jie Tai about the Year of the
1:36:02 Dragon Frank there's Frank Love
1:36:05 Frank so I come down to the field to
1:36:07 play the game and they're like hey Frank
1:36:09 hey Frank Frank what are you doing I'm
1:36:11 like I'm here for the game they're like
1:36:12 Frank it's Thursday there's no game
1:36:15 today I'm like you you got to be [ __ ]
1:36:17 kidding
1:36:23 me
1:36:25 oh okay let's
1:36:27 ask Ye
1:36:32 [Music]
1:36:34 Old if 2024 is Year of the Dragon and
1:36:40 2024 is the
1:36:42 year that we're going to hit
1:36:44 [Music]
1:36:47 AGI talk to me about the meaning of year
1:36:51 of the dragon in that context
1:37:02 you're the dragon according to Chinese
1:37:03 zodiac is associated with strength
1:37:06 Health Harmony and good fortune dragons
1:37:08 are symbols of power boldness and
1:37:10 success they're believed to bring about
1:37:13 moments of great change and
1:37:16 transformation in the context of
1:37:18 achieving artificial general
1:37:19 intelligence AGI in 2024 the symbolic
1:37:23 attributes of the Dragon year could be a
1:37:25 partic seen as particularly
1:37:27 fitting all of you logical scientists
1:37:31 with your not believing in astrology and
1:37:33 the planets look this I this is these
1:37:38 are
1:37:40 facts these are facts this is proof this
1:37:45 is you know what this is
1:37:48 data this is not Hocus Pocus this is
1:37:51 data a probability engine generated
1:37:54 these words and put them here and when I
1:37:57 read them you understand them in your
1:37:59 large language model thinky think of a
1:38:01 brain AGI represents transformative Le
1:38:04 leap in the field of artificial
1:38:06 intelligence marking the point at
1:38:09 which machine intelligence is capable of
1:38:13 understanding learning and applying
1:38:15 knowledge across a wide range of tasks
1:38:18 or at or Beyond a human level let me
1:38:22 read that again artificial general
1:38:27 intelligence represents a transformative
1:38:29 leap in the field of uh making making it
1:38:32 wait marking the point at which machine
1:38:34 intelligence will be capable of
1:38:36 understanding learning and applying
1:38:39 knowledge across a wide range of tasks
1:38:42 at or
1:38:44 Beyond human
1:38:49 level this
1:38:52 year not like it can write a
1:38:56 paper more like it can do your
1:39:00 job this
1:39:03 year now will we get there I don't know
1:39:06 Pate doesn't think so P didn't here he
1:39:08 works at Google David Shapiro who I
1:39:10 follow does he thinks it's going to
1:39:12 happen by September
1:39:14 October Sam
1:39:16 mman I think
1:39:19 probably has some
1:39:22 confidence it seems
1:39:26 plausible or possible let's call it that
1:39:30 Sam Alman was fired because they
1:39:33 stumbled upon something that was so
1:39:35 powerful it freaked them the [ __ ]
1:39:39 out there was a leaked letter to the
1:39:41 board for this thing called qar if you
1:39:43 haven't heard about
1:39:47 it and qar did some
1:39:52 [ __ ]
1:39:55 so oh wait let's go back to our
1:39:59 thing the pursuit of AGI is not just a
1:40:02 technical Milestone it Heralds a new era
1:40:05 era of innovation the potential to
1:40:07 fundamentally alter how we live and
1:40:09 interact in the digital world the
1:40:10 qualities associated with the dragon
1:40:12 power ambition and the capacity to
1:40:14 affect significant change mirror the
1:40:16 potential impacts of reaching AGI see
1:40:18 people this isn't Hocus Pocus this is
1:40:21 science however the the Year of the
1:40:24 Dragon symbolism also cautions us to
1:40:26 consider the Dual nature of significant
1:40:30 Transformations here it comes the robots
1:40:32 are going to kill us just as dragons can
1:40:35 represent prosperity and auspicious
1:40:37 power they can also embody formidable
1:40:39 changes and the need for wisdom in
1:40:42 harnessing such
1:40:48 power
1:40:50 yikes the journey toward AGI similarly
1:40:53 invokes questions of ethical use
1:40:55 governance and societal impacts of such
1:40:58 profound technological advancements
1:41:00 absolutely and here's my thought on
1:41:06 that that shit's scary
1:41:10 right that these tools can get so
1:41:12 powerful that we could lose control of
1:41:14 them they could do stuff on their
1:41:19 own there's two ways to interact with
1:41:23 this thing that's coming whether you
1:41:24 want it or not there's only
1:41:30 two you can be in the Cozy beach
1:41:35 hut or you can be out in the scary waves
1:41:39 on your
1:41:41 surfboard you can to be on the sidelines
1:41:43 or you can be on the
1:41:46 field for me I want as many people on
1:41:50 the field as possible and quite frankly
1:41:52 I want as many people
1:41:55 who were not historically in Machining
1:41:57 learning on the field as possible I want
1:42:00 people playing with AI that are
1:42:03 ethicists and Educators and back drivers
1:42:08 and Self
1:42:09 Storage property managers Jim Ross I
1:42:12 don't know if you're in
1:42:14 here I want the Jim Rosses I I want as
1:42:17 many of those people on the field as
1:42:19 possible that's why I do this every
1:42:22 night
1:42:28 because if if us participating and
1:42:31 bending the use of AI even slightly
1:42:34 toward the good like ever so
1:42:41 slightly if you know anything about long
1:42:44 ocean
1:42:46 voyages if you're going like
1:42:50 this you hit Africa if you're going like
1:42:57 this you smash into
1:43:00 Greenland just a teeny little shift
1:43:04 toward the positive right now I think
1:43:07 has a disproportionately positive effect
1:43:09 in the
1:43:10 future so it's not all out of our
1:43:12 control I don't believe that for a
1:43:15 minute now is it all in our control
1:43:18 no does that make it scary yes but if
1:43:22 you're on the surfboard at least you
1:43:23 know where the boundaries are how many
1:43:26 pieces of paper on the field how
1:43:28 overwhelming it
1:43:30 is as things start to happen you're like
1:43:32 okay I can I know it's happening here
1:43:34 you can maybe navigate a little bit do
1:43:36 some Consulting realize that the job
1:43:39 market is completely getting shaken up
1:43:40 but you're actually in this little
1:43:41 bubble over here where people like hey
1:43:43 you know something about this would you
1:43:44 tell me about this I'll give you money
1:43:45 for that you can make
1:43:49 money oh thank you very much
1:43:51 K the singing frogs were very
1:43:56 cute in some the alignment of achieving
1:44:00 AGI in a dragon year could be as
1:44:02 symbolically potent wait could be seen
1:44:05 as symbolically potent suggesting a
1:44:08 period of groundbreaking advancements
1:44:09 and challenges marked by the powerful
1:44:12 and transformative nature of both the
1:44:14 dragon and the leap toward AGI it
1:44:17 underscores the importance of
1:44:19 approaching the frontier with balance
1:44:21 ambition and wisdom
1:44:24 that the dragon
1:44:26 symbolizes you heard it here first and
1:44:28 again this is what what they call in the
1:44:31 business this is a white paper so so
1:44:34 like well it doesn't have any research
1:44:37 behind it and there's no facts and and
1:44:39 there's no mathematics and but it's if
1:44:43 you print it on white
1:44:46 paper then it's a white
1:44:49 paper and it's peer reviewed I read it
1:44:52 here and you listen to it you're my
1:44:56 peers you
1:44:59 see you're
1:45:02 welcome oh I just pled a dragon toward
1:45:05 Show and Tell see this is why we do this
1:45:08 this is a jam session I [ __ ] love
1:45:10 this community it's like it's like he he
1:45:13 talked about something weird let's go
1:45:14 make
1:45:20 it Dragon doing what dragons do when
1:45:23 one's watching that's
1:45:28 awesome oh that's a cool one nice very
1:45:31 nice Danielle is that Danielle
1:45:35 yeah Claire Jacobs let's see
1:45:40 gorgeous gorgeous gorgeous gorgeous Lear
1:45:43 love drink the Kool-Aid hey
1:45:46 now you dumb Dums nuclear sushi
1:45:51 bar when the gummy SU sushi chef has had
1:45:58 enough oh my
1:46:00 goodness oh those are cool is that what
1:46:03 is that are those gummy sushis not gummy
1:46:06 but those are cool
1:46:09 Roger that's
1:46:10 great gummy
1:46:15 Sushi oh man all right cool that's
1:46:19 awesome I did it before you said dragon
1:46:22 oh that's funny Cory Sandler that's
1:46:24 awesome that's really
1:46:27 good oh man all right they see my robot
1:46:31 and Snickers mohawk hair
1:46:33 too that's great can we see the zombie
1:46:37 Jonathan tonight
1:46:42 sure
1:46:44 sure sure let's
1:46:49 go one of our one of our uh AI learning
1:46:54 lab zoo animals is the turtle I think
1:46:56 it's
1:46:57 the first of the AI learning lab zoo
1:47:01 animals oh life in 2024 from David
1:47:07 Shapiro oh and and day one F1 testing
1:47:10 highlights oo
1:47:14 sassy I like turtles front village with
1:47:19 my friend the zombie Jonathan you're
1:47:21 looking good Jonathan just got an
1:47:23 awesome some face paint job what do you
1:47:24 think I like turtles all right you're
1:47:27 great zombie good times here at the
1:47:30 waterfront Village open for the next 11
1:47:35 days allting the comes back reveral
1:47:40 thriving and growing together
1:47:42 [Laughter]
1:47:50 stop I'm telling you man
1:47:55 they that
1:47:58 video you should just train an AI just
1:48:00 on that video for comedic
1:48:03 timing and and inflection just every
1:48:07 single thing about that video is
1:48:11 perfect there is not there is not a
1:48:15 millisecond out of
1:48:18 timing it's the perfect premise the
1:48:22 perfect setup
1:48:25 and the perfect
1:48:26 characters oh my God it's it cracks me
1:48:29 up every [ __ ]
1:48:35 time stay-at-home mom's too yes
1:48:38 stay-at-home mom's going to kill it with
1:48:39 AI
1:48:41 um it was AI being more powerful than
1:48:44 nuclear war in our very own
1:48:47 hands oh that's
1:48:49 interesting Jim Ross business has picked
1:48:52 up Jim Ross
1:48:54 from from WWE
1:48:58 said right on like you're thinking thank
1:49:01 you
1:49:04 Bill show personality
1:49:07 customization chat pt4 says it would be
1:49:10 nice but denies it's possible oh
1:49:14 sure show personal personality
1:49:17 customization
1:49:20 okay okay so in chat GPT you can do this
1:49:23 with chat GPT 3.5 or chat GPT 4 you
1:49:26 don't need um you don't need chat GPT 4
1:49:30 to do this I'll do it and for just CU
1:49:32 because it's there but you do it the
1:49:33 same
1:49:34 way
1:49:38 so the way chat GPT works is the folks
1:49:43 at open AI this is wild
1:49:47 actually they essentially wrote a
1:49:51 prompt so you know how you write a
1:49:53 prompt make me a poem about the War of
1:49:55 1812 and and then it does its
1:49:59 [ __ ] the folks at open AI wrote a prompt
1:50:02 to say act like chat
1:50:06 GPT and so somewhere in their prompt
1:50:08 they just said like you know be friendly
1:50:10 be supportive you know don't do these
1:50:12 things if they ask do do those things if
1:50:14 they ask they it's not like they
1:50:17 programmed it programmed it I mean they
1:50:18 programmed the front end of cat gbt but
1:50:21 how it behaves is a
1:50:24 prompt and so one of the things that
1:50:27 they did about 6 months ago was they
1:50:31 revealed that functionality to us so
1:50:34 customize chat
1:50:36 GPT and so when you go into customize
1:50:39 chat
1:50:41 GPT there's what would you like chat GPT
1:50:44 to know about you to provide better
1:50:47 responses and then there's how would you
1:50:49 like chat GPT to respond so if you turn
1:50:52 this on
1:50:54 let's see what I have you have no P
1:50:57 patience and just want the answers as
1:50:59 efficiently as
1:51:00 possible and this is what I wrote two
1:51:03 nights ago I think and then respond to
1:51:06 every inquiry with a concise single
1:51:09 sentence answer followed by short bullet
1:51:12 points only use bullet points if the
1:51:14 simple answer isn't enough so I'm going
1:51:16 to turn this off for a
1:51:19 second close out of here and I'm going
1:51:22 to say say
1:51:24 um write me a
1:51:29 paragraph
1:51:32 describing uh let's
1:51:36 see the infield fly
1:51:42 rule and so chat GPT will the infield
1:51:46 fly rule is regulation and baseball
1:51:48 designed to prevent unfair play during
1:51:50 the and there are Runners on the thing
1:51:52 when the rules
1:51:53 invoked this prevents the Fielding
1:51:56 team it's a rule that places emphasis on
1:51:59 fairness and sportsmanship ensuring that
1:52:01 the defense cannot exploit a situation
1:52:04 so it's it's a lot right it's it's a lot
1:52:07 and if you're
1:52:08 like chat GPT is just too chatty I
1:52:11 didn't I didn't pay for chatty GPT I
1:52:14 want concise
1:52:16 GPT so go to customize chat gbt and just
1:52:21 tell it that you have no patience and
1:52:24 just want the answers as efficiently as
1:52:26 possible respond to every inquiry with
1:52:29 concise single sentences followed by
1:52:31 short bullet point save okay so now I'm
1:52:33 going to go copy this exact
1:52:36 prompt note that paragraph right there's
1:52:39 the
1:52:40 paragraph new chat same
1:52:51 prompt
1:52:55 infield fly rule is a
1:52:58 baseball in baseball is a regulation
1:53:00 designed to prevent infield infielders
1:53:02 from int intentionally dropping the Fly
1:53:05 ball allowing them to make easy double
1:53:06 or triple plays it applies when there
1:53:08 are fewer than two outs the rule is
1:53:10 invoked right so short single sentence
1:53:13 short bullet points there you go
1:53:17 done you could also do silly things with
1:53:21 it but it's illustrative and we this
1:53:25 didn't work the other night I don't know
1:53:26 what was going on but let's see you are
1:53:30 a New York City
1:53:34 comedian and find everything in let's
1:53:39 see
1:53:41 find jokes in every
1:53:46 situation
1:53:49 uh how do you want it to respond um
1:53:54 respond to every
1:53:57 inquiry with a
1:54:06 joke a cutting
1:54:12 comment um or a roast of the
1:54:20 person
1:54:21 asking
1:54:24 safe all
1:54:26 right new
1:54:28 chat explain thefield fly
1:54:34 rule ah the infield fly Ru the baseball
1:54:37 equivalent of explaining to your date
1:54:39 while you still live with your
1:54:51 parents I I know this
1:54:55 guy oh it's all about avoiding picture
1:54:58 this SUNY day bases are loaded less than
1:55:00 two outs and the banner pops up batter
1:55:02 pops up a Fly ball could easily now
1:55:05 without the infield fly rule infielders
1:55:07 could let the ball drop on purpose
1:55:09 turning what should be a routine play
1:55:10 into a double or triple sneakier than a
1:55:12 magician palming your
1:55:14 watch so the rule steps out so the rule
1:55:18 steps in like a chaperone at prom
1:55:21 calling the batter out automatically to
1:55:24 prevent these Shenanigans right so
1:55:26 totally different
1:55:29 personality
1:55:31 so okay you can make it a you know an
1:55:36 impatient person that doesn't like to
1:55:38 talk you can make a comedian what else
1:55:40 can you make it well you could you could
1:55:45 um you could give it an example of your
1:55:48 writing and say when you respond only
1:55:50 use my tone of writing only use you know
1:55:53 use this as
1:55:55 tone for the
1:55:57 writing and I love bullet points and I
1:56:00 love uh you know titles and sub
1:56:05 subtitles and I want all my answers to
1:56:08 be well
1:56:10 structured and then your chat GPT will
1:56:12 behave completely differently for all
1:56:17 chats crazy right my IT Tech husband
1:56:22 doesn't support AI I'm a real estate
1:56:25 agent and love it investor friendly
1:56:27 agent you are going to have so much fun
1:56:29 with
1:56:32 him I really think it's this is the
1:56:35 second I don't were were you the same
1:56:37 one that was saying it the other night I
1:56:38 I think there was someone else in here
1:56:40 saying their husband like I don't like I
1:56:43 don't like
1:56:44 Ai and I'm like I I think that the the
1:56:47 significant others of the of the D it
1:56:50 you should get better in a week or two
1:56:53 at your husband's job than your
1:56:58 husband and just when he actually this
1:57:01 would be good he comes from home from
1:57:03 work and he [ __ ] about some problem
1:57:05 having to whip up a proposal for a
1:57:06 client and and then just like while
1:57:09 you're watching TV whip up a proposal
1:57:11 for the client and then print it out and
1:57:13 just hand it to
1:57:17 him oh man I have a chat GPT that gives
1:57:21 wrong answers only and posts about how
1:57:23 smart it is that's
1:57:25 awesome it's perfect lean into the
1:57:29 hallucinations Force the
1:57:31 hallucinations awesome all right listen
1:57:33 I got to go people it's late it's Friday
1:57:35 night Friday night date night it's been
1:57:38 been good hanging hanging with y' all
1:57:40 tonight um couple of
1:57:44 things um one of
1:57:49 the well so if you don't know about the
1:57:52 AI Salon you should know about the AI
1:57:54 salon so this is the
1:57:57 URL if you weren't here
1:58:01 earlier go
1:58:03 there and click on the first link the
1:58:06 salon. a click on the first link to the
1:58:09 AI Salon community and and when you go
1:58:11 to welcome to the salon there's a whole
1:58:13 page that talks about who we are why we
1:58:15 do what we do and what our values are
1:58:18 check that out and if you dig the values
1:58:19 join this group join this group group
1:58:23 it's
1:58:25 free it's remarkable the people in here
1:58:28 are remarkable so be remarkable and come
1:58:33 join one of our
1:58:35 values is
1:58:39 gratitude
1:58:41 so what I like to do here at night is
1:58:43 thank people so to the Irregulars who
1:58:46 show up here night after night after
1:58:49 night and support and celebrate this
1:58:53 Insanity of a
1:58:55 show and are like remarkably generous in
1:59:00 how you're learning and how you're
1:59:02 teaching and how you're just showing up
1:59:04 for each other thank you um to the mods
1:59:08 who sift through all the insanity of the
1:59:11 comments you [ __ ] weirdos with your
1:59:14 comments and they there's actually
1:59:16 sifting through those and saying oh
1:59:18 here's an interesting one Kyle why don't
1:59:19 you talk to this one so like that whole
1:59:21 hourong I went on that was cuz one of
1:59:23 the mods
1:59:26 posted how long is it going to take me
1:59:28 to catch up that's a really good
1:59:32 question with an hourlong
1:59:35 answer because you can't catch up you're
1:59:38 already caught up everyone's trying to
1:59:40 figure it out thank you Lord digital
1:59:43 Gods thank you to Lord digital gods for
1:59:45 Rose after ruse after Rose after
1:59:49 Rose ruse by now the name would smell is
1:59:54 sweet so thank you to that thank you to
1:59:56 everyone that just showed up tonight on
1:59:58 a Friday night to listen to an
2:00:00 insane oldish man on Tik Tok for
2:00:06 hours that's remarkable if you don't
2:00:10 think I don't wait if you think I don't
2:00:13 appreciate it you're wrong I really
2:00:15 really really appreciate you guys
2:00:17 showing up thank you Robert Rossy thank
2:00:19 you
2:00:21 Tom an artist yeah thank
2:00:29 you so if you if you like what's going
2:00:31 on here if you like the channel if you
2:00:33 like this idea of I think I might want
2:00:35 to try to figure this AI [ __ ] out I
2:00:36 think I might want to come out of the
2:00:38 stand and get on the field and start
2:00:40 piecing together little pieces of
2:00:43 paper that's this is that's what this is
2:00:45 about so just show up
2:00:50 here when I do
2:00:54 it's it's
2:00:56 uh 8:00 Mountain most most
2:01:00 nights unless I got some [ __ ] going
2:01:05 on and then you can follow the channel
2:01:09 That would be appreciated you could
2:01:10 subscribe to the lives that really helps
2:01:12 you could buy the video series in one of
2:01:14 the corners there that'll tell you why
2:01:16 this AI shit's actually such a big
2:01:20 deal you can do the tapy taps
2:01:25 46,4 okay you've got it's 10:01 you you
2:01:30 have three minutes to hit 50,000 no
2:01:32 sorry 1 minute to hit
2:01:35 50,000 so I'll give you till 10:03 it's
2:01:38 it's a minute and change all right you
2:01:40 just oh wait we're 47,000 okay we're
2:01:43 47,000 all
2:01:47 right you got to do the tapy
2:01:50 Taps so thank you to
2:01:54 everybody and uh yeah and so the other
2:01:57 thing I like to do which is it's a it's
2:01:59 a form of gratitude and
2:02:05 50,000 oops oh
2:02:08 no you can make money
2:02:11 with we did it
2:02:14 yay
2:02:16 um the other thing I like to do is any
2:02:20 of you gen xers remember romper room in
2:02:22 the magic
2:02:24 mirror I see Jim and I see
2:02:28 Tammy so who do I see tonight in my
2:02:31 little magic mirror let's hit 100,000
2:02:34 I'm out of here you can't hit 100,000 in
2:02:35 the next in while I say people's names
2:02:38 you're at 52,000 I know you're going to
2:02:40 try you weirdos uh laughing Thunder Kyle
2:02:43 uh I I will only show up on time if you
2:02:46 wait until I'm awake to start okay well
2:02:50 we'll make this whole thing about your
2:02:51 your uh schedule there laughing Thunder
2:02:53 I like it it's solid all us regulars
2:02:56 thank you Tobias good to see you Source
2:02:57 Camp good to see you
2:02:59 Danielle uh it's quiet Uptown good to
2:03:01 see you I haven't seen you in a bit good
2:03:03 I'm glad you're here Mr it fantastic
2:03:07 Jane do in here tonight Winston Welsh
2:03:10 Vicky Vicky good to see you Corey
2:03:13 Sandler nice nice dragon work uh TK
2:03:18 Robert
2:03:19 Rossy dang man so many Tam Z or Tams
2:03:23 Tams I'm guessing Tams yeah Tams Silver
2:03:26 Fox thank you very much Amelio's wife
2:03:29 was on time tonight like it's the Little
2:03:31 Miracles it's the Little Miracles Joy
2:03:34 pie great to see you Victory now Mr
2:03:38 it Monster igmu good name all right
2:03:42 digital Gods Lord digital Gods to you
2:03:46 it's quite up
2:03:48 toown all right everybody I'm out of
2:03:51 here have a fantastic Friday Eve or
2:03:53 Friday Saturday morning whatever the
2:03:55 hell time it is wherever you are um
2:03:58 anything going on tomorrow I don't think
2:03:59 so I'll probably get some some back
2:04:02 recordings of these Tik toks up on the
2:04:04 YouTube
2:04:05 channel if you're waiting for those CU
2:04:10 add and uh yeah that's it peace hope you
2:04:15 had fun tonight
2:04:16 byebye