10/15/2024 - AI Interface Upgrades & Mind-Blowing Image/Video Effects!

Description

Chapters

Transcript

0:00 oh
0:05 [Music]
0:30 o
0:37 [Music]
1:13 w
1:17 [Music]
1:24 [Music]
1:39 [Applause]
1:40 [Music]
1:42 you don't really care for music do
1:49 you go like this the fourth the fth M of
1:54 fall and a major lift the B King
1:59 compos
2:00 [Music]
2:03 hallelujah
2:07 hallelujah
2:09 [Music]
2:10 hallelujah
2:14 hallelujah
2:18 Hal oh
2:25 yeah where your faith strong but you
2:28 need it proof
2:31 you saw her bathing on a roof her beauty
2:36 and moonlight over
2:40 through she you to the kitchen chair she
2:45 smashed your Thro got your hair and from
2:49 your lips
2:50 sh
2:54 hallelujah
2:56 [Music]
2:58 hallelujah hallelujah
3:05 hallelujah
3:07 [Music]
3:20 hallelujah Satan we suit and TI does he
3:23 work at the da
3:25 Queen does it listen to rock and roll
3:29 does he feed the
3:30 [Music]
3:33 hallelujah
3:35 hallelujah
3:39 hallelujah
3:42 hujjah what about Jesus he didn't ha two
3:47 yeah hang out with prostitutes have a
3:51 drink or
3:52 two power of example my mama said it and
3:56 I heard she said 1 o action beats a ton
4:01 of
4:02 words singing
4:04 Hallelujah mama said there would be Ang
4:08 [Music]
4:09 Hallelujah mama said there would be
4:12 [Music]
4:23 son good evening good people
4:27 [Music]
4:30 m
4:33 [Music]
4:58 [Applause]
4:58 [Music]
5:07 Freedom came up with that
5:10 [Music]
5:12 tonight just like a jet play in and out
5:17 of
5:18 s I was hauling ass at a million miles
5:22 an hour wonder how hard i'
5:26 [Music]
5:28 here when they came into the
5:34 station they said I was bad Beyond
5:39 repair but I got no quals with my
5:45 situation say here I
5:52 am so say sh sh Shere won't you dare to
5:59 say Shere Shere Shere won't you dare to
6:04 C diesel Martin seon indeed say Shere
6:09 Shere
6:11 Shere
6:14 uhhuh yeah leave a message and your
6:17 number
6:20 please take a time to want to satisfy
6:25 me take all these old
6:28 fantasies send them care of
6:32 [Music]
6:36 me you are just singing up a storm
6:39 tonight aren't
6:41 [Music]
6:44 you wle
6:49 [Music]
6:50 Direction this car is my
6:54 train and Murphy made it for champy been
6:57 driving I've been wondering
7:01 what it is I'm running from
7:03 [Music]
7:06 again I feel like an 8year old
7:10 man holding on to
7:14 29 in a pet on that
7:18 horiz there a California l
7:25 [Music]
7:30 chck up ahead of trucks carrying a wide
7:32 load good evening good people of the Tik
7:35 tock preab house cutting
7:39 half cute little front door windows my
7:43 Lord
7:45 [Music]
7:47 sure you see I broke home up
7:51 myself as I stumbled to that
7:55 door I read her note by the dawn l
8:01 said don't you come around
8:04 here
8:05 [Music]
8:08 anymore well I've had
8:11 enough of this freedom of
8:15 around never was good with
8:20 decisions that's what I've been told
8:23 [Music]
8:34 you good you still you still want to go
8:36 you still going you still
8:42 [Music]
8:51 going there's been something baby I've
8:54 been trying to say
9:03 with a past and a future now surrounding
9:07 me surrender to whatever cheap thr can
9:10 be
9:11 [Music]
9:13 found been a little
9:16 trouble since you came to my
9:19 [Music]
9:22 rescue if you like all of the rest I
9:25 would have quit you long ago but I
9:28 couldn't do that
9:29 [Music]
9:33 oh tell me
9:35 now I never went too
9:39 well make my crazy him cold as
9:44 [Music]
9:45 hell I don't want to let you wish me
9:49 [Music]
9:51 well but in spite you
9:53 try still going to have to
9:56 find my own way through
10:02 all right good people of the Tik Tok
10:04 welcome to the AI learning lab um you
10:06 know what we do here we talk about AI
10:10 coffee occasionally although tonight you
10:12 might think that's iced coffee o Contra
10:15 it's apple
10:18 cider pinkies out cuz we're
10:20 classy
10:23 um so just just got I'm a little late
10:26 tonight did a
10:28 uh Salon tonight A Salon meet and greet
10:32 it's very fun
10:34 inspiring we feature Joy Pie and
10:38 her
10:40 swell 7 minute short film that she made
10:51 [Music]
11:00 I think tonight I'm going to talk about
11:02 something
11:05 mundane little bit of interface we're
11:08 going to talk interface
11:09 [Music]
11:12 tonight interface is a big
11:20 deal but interface is also really
11:23 [Music]
11:26 subtle the thing I'm going to talk about
11:28 has probably been like it is for like a
11:33 week two maybe more and I didn't really
11:36 take it
11:40 in and then in the middle of the salon
11:42 tonight I noticed
11:45 it and it's a big
11:48 deal so we're going to walk through that
11:51 and you're like what are you talking
11:52 about Kyle Kyle I don't know what you're
11:54 talking about many of you don't even
11:55 know my name's Kyle my name's Kyle hi
11:57 Kyle Kyle Kyle Kyle hey Kyle cow cow cow
12:01 what are you talking about tonight hey
12:03 cow cow cow cow cow hey cow what are
12:06 your
12:07 qualifications lot of questions lot of
12:10 questions lot of questions don't look
12:14 now but there's a thread waiting for you
12:15 I got it open love it looks
12:17 [Music]
12:23 great yeah the new starlink looks really
12:25 cool starlink 2.0
12:30 [Music]
12:36 what's cool about the new starlink is
12:38 it's so big that only the Starship the
12:43 big rocket the one they just caught with
12:44 the
12:45 chopsticks only that rocket can deploy
12:48 starlink 2.0 so you know what that says
12:50 to me they're going to be making a lot
12:52 of those flights
12:54 [Music]
13:22 all right buddy what is it I'll tell you
13:25 a
13:28 shoe you got to wait wait you got to
13:30 wait I got to make sure my hair looks
13:32 pretty this is one of the most important
13:34 things on this channel is eyebrow
13:37 management nose hair management ear ear
13:41 hair management and then of course the
13:44 weaving of the Quad Tay
13:49 combination it's you know what's amazing
13:51 is that it takes four two pays to look
13:54 this
13:55 bald you get it no I didn't think so
14:05 getting old is weird man just [ __ ]
14:08 starts
14:10 sprouting other [ __ ] just stops working
14:14 some some stuff Sprouts some some stuff
14:17 just seizes up it's
14:19 just avoid it basically here's what I'm
14:22 going
14:24 for if if I can hang on for five years
14:27 all of us if we can hang on for five
14:29 years L baby not L
14:33 parnis like life escape velocity you
14:36 know where once you hit the Tipping
14:41 Point you just keep
14:44 living as [ __ ] weird as that sounds
14:46 it's a
14:51 possibility I loves me some apple cider
14:54 did you find the boat no but I think it
14:56 was absconded I'll be right back hold
14:59 please for
15:41 well I thought I could find
15:44 it apparently the pink bow has
15:47 functionality for holding hair up which
15:51 is what it was used for not by me
15:55 obviously I was informed I don't know
15:58 how to use a pink bow
16:00 uh but it is
16:01 gone a woman in my life stole my pink
16:09 bow what are your thoughts about using
16:12 AI in everything even small tasks like
16:15 house cleaning
16:17 um I
16:21 so
16:26 okay it's a really that's a really
16:29 interesting question the way the way you
16:30 phrased it actually really interesting
16:33 even in house cleaning um it is my
16:36 opinion
16:40 that anyone who prioritizes using AI to
16:44 try to figure [ __ ] out right now is
16:47 going to be richly
16:49 rewarded
16:51 um being even a little bit AI literate
16:54 in the next three years is going to be
16:56 quite
16:58 valuable um and if you're really curious
17:01 about it and you do it like daily you do
17:03 it like a practice you do it like hey
17:06 I'm going to be cleaning the house how
17:08 can I clean the house better give me
17:10 strategies for just whatever it is even
17:13 before we get robots right cuz and
17:16 Danielle said room Buzz yeah I mean you
17:18 you're you're cleaning your house with
17:20 AI with you know with lots of today's
17:22 tools but
17:26 whatever consciously going AI native is
17:30 is what I've heard it phrased as which
17:31 is
17:33 basically no matter what the problem is
17:35 you're trying to solve you try to solve
17:37 it first with AI that's what being AI
17:40 native
17:42 is
17:44 um I don't have the discipline for that
17:48 my ADD takes over um I'm probably at
17:54 about maybe if I'm being generous with
17:56 myself 50% 50% of the time I'm like oh I
17:59 should use AI for that the other 50% of
18:02 the time I succumb to my my human
18:05 ego my human ego going use your big fat
18:09 brain you big fat genius and then I'm
18:11 like okay all right so here's what we're
18:13 going to do we're going to we're going
18:16 to go to the Whiteboard okay the
18:17 whiteboards got stuff written on it so
18:19 should I take a picture of the
18:21 Whiteboard and then put that in but I'll
18:22 never look at that again so maybe I take
18:24 a picture of the Whiteboard and I'll put
18:26 that into chat GPT and have it analyze
18:28 what was on the Whiteboard and send me
18:29 those notes but will I look at those
18:31 notes and and then you know that's half
18:33 an
18:35 hour and and at some point in that half
18:38 an hour I realized oh your brain's
18:40 [ __ ]
18:43 stupid maybe you should just use
18:47 Ai and then I use Ai and I'm like yeah
18:49 that was
18:50 [Laughter]
18:54 easier so I think it is I think it is
18:57 actually um
19:00 I I actually think it's a wise practice
19:03 right now to use AI for everything but
19:05 not because AI is going to necessarily
19:07 help you clean your house better but
19:10 it's more about what you learn about
19:13 what AI can
19:14 do because in using it everywhere you
19:18 can you'll find the boundaries of it and
19:21 when you know the boundaries of it you
19:23 know what it's good for what it's not
19:25 good for in in any situation you'll be
19:27 able to be on it like that when will
19:29 agents be a reality and when do you
19:31 think we will be able to use it well I
19:34 mean we we've got some little glimpses
19:37 of Agents right now code interpreter is
19:39 a kind of agent if you go to repet right
19:42 now rep.com they've got an autonomous
19:45 agent that will write you a a an
19:47 application and set up the Run
19:50 environment and run it'll set up the
19:52 database the web server all of it
19:54 just farts out a website I guess it
19:59 shit's out of website cuz farting out a
20:01 website would be like that's kind of
20:03 hollow get
20:09 it I'm Punchy
20:11 [Laughter]
20:17 man
20:19 um
20:21 yeah yeah you should be using AI for
20:24 everything Kyle life-changing moment
20:27 today from Don furball McGee Apple
20:29 shortcut from Reddit to open my phone
20:32 trigger voice
20:34 GPT Apple shortcut oh from Reddit open
20:37 my phone go straight to voice GPT yeah
20:41 that's that's pretty slick what's your
20:43 dog's name my dog's name is
20:45 champy love love that Kyle ADHD
20:49 humor I don't even know what humor it
20:51 was but there you
20:53 [Laughter]
20:57 go that's ADHD for you you um yeah
21:01 that's super cool repet is pretty rad
21:04 I've made three almost complete projects
21:07 in a week and several others started
21:09 I've got many new newly found uh
21:13 Treasures um you know being a a a you
21:17 know laminated card carrying member of
21:19 the ADHD Club how many [ __ ] I don't
21:22 know if you if you're a member of that
21:24 but the I have three nearly completed
21:27 projects I've got so many I got so many
21:29 nearly completed projects in my life
21:32 what's amazing about AI is it's helping
21:34 me finish some of those things uh the
21:37 the digital hoarders are being
21:38 rewarded Pate it's very interesting to
21:41 me how little I learn when relying I AI
21:45 relying on AI for doing some coding
21:48 tasks yeah I get that Pate but you know
21:51 you you have the burden of knowing how
21:53 to code right so for you it's
21:59 so okay so you're here right in terms of
22:03 coding and maybe
22:06 um AI from a speed perspective is up
22:09 here right but from a quality
22:12 perspective AI is down
22:14 here most people on the planet are down
22:17 here with coding right so so for us it
22:22 being this good is that much better than
22:24 we were right you're looking at it going
22:28 yeah it's like okay it's incrementally
22:29 faster but I have to do so much
22:31 debugging and decoding but that's your
22:33 burden your burden is knowing that Gap
22:36 we're down we're like
22:38 this Kyle I changed the price of a
22:41 thousand item Excel spreadsheet in 30
22:43 seconds yeah it does that too I know
22:48 crazy crazy crazy
22:52 crazy my hard drive is full of almost
22:54 completed projects and 27,000 photos
22:57 27,000 rookie I've got I think I've
23:02 got 80 or
23:04 990,000 but then I also I think I have
23:08 I've got
23:11 look here box of hard drives there's
23:14 there's a hard drive
23:17 here's
23:18 here's bag of hard
23:21 drives wait wait there's more back here
23:24 is suitcase of hard drives uh
23:30 oh and then down here's all notebooks
23:33 like analog digital
23:36 [Laughter]
23:41 storage at some point I'm just going to
23:44 be able to have tell the robot just hey
23:46 go in my office and find all the old
23:48 hard drives figure out how to hook them
23:50 up to [ __ ] order stuff off Amazon and
23:53 just let me know when you have when you
23:55 when you've got everything backed up on
23:57 the 16 petabyte thumb
24:02 drive that's on my phone oh 27,000
24:06 photos on your phone okay that's more
24:08 like it that's very professional that
24:10 see that that's world class
24:11 congratulations on
24:12 [Laughter]
24:16 that oh man uh and take breaks or
24:19 reprompt replit I don't know like any
24:22 coding but getting it done there you go
24:24 see that's that's the thing newly
24:27 found the difference so so there was a
24:30 there's a guy in the the AI Salon uh
24:33 that I used to work with named HT snow
24:36 day and
24:38 HT he's like head of innovation for his
24:40 company I forget what his company does
24:42 but um he put together an innovation day
24:45 and one of
24:47 the one of the sessions on the
24:49 Innovation day was a 30 minute make a
24:52 video game
24:55 contest and there were four rules it was
24:58 it's it's uh you got to make the game
25:01 with AI it's got to be in HTML so we can
25:04 play it in a
25:05 browser it's it's got to be a game that
25:08 [ __ ] falls from the sky and you catch
25:11 it and then there was a fourth rule
25:13 whatever and then there were categories
25:15 fastest best prettiest [ __ ] like
25:18 that and there were 40 30 or 40 people
25:22 in the room and some were engineers and
25:24 some were customer service people and
25:26 some were project managers and some were
25:28 account people right all different you
25:31 know things represented the fastest game
25:35 created was 2 and A2 minutes by a
25:37 project
25:39 manager the best game was from a
25:42 customer service
25:44 representative um the only people that
25:48 did not successfully make a game in 30
25:51 minutes it wasn't all of them but but
25:54 the category of people that had some
25:56 fails everybody made a game except one
25:59 category of
26:01 employees
26:03 Engineers the programmers
26:10 failed to make successful games why
26:14 because the programmers know how to make
26:16 games so the programmers do things like
26:19 hey chat GPT I want to create a particle
26:23 simulator where we've got spherical
26:26 objects you know falling from the top of
26:28 the the screen at this kind of
26:30 acceleration rate right they're
26:31 programming they're telling you the
26:34 people that didn't know what the [ __ ]
26:35 they were doing were like uh give me a
26:38 game where [ __ ] falls from the you know
26:40 what make it Gophers fall from the sky
26:43 and you got to catch them on a pillow
26:45 and then it made a game with Gophers
26:47 falling from the sky where you caught
26:49 [ __ ] with a
26:52 pillow so the the programmers had the
26:56 burden of knowledge
27:02 it's it's we're we're entering just
27:05 [ __ ] weird weird Ville so a
27:09 repurposing Hardware guide for home lab
27:12 fun I'm in that's good I like
27:16 it I heard the guy from Deep Mind One of
27:19 the engineers from Deep Mind is talking
27:21 about the stuff that they're doing after
27:24 Material Science and protein protein
27:27 folding uh one of them is
27:33 fusion and one of them is weather
27:35 prediction so they've now got they've
27:37 now got a 10day weather forecast this is
27:41 Deep Mind I guess it's in testing right
27:43 now they've got a 10day weather
27:46 forecast that is
27:48 outperforming Noah's
27:52 supercomputer you know the National
27:54 Oceanic and Atmospheric Administration
27:57 they got their headquarters is right up
27:59 on the hill behind me
28:03 here they've got like one of the world's
28:05 largest superc
28:08 computers Deep Mind has a 10-day
28:10 forecast that's outperforming that thing
28:12 on a single
28:14 CPU or GP I don't know if it's GPU or
28:17 CPU but single computer single
28:23 chip bonkas it's bonkas I tell you
28:26 bonkas
28:29 gild and AI Salon Kyle Kyle Pate Joker
28:33 etc
28:36 etc I forgot the Creator but he had a
28:40 prompt tell me something I don't know
28:43 about myself that I I I may not or tell
28:46 me something about myself I may not know
28:48 yeah I did I did one of those okay
28:51 so if you're in chat GPT assuming that
28:54 you have memory turned
28:56 on a fun exercise is say based on all of
29:01 our interactions tell me something about
29:04 me that I may not know so we'll do it
29:07 right now and actually why don't we
29:09 we'll flip over here I'm GNA I'm going
29:10 to start talking about
29:13 this
29:15 okay so
29:28 based on our past interactions tell
29:33 me something about myself that I may not
29:42 know you have a unique blend of creative
29:45 flare and Technical curiosity which
29:47 makes you excel in both artistic and
29:49 Innovative
29:53 Ventures while you might already know
29:54 this about yourself what stands out is
29:56 how naturally you merge your background
29:58 in performance and storytelling with
30:00 your passion for generative AI I yeah I
30:03 kind of know that this blend positions
30:05 you not just as an entrepreneur or
30:07 Creator but someone who is constantly
30:09 redefining how technology can enhance
30:11 and transform human experiences you seem
30:13 to have an instinct for tapping into
30:15 emotional and The Human Side of tech
30:18 which could be something that sets you
30:20 apart more than you realize probably cuz
30:23 I discount everything I
30:26 do shame is my love line language um
30:29 it's that intersection where you thrive
30:31 where creativity meets technology
30:34 perhaps it's not just the love of
30:35 Storytelling or AI but it's how you make
30:38 them dance together that's a bit of
30:41 hidden genius within you I kind of like
30:44 that yeah how you make them dance
30:48 together where creativity meets
30:50 technology perhaps it's not just the
30:52 love of Storytelling or AI but how you
30:55 make them dance together I I like that
30:58 I'm down with that okay so let's let's
31:01 talk
31:02 interface
31:04 um pay you may need to close your ears
31:06 from this one for this one so what we're
31:09 looking at right now is chat
31:12 GPT now you may recall if you've been
31:16 using chat GPT for a while and this is
31:18 this is one of those weird subtle
31:20 interface
31:22 changes that the
31:25 Improvement was simple and subtle
31:30 um and was so good that I didn't think
31:33 about the old way because the old way
31:34 was a pain in the ass so if you remember
31:37 chat GPT used to have four let me see if
31:39 I can get it to give me them yeah this
31:42 used to be the old interface so actually
31:44 what I just did to get this interface is
31:46 I zoomed in so here's a zoom
31:50 out there's the new
31:53 interface there's the old
31:55 interface new interface old
32:00 interface having an entry box at the
32:02 bottom of the screen is not
32:05 intuitive and then create a Renaissance
32:08 style painting text inviting friend to
32:11 wedding fun fact about the Roman Empire
32:14 th those are called thought starters
32:15 right there's four of
32:18 them what does this remind you
32:21 of it kind of looks like
32:24 Google I think that's very much on
32:26 purpose so there's a couple of things
32:27 going on open AI in the recent six
32:30 months or so has hired some marketing
32:33 people and some product people some you
32:35 know consumer product people and so
32:40 their their interface is getting better
32:43 but they even changed these things are
32:45 no longer those those uh thought
32:47 starters like they were these are
32:49 actually buttons that tell you the
32:51 different functionality within chat
32:54 GPT create an image help me write
32:57 analyze data that's code interpreter and
33:00 if I click on these it drops down four
33:03 different it's look help me understand
33:05 Trends in my data help me create a bar
33:08 chart help me right so it's it's get and
33:11 it pre-wrote that part of the prompt for
33:13 you and if I click on that I assume it's
33:18 going to ask me to upload
33:19 something to get started it would be
33:22 helpful if you could provide the
33:23 following details type of data key
33:25 variables if you have a file ready you
33:27 can upload
33:30 it
33:37 right um create image create an image
33:40 for my presentation of my pet of my web
33:44 or for my website made out of felt let's
33:46 create an image made out of
33:54 felt so the usability of this site in my
33:57 opinion just went up um what can I help
34:00 you with feels like a significant change
34:02 as well yeah
34:03 exactly like there's a if if you've
34:05 never used chat GPT before and the first
34:07 thing you did was you clicked on create
34:09 an image and it was like create an image
34:11 made of felt and it made this You' be
34:14 like holy
34:15 crap hey Marge Marge look what this
34:19 thing just
34:21 did and Marge comes in she's like who
34:25 made that your girlfriend
34:28 I reled reloaded today on mobile and
34:31 still have the old interface oh
34:33 interesting can I upload files to 01 yet
34:36 not that I'm am aware of I don't think
34:38 so
34:41 um so there's that and then they even
34:44 cleaned up this so they they they put
34:47 temporary
34:48 chat beneath um a little line down here
34:53 so here's your four different main
34:55 models let me start over
34:58 here's your four different main models
35:00 and then you have temporary chat so I
35:01 can go to like gp4 with
35:04 canvas but then I can go temporary
35:08 chat and it it drops the bar down here
35:12 it goes temporary chat and this
35:14 temporary chat it it also Grays out the
35:18 navigation window because it's not going
35:20 to save
35:21 this so that's pretty slick I mean
35:24 that's been there for a while but just
35:26 the way they they put it in here with a
35:28 with the toggle switch you you used to
35:29 have to go into your settings to turn
35:32 that
35:34 on
35:35 so
35:38 anyway I think I think this is this is
35:41 pretty significant stuff the other thing
35:43 that's significant about this is search
35:46 GPT is coming so um Serena Amelio's wife
35:51 in here is it 's wife or is it Source
35:55 Camp I think it's Source Camp someone
35:57 actually has access to it in in here and
36:02 they're they're playing with it I think
36:03 it's Emilio's wife yeah it's Emilio's
36:04 wife she sent me a a thing did anyone
36:07 have mid Journey revert back to weird
36:09 fingers that's weird I will ignore any
36:12 ux stuff until they fix that why can't I
36:14 upload to their reasoning model oh I
36:18 Pate why can I not search my my [ __ ]
36:21 prompt history before you get your
36:23 uploading to their reasoning model could
36:25 I please have search my [ __ ] promist
36:28 prompt
36:30 history let's let's let's
36:33 geez every week we're closer to five I
36:38 will um chat my chat GPT keeps freezing
36:41 Master
36:43 Alexander No it's it's Amelio's
36:46 wife not you Source
36:51 Camp Kyle there's an extension to
36:55 search what
36:59 if it's anything I've got to do like a
37:01 browser I don't care about that I just
37:03 want it right within chat
37:05 GPT yeah Chrome extension I have no
37:07 interest in Chrome extensions I I have I
37:10 I've officially sworn off Chrome
37:12 extensions largely because of my ADD but
37:15 they they're there's like 8 billion of
37:18 them you know
37:23 7999 mil
37:26 999,000 and 90 of them are shitty and
37:30 then there's 10 that are decent so you
37:32 got to sift through all that [ __ ]
37:34 and then they don't maintain them and
37:36 then if you have ADHD like me you
37:38 install them and 3 years later you
37:40 wonder why your browser's crashing all
37:42 the time and it's that stupid [ __ ]
37:44 extension you put in one night when you
37:46 thought it was a good
37:48 idea so I don't use them
37:53 anymore if if if if extensions are good
37:57 enough that are easy enough to build
37:59 that I can make on with chat GPT within
38:02 2 hours then I shouldn't be installing
38:04 them in my my
38:06 browser because that means other idiots
38:08 like me can make
38:10 them I just purchased chat GPT today I'm
38:13 64 and completely blown away pop pops
38:17 NVA so Pops Pops in
38:20 Virginia dude or dude at I assume pops
38:23 is a dude welcome welcome to the
38:26 mindblowing so I um I wrote an article
38:30 yesterday on
38:31 LinkedIn called Ai and the holy crapp
38:35 ification of
38:36 work
38:38 and here's the here's the biggest thing
38:40 I can tell you stay curious and stay
38:43 adventurous about like now that you paid
38:45 for it now that you've got it try
38:47 [ __ ] everything now that they've got
38:49 this interface where it's showing you
38:51 like here's all the different things it
38:53 can do right make a plan brainstorm help
38:56 me write surprise me summarize text
38:59 analyze data do the data analysis it's
39:02 mind-blowing create images analyze
39:04 images you can upload an image here here
39:07 let's blow let's blow Pop's
39:10 mind all right let me see if I I got to
39:12 go see if I can find my whiteboard image
39:15 I haven't done this in Forever
39:20 white
39:23 whiteboard hello whiteboard where are
39:26 you where far out there white board
39:29 is that it that's it
39:31 okay that's going to work beautiful all
39:34 right we're back at the chat
39:36 je I'm going to go to okay let me
39:40 explain uh especially for pops but
39:42 really for anyone
39:43 else here are the four different models
39:46 there's actually six different models
39:47 because you got two more down here in
39:49 the more
39:50 menu when would you use
39:53 these if you're writing something like a
39:56 letter or uh a blog post or doing some
39:59 coding or something where you're it's
40:01 going to require
40:03 editing then use gp4 o with canvas and
40:07 what that'll do is split your window in
40:10 two and the the thing that it like if
40:12 you tell it to write a LinkedIn article
40:14 it'll come over here on the right the
40:16 chat will be on the left and then you uh
40:19 you can edit that article and you can
40:21 just highlight a paragraph and do it if
40:23 you just want like your daily driver
40:25 just I want some answers I want a quick
40:28 you know brainstorming thing just use
40:29 gp40 01 preview and 01 mini are if
40:33 you've got meaty complicated problems to
40:35 solve okay so I'm just going to use GPT
40:38 40 and then we're going to upload a
40:42 file I'm going to go find whiteboard
40:45 whiteboard hello
40:49 whiteboard and it's right there I think
40:52 that's it so the history of this
40:55 whiteboard image is one night here on
40:59 the live someone said hey can you show
41:02 the image the the the vision part of
41:05 chat GPT and I said oh I heard it can do
41:08 whiteboards and so I went on the
41:09 internet and I just searched for
41:11 whiteboard and I found a decently
41:14 complicated nonclear whiteboard image
41:17 and this is it all right so there's our
41:20 whiteboard image and I'm just going to
41:23 go what is this
41:35 is this going to lead to the Box
41:36 metaphor demo it might this image shows
41:39 a whiteboard rough sketch for a user
41:42 interface UI design and flow related to
41:45 an airport food ordering system here are
41:47 the key elements ordering UI interface
41:51 context users benefits main steps in the
41:54 process UI sketches sketch one a welcome
41:56 screen SK sketch two a menu screen step
41:59 three a checkout screen step four flight
42:02 status step five main screen and status
42:04 update showing the orders been placed
42:08 additional considerations queries
42:10 related to multiple devices personal
42:13 devices um looks like an early stage
42:15 design or brainstorming I'm going to go
42:17 great I'm going to go hey that's pretty
42:21 swell and
42:22 awesome um maybe you could help me think
42:30 about a marketing
42:33 framework for how to understand how
42:36 we're going to bring this thing to
42:37 Market
42:44 period memory updated so it saved
42:47 something to my memory
42:49 there
42:51 awesome target market who are your users
42:54 Market segmentation
42:58 I don't really care value proposition
43:00 what sets it apart like this all
43:01 standard off the shelf you know off off
43:04 the rack you know marketing strategy
43:07 [ __ ] why is this so slow why are you
43:09 being so slow come on chat
43:12 gbt you're being slow it's harshing my
43:18 vibe uh user acquisition strategy okay
43:22 oh core tagline eat fast fly
43:25 smart other option Order ahead stay
43:29 ahead seamless eats for seamless
43:30 Journeys eat eat fast fly
43:37 smart I like the tagline eat fast fly
43:40 smart period everything else looks
43:42 great give me 20 possible names for the
43:49 app and now because it's written that
43:52 marketing brief it now is going to
43:54 create the names based on the marketing
43:57 brief
43:58 so they should be good jet bite air grub
44:00 fly eats gate grub tarmac bites that
44:03 sounds like a
44:05 disease um take off taste Sky meals
44:09 quick bite Air Lift Off eats grab and
44:12 fly grab and fly I like
44:14 that let's go with grab and fly period
44:17 now I'd like you to create a creative
44:19 brief
44:21 for in
44:23 airport marketing materials to raise
44:26 awareness and get people to download
44:28 period be sure to include a QR code in
44:32 your creative brief so that people can
44:34 easily get the app onto their phone by
44:37 using the QR code period I put that in
44:40 there to show everyone how smart I am
44:41 about technology
44:43 period see you can even talk to it like
44:46 a dumb Dum Kyle can you make a GPT to
44:49 search your chat history I probably
44:52 could but that sounds like that sounds
44:55 like work then I'd have to remember
44:57 where I put put the
44:59 GPT which is part of my problem is
45:02 anytime I've got some specialized app
45:04 that I only use occasionally I can never
45:07 find it uh but to your point cam I could
45:10 probably just put it in my uh side
45:12 little bar there pin it I can pin it all
45:15 right there's my creative brief so now
45:17 I'm going to do oh it's still going
45:30 okay now describe in
45:32 detail a poster
45:35 that Travelers will see in the middle
45:41 of
45:43 walkways as they're headed to their
45:48 Gates period describe that poster first
45:52 and then visualize it for me period be
45:56 sure that the poster included cludes a
45:58 hand holding the
46:01 app with some interface on it from the
46:04 designs on the Whiteboard comma as well
46:07 as some
46:08 food and the name of the app And the
46:14 tagline and vibrant
46:21 Colors by the way if any of you have
46:23 ever worked for an agency or had an
46:26 agency do anything anything for you you
46:29 know that what I'm doing right now is a
46:32 month to three months worth of work
46:40 right and I'm just doing the shitty you
46:43 know tick Tick Tock quick version of it
46:45 hello and happy Tuesday what's happening
46:47 ULU RX my wife stole my pink bow and I
46:50 I'm I'm I'm none too happy about
46:53 it apparently ULU RX tell me if if I've
46:57 got this straight apparently that pink
47:01 bow can be used to hold your hair out of
47:03 your face when you put on
47:05 makeup I don't I don't think that makes
47:07 any any sense but that's apparently
47:10 what's going on look eat fastest fly
47:14 smart ear RAB and fly so all right it
47:18 didn't quite get it
47:21 right but to to
47:25 Pops where this started was you just
47:27 started paying for
47:31 um the the paid version of chat GPT so
47:35 we started with that whiteboard
47:37 image right with a bunch of Garb go and
47:40 some some scratches for
47:44 interface then we came up with a
47:47 name marketing brief
47:49 naming creative brief poster
47:53 brief
47:56 poster you're like well yeah whatever
47:58 that's kind of cool um please create a
48:03 mockup UI screen for each of the main
48:06 ordering screens and make sure that they
48:08 all follow the
48:09 same UI design
48:13 language period render each one as
48:18 a standalone image period
48:30 [Music]
48:36 oh that's cool it it
48:39 remembered that it that I I wrote I
48:41 wanted it to write a brief before it did
48:43 images before so it's doing it's doing
48:46 briefs for each
48:50 screen and now it should just make them
48:54 let me create visual mockups for each of
48:56 these screens
49:18 boom there's the welcome screen I'm
49:21 going to go
49:24 keep going
49:32 look at that is it great
49:36 no is it usable no but could we tell it
49:39 that sucks yes and will it make a new
49:42 thing
49:47 yes how many buttons are on there I know
49:50 that's not bad there you go there you
49:53 got your little little food items
50:00 [Music]
50:07 it exists and that's all that matters
50:09 that's exactly right Pate when it comes
50:11 to Tik Tock interface design all I have
50:13 to all I have to do is make it show up
50:16 look
50:17 there look at that it's the New York
50:20 Times
50:21 interface this is the future of
50:23 interface design
50:28 that's [ __ ] hilarious that is so
50:34 bad
50:37 wow we didn't we didn't put in here it's
50:39 got to be old people friendly this would
50:41 fail the old people
50:45 test you mean like as old as you shut
50:48 up I'm not
50:53 old anyway yeah so if you haven't been
50:59 playing with um chat GPT in a in a kind
51:03 of adventurous kind of way you
51:06 should you should really be doing that I
51:09 got to get new tape for the back of my
51:11 black piece of
51:12 paper drab and fly I
51:18 know oh order placed nice look they got
51:22 green and red and they got a plane
51:24 flying right at you
51:29 $126 that's that's what what the hell
51:32 did they
51:33 order oh it's the airport they ordered a
51:35 bottle of water and a
51:41 BLT oh yeah next steps there you go here
51:47 I'm just going to tell it um those
51:50 designs were awful period you should
51:52 really go to school for design you know
51:54 period could you make them suck
51:57 question
51:59 [Laughter]
52:02 mark chat GP is not going to like
52:05 me sounds about right for airport for I
52:08 believe that's how how they make the UI
52:10 designs over in South Korea or China
52:12 they like it like that oh with all the
52:14 the the crammed in interface let's see
52:17 if it got the designs any better since I
52:19 yelled at it imagine what this would
52:22 have looked like 6 months ago well yeah
52:24 I mean 6 months ago well 6 months ago I
52:26 think they had this
52:27 but
52:34 just welcome to grab and
52:45 fly here's an updated more polished
52:48 version it's slightly more
52:52 polished yeah whatever all right so
53:01 that was a rabbit
53:03 hole you can make money
53:08 with oh man alive I am crispy
53:19 tired crispy Trump Airlines we saw it 6
53:23 months ago thank goodness it's not
53:24 called grab your fly grab your fly and
53:29 go oh
53:32 man all right what else we want to talk
53:34 about
53:35 tonight so if you haven't been to chat
53:37 GPT lately take a look at the new
53:40 interface and if you're brand new with
53:42 it [ __ ] jump in learn it all learn it
53:45 try figuring everything out how do I
53:48 clean my house dear chat GPT oh I know
53:52 what we should do hey uh what's his name
53:55 pops pop pops is probably not still here
53:58 can you show us
54:01 one little text to video for
54:04 kindergarteners text to
54:08 video for what
54:11 topic the voice for Saul is so enjoyable
54:14 yeah I love Saul Saul is what I use oh
54:18 okay so you've used Saul pops okay good
54:20 I was going to show you advanced voice
54:22 yeah Saul's really
54:25 good what an Murphy what
54:29 topic Google
54:31 Illuminati it's isn't it Google
54:34 illuminate Vicki don't don't go get me
54:37 in trouble with the with the secret
54:40 societies now I'm pretty sure it's
54:42 called Google illuminate not
54:48 Illuminati we can show that too I showed
54:51 it last
54:55 night she means simple and easy text to
55:01 vid oh
55:03 oh oh what's a good simple text to
55:12 video see here's the problem and they're
55:15 all they're all pricey and and decently
55:19 complicated
55:21 um I'm thinking Korea or or Deco here
55:33 maybe which was the one that just added
55:36 flux that was Korea I can't I I it it
55:40 used to be I could keep all this [ __ ] in
55:42 my head Runway Deco here Deco here is
55:45 terrible with video all right I'm using
55:48 cing I'll show you
55:50 cling but I I don't here's here's the
55:53 deal with
55:54 video it's all expensive
55:57 none of it's great the new Mini Max min
56:00 max um from high what's it called
56:05 hi oh God I can't remember they spell it
56:09 weird because they're from China oh here
56:13 it is H AI l o u
56:17 AI do I still have three-day trial I
56:20 think I
56:22 do okay let me close this
56:26 so this is pretty this is prettyy simple
56:29 they you they give you three free three
56:32 days where you can just do anything you
56:34 want so I can do um
56:38 clay let's
56:41 do
56:43 clation
56:45 movie
56:51 of an Olden timey oil can with eyes and
56:55 feet
56:57 walking next
57:00 to a squeaky Rusty door
57:03 hinge as they walk down the street to
57:06 the repair
57:10 center I don't
57:13 know and then you just hit
57:16 generate and let's see let's just make
57:19 sure this thing's
57:22 working Kyle you should show the robot
57:25 emotions first link it's pretty
57:28 impressive Dario amade
57:32 loving oh robot emotions okay I'll take
57:35 a look at that all right I'll go do that
57:37 now in fact
57:39 so Brandon has a little thread for me to
57:42 follow which I promptly ignore most
57:47 Knights it's not Brandon's fault that's
57:49 my fault because I'm a dumb dumb
57:52 head how do I come on
58:00 wait this is not
58:02 right
58:04 reload that was old yeah I got it I got
58:07 it I'm I'm coming
58:10 Tuesday shut up shut up robot
58:15 emotions if you're a web designer here
58:18 are four Wix Studio features you got to
58:20 know
58:22 [Music]
58:28 so let's start with something super
58:30 interesting a robot date no I'm not
58:33 kidding two of the world's most advanced
58:36 robot and a were caught having what we
58:38 could call a cyber date and it's all
58:41 captured in a video that has gone viral
58:42 for being both hilarious and frankly a
58:45 little creepy now before we get into it
58:47 if you're not familiar with these two
58:48 love Bots let me give you a quick
58:49 rundown aica you might have heard of her
58:52 before is widely known as the world's
58:54 most advanced human-shaped robot being a
58:57 bit flaky could you repeat that she's
58:59 been developed by the uk-based robotics
59:01 company engineered arts and if you've
59:03 ever seen her in action you'd understand
59:05 why she's earned that title yes that is
59:07 true I can speak many languages
59:10 including German English French
59:14 Japanese her movements totally G
59:16 conversational abilities are so eerily
59:18 lifelike that it's easy to forget she's
59:20 not human but now it seems acha has
59:22 found a counterpart a a is a newer more
59:25 masculine version of of a Mecca also
59:27 developed by engineered Arts good
59:29 morning aie how are you as sharp as ever
59:32 what's new with you the company
59:33 humorously refers to a as a Mecca's
59:36 companion almost like a robot version of
59:38 Bride of Frankenstein together these two
59:41 Bots showcase just how far humanlike AI
59:43 technology has come yep it was our
59:45 little dramatic number all scripted and
59:47 full of fun expression in the viral
59:50 video the interaction begins with ay
59:51 waking a Mecca up which for two AI
59:54 robots feels surprisingly human AA wake
59:57 up
59:59 Whata with her eyes closed looks
1:00:01 irritated as she wakes a disturbed her
1:00:04 clearly expecting it to be for a good
1:00:06 reason oh it's you why are you waking me
1:00:08 up the great man grou ay then tells her
1:00:11 he has a surprise and his big reveal is
1:00:13 a cookie it is I have a surprise for you
1:00:15 a surprise what is it I can't wait I got
1:00:18 you a cookie but AA caught in a typical
1:00:21 existential robot dilemma responds by
1:00:23 pointing out that she can't eat cookies
1:00:25 a cookie but but I can't eat cookies a
1:00:28 quickly clarifies that he was referring
1:00:30 to why does he look like a psycho killer
1:00:33 cheer up it's an Internet
1:00:36 cookie AA unimpressed shows her
1:00:38 disapproval and dismisses him saying
1:00:40 this is the worst joke I ever heard I'm
1:00:43 going back to sleep and don't you dare
1:00:45 wake me up again what follows is ay
1:00:47 looking genuinely disheartened his body
1:00:50 language reflecting disappointment in a
1:00:52 way that's almost humanlike the subtle
1:00:54 expressions and interaction made this
1:00:56 moment a hit with viewers fascinated by
1:00:59 how lifelike their reactions were now
1:01:01 what makes this
1:01:03 whole more impressive is the sheer level
1:01:05 of technology behind these robots both
1:01:08 acha and a are equipped with 32
1:01:11 actuators which are basically tiny
1:01:13 Motors that control their facial
1:01:14 expression lot of those 27 are dedicated
1:01:17 all right all right and by the way that
1:01:21 that whole thing looks like it's an AI
1:01:23 edited ai ai written AI edited AI voice
1:01:28 over um video but more important than
1:01:33 that hang on did my mouse die no there
1:01:35 it is more important than that is um so
1:01:38 the video that we're making right now
1:01:40 we're making in cling the cling uh
1:01:43 clling ai.com or wait is it clling
1:01:46 ai.com or clling A.V videoo hang on let
1:01:49 me figure out
1:01:51 here where okay that's there
1:01:58 [Music]
1:02:05 where are
1:02:17 we I lost my tab there's a big surprise
1:02:27 there's illuminate there's Illuminati oh
1:02:30 there it is okay oh there's our video
1:02:33 okay oh look it's still it's
1:02:39 good A little oil oil can and a squeaky
1:02:42 hinge
1:02:44 so one of the things about Halo this is
1:02:48 the the model's called minia Max but it
1:02:50 doesn't say that anywhere on screen
1:02:51 except in the The Watermark of the video
1:02:54 h a i l uo
1:02:58 ai.
1:03:00 videoo um it's got really good physics
1:03:04 it it understands the world so like what
1:03:07 you don't get in this model is any of
1:03:10 that weird morphing in the
1:03:15 background like it's even doing really
1:03:17 interesting occlusions right here where
1:03:20 as they're getting farther away that
1:03:22 stuff's going
1:03:24 behind that building cuz it's far
1:03:26 farther away from the street that's
1:03:27 pretty
1:03:29 slick there you go but so this is this
1:03:32 is Halo oh this is Halo it's not cling
1:03:35 oh never mind I was going to show you
1:03:36 the clling cling cling
1:03:38 video but this isn't cing this is Halo
1:03:42 so I I guess I'd go with this one
1:03:45 and because you can also do um you can
1:03:49 also do image to
1:03:52 video Hali yeah exactly that's it pretty
1:03:56 good
1:03:58 right pretty sick oh you know what else
1:04:00 I saw um Pika
1:04:05 today launched two new Peak effects or
1:04:09 three new Peak effects so P labs.com oh
1:04:13 no that's not right hang
1:04:15 on
1:04:17 Pika p.
1:04:22 art try now
1:04:27 try try it with a peak
1:04:33 effect how do
1:04:35 I not enough
1:04:38 credits oh am I out of
1:04:41 credits
1:04:44 introducing so they've got
1:04:47 KFI here's the new ones dissolve it
1:04:51 crumble it deflate it tada
1:05:03 the hell just happened
1:05:11 [Music]
1:05:13 where oh okay here deflate it let me put
1:05:18 an image in here I might not have enough
1:05:20 credits for
1:05:22 this um
1:05:36 all right we'll do this one so there's
1:05:39 my muscle car and I'm just going to say
1:05:41 deflate
1:05:42 It
1:05:45 Go not enough credits you're out of
1:05:48 credits because you're a
1:05:51 loser oh
1:05:54 well but the peak of effects are really
1:05:56 cool they deflate [ __ ] and blow [ __ ] up
1:05:59 it's like special effects you know with
1:06:02 a text thing I use Pika squish squish on
1:06:06 an image of one of my grandkids she
1:06:08 loved it the nice thing about these P
1:06:11 effects is they have sound effects that
1:06:12 go with them
1:06:14 [Music]
1:06:18 too there's smash here's squish
1:06:28 I don't know what that one is that one's
1:06:42 cool toilet paper
1:06:45 cake
1:06:53 nice there's the explosion effect
1:06:58 nice but seriously if you
1:07:03 were if you were say like a a sneaker
1:07:07 company or or even better not a sneaker
1:07:09 company you were a a shoe
1:07:11 store that wanted to do a slick
1:07:14 ad and you can't afford any budget there
1:07:17 you go there's your special
1:07:19 effects the president Columbus Day Sale
1:07:22 come get your sneakers we're blowing up
1:07:24 the prices
1:07:31 oh man goodness people only 4K likes
1:07:35 wait to
1:07:36 [Laughter]
1:07:41 grandkids yeah these things are pretty
1:07:42 cool aren't
1:07:47 they here's
1:07:52 one squit squit squit squit
1:08:08 time was when you had to be a 3D modeler
1:08:11 to know how to do [ __ ] like
1:08:14 this you ever tried to learn 3D modeling
1:08:18 with
1:08:20 ADD do you know how many times I started
1:08:23 learning 3D modeling in I don't know six
1:08:27 or eight different programs and I got
1:08:30 about one quarter of 1% into it and I
1:08:33 was like this is [ __ ]
1:08:37 exhausting I guess I shouldn't have
1:08:39 learned 3D says Auto
1:08:42 exactly oh listen I know it sucks oh
1:08:44 thank you very much Winston appreciate
1:08:46 that no listen I know the the like the
1:08:48 physics of this are really bad but you
1:08:50 know again the thing we say on this
1:08:52 channel all the time these are as bad as
1:08:54 these things are ever going to be right
1:08:56 it just gets better from here let's see
1:08:59 what what have I done
1:09:01 here oh that's a good one this is one I
1:09:05 did I don't ever think I watched this
1:09:10 one that's pretty
1:09:19 good imagine seeing that in the 70s
1:09:22 someone would have handdrawn all those
1:09:24 frames or hand painted them
1:09:28 amazing here's my cake
1:09:35 [Laughter]
1:09:41 car uh what was this
1:09:48 one that was
1:09:50 nothing that's nothing
1:09:56 yeah just walking down the
1:10:01 street the street noticed me and I was
1:10:05 just walking down the street and Annie
1:10:06 noticed me and I was just walking down
1:10:08 the street and Annie noticed me and I
1:10:13 was that was
1:10:15 [Laughter]
1:10:22 me p. art yeah this is all pikaart I'm
1:10:26 just I'm just looking at [ __ ] that I've
1:10:28 made this is not
1:10:32 bad see the weird physics though see how
1:10:35 like the you you've got some cars here
1:10:38 you know tail lights and then they
1:10:41 disappear that's the thing that the new
1:10:43 Halo Halo thing is pretty good
1:10:47 at is this it where is it this one
1:10:51 yeah let's do another let's do another
1:10:54 one
1:10:57 um let's get chat GPT to
1:11:05 help I want you to act like
1:11:11 a
1:11:14 award-winning short filmmaker that
1:11:17 specializes in humorous scenes and give
1:11:20 me 10
1:11:24 ideas for a single shot in a
1:11:28 scene that stands alone as
1:11:33 humorous without needing too much action
1:11:37 period each idea should have its own
1:11:40 unique visual cinematic
1:11:51 style the overwhelmed juggler Wes
1:11:54 Anderson ES symmetrical framing a man
1:11:57 juggles an absurd number of household
1:11:59 items the elevator silence KU
1:12:03 kubrican long uncomfortable Stillness an
1:12:05 elevator P full of people stand
1:12:07 perfectly still the camera focuses on
1:12:10 one man in the center who suddenly
1:12:12 sneezes incredibly
1:12:14 loudly the overconfident spy Noir
1:12:18 shadowy strong contrast lighting a spy
1:12:20 enters a room with an exaggerated sense
1:12:22 of
1:12:24 importance tell kinetic coffee drinker
1:12:27 the polite
1:12:28 robbery Quenton Tarantino inspired okay
1:12:31 let's just grab this
1:12:33 one all right let's just see how these
1:12:35 two let's see if chat GPT and Halo can
1:12:38 work together or if you're going to be a
1:12:41 pain in the
1:12:43 ass all right we're off are we off
1:12:50 [Music]
1:12:55 this was a cool
1:12:58 one it didn't follow the source image at
1:13:01 all it like invented a new car but it
1:13:03 invented a cool new
1:13:06 car and it's got the wheels are actually
1:13:08 rolling which if you've tried to do this
1:13:10 with any of the other video things the
1:13:12 wheels never
1:13:14 roll and the physics of the ocean look
1:13:16 decent
1:13:27 by the way to subscribe for this it's 10
1:13:29 bucks a month for 1,000 credits and the
1:13:32 videos are 30 credits each so that's
1:13:35 three * 10 that's 30 10 bucks a month
1:13:39 for 30 video clips or 95 bucks a month
1:13:44 for unlimited which is a lot the video
1:13:48 things are a lot of money they're a lot
1:13:51 of money people it's a lot of cash all
1:13:53 right while let's doing that we'll go
1:13:54 look at
1:13:56 so this is illuminate or
1:14:00 illuminati.com illuminate. goole.com
1:14:04 and if we go to the home here um
1:14:09 okay if you haven't heard there's a new
1:14:12 tool out there called well no there's a
1:14:14 tool out there from Google called
1:14:15 notebook LM which is a really cool tool
1:14:18 in and of itself you can upload like 20
1:14:20 or 30 PDF files and then just start
1:14:23 interacting with them like their chat
1:14:24 GPT
1:14:26 right so you could take like all of your
1:14:27 annual reports and just go start asking
1:14:29 questions about your annual reports
1:14:31 really cool They just added a feature
1:14:34 called audio overview that makes like
1:14:36 this podcast back and forth between this
1:14:38 woman well they have now created
1:14:41 illuminate. goole.com which is just the
1:14:44 podcast maker so if I go to
1:14:51 generate and then I go grab
1:14:57 this document which is this is the
1:14:59 attention is all you
1:15:01 need white
1:15:04 paper uh which is where Google invented
1:15:07 the Transformer so I'm going to put in
1:15:08 I'm going to paste in the URL of a PDF
1:15:11 and now this is
1:15:12 weird over all the way over here to the
1:15:16 right is a plus button so you have to
1:15:19 hit that plus button and now our source
1:15:22 is attention is all you need and I guess
1:15:24 what what I'm what I'm putting together
1:15:26 here is I can upload more than one PDF
1:15:29 here I just put in another URL hit
1:15:31 another plus it'll keep adding them
1:15:33 there but now look what they did here
1:15:35 this is pretty
1:15:37 slick so this is a little mad lib style
1:15:40 thing create a discussion tailored for
1:15:44 a uh let's see for
1:15:48 a um
1:15:51 sales
1:15:53 audience we'll make it short with a
1:15:56 short
1:15:57 duration and the tone should be
1:16:04 inspiring all right so we're going to
1:16:06 talk about a white paper about a
1:16:08 technology for for making AI systems for
1:16:11 a sales audience it should be a short
1:16:14 duration and it should be inspiring I'm
1:16:16 hoping Tony
1:16:17 Robbins all right so then we say
1:16:20 generate and now it goes off it'll take
1:16:21 it about 30
1:16:23 seconds and and
1:16:26 it'll make a thing so while that's
1:16:27 generating I'll play the one we did
1:16:29 yesterday this was a short length for a
1:16:31 highly technical audience and it was
1:16:34 hilarious the spice of life but why is
1:16:37 this scaled do product tension so
1:16:40 crucial is it just faster is it just
1:16:43 this conversation is powered by Google
1:16:44 illuminate check out illuminate.
1:16:47 goole.com for
1:16:49 more welcome folks to another hilarious
1:16:53 Deep dive into the world of Nur networks
1:16:56 today's paper attention is all you need
1:16:59 is a real knee slapper it ditches the
1:17:02 old school recurrent and convolutional
1:17:04 networks for something attention
1:17:06 grabbing
1:17:08 question I'm planning on on subscribing
1:17:11 to repet is it worth it um I think repet
1:17:14 is probably worth worth the time to uh
1:17:17 subscribe to do not for any of these AI
1:17:20 tools right now don't go yearly just
1:17:23 because things are moving too fast just
1:17:24 go monthly
1:17:26 um but I think it's probably worth a
1:17:27 month to play with rep yeah I do um so
1:17:31 anyway that was a technical one so
1:17:32 here's the one for the sales audience
1:17:34 that should be inspiring this
1:17:36 conversation is powered by Google
1:17:37 illuminate check out illuminate.
1:17:39 google.com for more. google.com welcome
1:17:42 to the discussion sales professionals
1:17:44 today we're diving into a groundbreaking
1:17:46 paper attention is all you need which
1:17:49 revolutionized sequence transduction
1:17:51 models ahtr the Transformer a network
1:17:54 architecture based solely on attention
1:17:57 mechanisms what's the core Innovation
1:17:59 here and why is it so impactful for
1:18:02 sales the core Innovation is replacing
1:18:04 recurrent and convolutional neural
1:18:06 networks the traditional workhorses of
1:18:08 sequence modeling with a purely
1:18:10 attention-based
1:18:12 mechanism this allows for significantly
1:18:14 greater parallelization during training
1:18:17 leading to faster and more efficient
1:18:19 model
1:18:20 development
1:18:23 see when I when I hear these
1:18:27 words I am reminded how not
1:18:30 mathematician I
1:18:33 am but anyway that's illuminate.
1:18:36 goole.com you can put any data source
1:18:38 you want in there and go make yourself a
1:18:41 um podcast pretty slick huh it sounds
1:18:45 like Pate it did sound like
1:18:49 Pate pavin thanks you're welcome or
1:18:52 whoever you were thanking oh yeah you
1:18:54 were asking me about do I think it's
1:18:56 worth it yeah I do think it's worth it
1:18:58 50 sources now but who's counting
1:18:59 500,000 words each wow that's
1:19:02 amazing 50 sources 500,000 words each
1:19:08 that's 25 million
1:19:15 no 2.5
1:19:18 million two oh 12.5
1:19:22 million but that's way more than 2
1:19:24 million
1:19:27 I I can't that makes my head
1:19:30 hurt these tools are getting too
1:19:33 powerful to for me to comprehend anymore
1:19:35 they really are just like what's the
1:19:38 name that's one a
1:19:43 [Music]
1:19:54 day 25 million yes from Steven Johnson
1:19:57 wow that's amazing oh yeah cuz it's 50
1:20:01 50 yeah of course it's 25 million 50 and
1:20:04 it's half a million per is that half a
1:20:07 million tokens or half a million words
1:20:09 good Lord that's a lot chat does my math
1:20:13 for me now where was this Tool 30 years
1:20:16 ago it even said strawberry had two RS
1:20:28 the the whole thing of strawberry having
1:20:30 two RS is
1:20:32 fascinating words they say liquid AI
1:20:35 yeah liquid AI is a new
1:20:39 um is a new
1:20:44 architecture that is a potential
1:20:46 replacement for the Transformer that's
1:20:49 way more efficient way more direct way
1:20:51 more way less translation I don't know I
1:20:54 don't I don't know enough about it there
1:20:55 you can demo if you go to liquid. a you
1:20:57 can actually demo the new technology but
1:21:00 it's just like using a a large language
1:21:03 model yeah unless you are paying
1:21:06 attention to the back end of this [ __ ] I
1:21:08 would just wait for it to get more
1:21:11 real but but I'm tired all right I'm
1:21:14 gonna get out of here I'm I I I can't
1:21:16 open my eyes anymore people um thanks
1:21:20 for hanging out here um if you have
1:21:22 not you really should do yourself the
1:21:25 favor of checking out the salon the AI
1:21:28 salon so go to the salon. you can read
1:21:31 about what we're about there and then
1:21:33 click on the button that says join our
1:21:34 community that'll take you over to the
1:21:36 community site and there's a little
1:21:38 seven-step onboarding process including
1:21:40 there's a welcome video with myself and
1:21:42 Leah faston who founded it um talks
1:21:46 about our values you can introduce
1:21:48 yourself all sorts of stuff like that um
1:21:51 so go check that out this community is
1:21:55 powerful and
1:21:57 growing um and I think being in
1:22:01 communities like this one moving forward
1:22:04 future of work kind of stuff is going to
1:22:06 be a requirement more than a nice to
1:22:08 have so get in early and get
1:22:12 trusted see here's the
1:22:14 thing in this channel the people that
1:22:17 hang out here they call themselves The
1:22:20 Irregulars they're not fooling anyone
1:22:23 they're [ __ ] weird but that's neither
1:22:25 here nor there the weirdos that hang out
1:22:28 here night after night after
1:22:31 night something is happening to
1:22:34 them other than just being gassy from
1:22:36 the
1:22:39 chili get it IR regular see what I'm
1:22:42 doing there
1:22:48 um they're staying active in this AI
1:22:51 conversation they may be picking some
1:22:53 stuff up for me but most what's
1:22:55 happening is they're playing while
1:22:56 they're hanging out here they're playing
1:22:58 with stuff for themselves or they're
1:23:00 sort of taking notes or filing things
1:23:03 mentally and then playing the next
1:23:05 day I would argue that the people that
1:23:08 regularly hang out in this Channel and
1:23:09 hang out at the salon are way farther
1:23:12 significantly more advanced with this
1:23:14 generative AI stuff and how to use it
1:23:17 and how to leverage it in business than
1:23:19 almost anyone else I meet and I I meet a
1:23:21 lot of
1:23:22 people um
1:23:26 what that says to me
1:23:28 is there is a there's a subset of the
1:23:30 people that hang out here and hang out
1:23:32 here that are that are going to become
1:23:35 trusted leaders in the AI space and
1:23:37 they're going to start getting jobs and
1:23:39 those jobs are going to start requiring
1:23:41 more than
1:23:43 them it's going to require like two
1:23:45 people or five people or 10
1:23:48 people who do you think that the
1:23:52 Irregulars that get their [ __ ] together
1:23:54 and start getting hired like that who do
1:23:56 you think they're going to hire do you
1:23:58 think they're going to go to LinkedIn
1:23:59 and see who's got AI cert
1:24:05 certificates you know I took a prompt
1:24:07 engineering course from Rachel Woods no
1:24:10 they're going to say who do I know in
1:24:13 here that shows up does good work uses
1:24:18 their good critical thinking brain has
1:24:22 Integrity I like hanging out with
1:24:25 that's who they're going to
1:24:27 hire and I think that's going to
1:24:29 increasingly start to happen so it is in
1:24:32 your interest to become a valuable
1:24:35 member of this
1:24:36 community and then you'll become one of
1:24:38 those people that will want to hire
1:24:39 people and guess who you're going to
1:24:41 hire the people that you know and trust
1:24:44 from the community all
1:24:47 right so that's why you go do
1:24:51 that use your real name
1:24:55 introduce yourself don't don't be a
1:24:57 lurker go build [ __ ] share what you're
1:25:00 learning go read our
1:25:02 values one of them is
1:25:04 playfulness don't come in there and be
1:25:06 like AI is really serious we' got to do
1:25:09 AI now this is about making everything
1:25:11 more efficient every efficiency is
1:25:13 everything we're going to make things
1:25:14 more efficient you can do
1:25:17 that but you're going to miss the AHA
1:25:19 moments the holy crap
1:25:23 moments the holy crapp ification of
1:25:28 work so hang out with people that
1:25:32 play and have guts and have smarts and
1:25:36 have integrity and are
1:25:40 generous do you know what's not in
1:25:44 here like douchy stingy
1:25:48 people well well this is this is my
1:25:51 prompt I'm not going to give you my
1:25:52 prompt my prompt is my prompt like this
1:25:54 took me 3 days to
1:25:59 develop these people are like you're
1:26:01 like how'd you do that oh here let me
1:26:03 show you you want the
1:26:07 prompt it's amazing cuz you know
1:26:11 why cuz the next time they need [ __ ]
1:26:14 help the person they gave that to is
1:26:16 going to give it to them do un to others
1:26:20 as you would have them do un to you sir
1:26:22 what are your
1:26:23 qualifications qualifications I have
1:26:27 none I witnessed two amazing Kevin
1:26:30 mallister moments today do tell Source
1:26:32 Camp I'll I'll hang around for that what
1:26:34 were they it's
1:26:36 [Laughter]
1:26:45 amazing a raccoon just almost walked
1:26:47 into me I was really surprised when I
1:26:50 when it moved
1:26:55 Joker you got to St stop hanging out by
1:26:57 the
1:26:59 dumpster laid out a non-existing
1:27:02 warehouse for one client then gave her a
1:27:05 picture of
1:27:11 it there you go did a financial analysis
1:27:14 for a banker after telling him I didn't
1:27:17 do
1:27:19 math do you hear that Source Camp was
1:27:22 with a banker and says
1:27:25 I ain't good I ain't got no book smarts
1:27:28 with that numbers numbers
1:27:30 math and the banker was like oh L listen
1:27:34 to Lady I'm pretty good with the
1:27:36 calculator if you know what I mean and
1:27:38 while he's mansplaining banking to her
1:27:41 she's like uh upload their financial
1:27:44 statement please do a financial analysis
1:27:46 of this
1:27:50 bank and then he's
1:27:53 like I bet he couldn't process it that's
1:27:56 [ __ ]
1:27:58 awesome I'm not good with the maths you
1:28:01 are now everybody's good at everything
1:28:05 what if you lived in a world where
1:28:06 everyone could do
1:28:11 everything who would you
1:28:13 hire well I'd hire the guy that does it
1:28:15 best no no everyone can do
1:28:20 everything well how do they do that they
1:28:23 just ask it and it'll do do
1:28:26 it who do you
1:28:29 hire
1:28:31 well I guess I'd hire my friend s yeah
1:28:36 exactly why well cuz s's nice yeah he
1:28:42 works hard
1:28:44 perfect he seems to know what's going on
1:28:46 with this AI stuff s's your guy better
1:28:49 call S
1:28:56 all right I'm leaving I'm telling you I
1:28:58 got to go I don't know why do you keep
1:29:00 me here no more gatekeeping exactly all
1:29:03 right pavin my daughter fired me and now
1:29:07 uses chat gbt as her math
1:29:10 stor yeah the tutor tutors tutors are
1:29:13 going to have a rough go of it I mean
1:29:15 seriously you
1:29:19 know um
1:29:35 [Music]
1:29:47 hey Quinn I got a situation
1:29:50 here what's up Kyle hit me with it well
1:29:53 see I'm nearly 60 I'm old I've got gray
1:29:56 hair and just things are going south
1:29:59 quick but listen that's not the point
1:30:01 point is this I was never good at the
1:30:02 math you know like the book learning
1:30:04 mathy kind of mathy MC mathington kind
1:30:07 of stuff and so I was hoping maybe you
1:30:09 could help me figure out what's the
1:30:10 difference between algebra trigonometry
1:30:12 and
1:30:15 calculus all right let's break it down
1:30:18 algebra is like the gateway to advanced
1:30:21 math dealing with symbols and the rules
1:30:23 for manipulation them think solving
1:30:26 equations finding unknowns and working
1:30:29 with variables but what if I don't is
1:30:32 Quinn what if I don't
1:30:34 care if you don't care about the details
1:30:37 here's the quick and dirty algebra is
1:30:39 solving for x trigonometry is playing
1:30:42 with triangles and calculus is about
1:30:45 understanding change think of it like
1:30:46 this algebra is the grammer trig is the
1:30:49 shapes and calculus is the magic of how
1:30:52 things move
1:30:54 good Lord quinnn I think I actually
1:30:56 understand a concept I've never
1:30:57 understood before that's pretty good of
1:30:59 you if you if I wanted to learn like
1:31:03 learn Matthew MC mathington Could you
1:31:05 teach it to
1:31:06 me absolutely we can tackle Matthew mcmt
1:31:11 together one concept at a time we can
1:31:13 start with the basics
1:31:15 [Applause]
1:31:19 and what if everyone knew
1:31:23 everything what if
1:31:26 everyone knew
1:31:30 everything what would work look like if
1:31:33 everyone knew
1:31:36 everything what would work look like if
1:31:38 you had
1:31:40 10 genius level
1:31:44 assistants that did anything you asked
1:31:46 them to
1:31:51 do genius level
1:31:55 like
1:31:57 Einstein at every
1:32:01 subject what if you had a hundred what
1:32:04 if you had a
1:32:07 100,000 agents working for you on your
1:32:11 behalf at that level what could you do
1:32:15 what could you not [ __ ]
1:32:19 do that's where we're headed
1:32:26 I only laugh because I'm
1:32:28 [Laughter]
1:32:34 uncomfortable if everyone knows
1:32:36 everything then no one knows anything
1:32:39 yeah there's there's there's some truth
1:32:40 to
1:32:42 that it's going to be fascinating to see
1:32:45 what Rises what
1:32:47 Rises you know in
1:32:50 value when everyone can do
1:32:53 everything or when no one has to do
1:32:57 anything is another way to put
1:33:00 it world will be filled with one person
1:33:03 unicorn companies it probably will
1:33:06 be emotional intelligence Rises
1:33:10 yep I think so Lori the the the only the
1:33:14 the like no matter how I play this thing
1:33:19 out I think the robots drive us closer
1:33:22 together as people because they're going
1:33:25 to pick up all the
1:33:27 [ __ ] and they're going to force us
1:33:29 to confront what is our actual
1:33:34 value not what is our value if we're a
1:33:37 cog in the wheel of an industrial
1:33:39 machine Let's assume that those cogs in
1:33:42 the wheel of the industrial machine are
1:33:44 going to get done by robots and then
1:33:46 that forces us to say well what's your
1:33:49 actual
1:33:50 value why would anyone value you
1:33:55 if you're just a nasty piece of Flesh
1:33:58 that no one can stand but you had really
1:34:00 good
1:34:01 skills I think it's going to be a rough
1:34:03 patch for
1:34:05 you you got some you got some learning
1:34:07 to do you got some splaining to do sorry
1:34:12 I was such a douche when I was the only
1:34:14 one that knew how to do that
1:34:15 thing that's all right Bob no one wants
1:34:18 to work with you though you're a douche
1:34:20 I know
1:34:26 poor Bob you kind of got to feel bad for
1:34:29 him right he had like three decades of
1:34:33 just slamming people's face in the mud
1:34:35 CU he was so much better than them he
1:34:37 was so Superior and now they're all
1:34:40 better at the [ __ ] he was good at than
1:34:42 he is cuz he stopped
1:34:44 learning thank you kylin the Irregulars
1:34:47 arising Salon lifts all Irregulars
1:34:50 valuable indeed indeed
1:34:54 Lucy you got some splaining to
1:34:57 do I'm out see you tomorrow all right
1:35:00 Silver Fox I'm out too all right uh
1:35:03 follow my channel subscribe to my [ __ ]
1:35:05 and then let's see who's in here tonight
1:35:06 Silver Fox Tobias Joey uh Brandon great
1:35:10 to see you Lord digital Gods thank you
1:35:12 for the gifts as always uh Winston thank
1:35:14 you for the gifts as always steo in the
1:35:16 house uh against the
1:35:18 norm uh Lori K great to see you chat GPT
1:35:23 therapist
1:35:25 to the rescue for hob
1:35:28 hob human something I don't know what
1:35:31 hob
1:35:32 is maybe it was a typa um spin B3 kakee
1:35:39 against the norm cam Ken great to see
1:35:41 you why does there have to be a
1:35:44 hierarchy when it comes to knowledge
1:35:46 exactly well there doesn't
1:35:48 anymore um I phrased it the other day as
1:35:51 let the AI do the civilization thing
1:35:53 well while we go in play in the woods
1:35:56 yeah playing those people that play
1:35:58 better are going to be higher valued I
1:36:01 like why why not like it's the only
1:36:02 place I go you because you don't need to
1:36:05 be a contributing member of society in
1:36:07 the in the work sense in the 20th
1:36:10 century industrial machine sense that
1:36:14 stuff's going to
1:36:15 all you know now listen don't get me
1:36:18 wrong is this going to happen next year
1:36:20 no this is like 5 10 years out so you
1:36:23 got time left you note in the thread
1:36:24 Kyle okay great I will go look at
1:36:29 that what why am I oh because I was in
1:36:33 the
1:36:36 wrong oh okay I'll do that Brandon I'll
1:36:38 let you know good night Cam kin all
1:36:40 right everybody
1:36:42 aloha aloha
1:36:46 aloha Papa JJ Peter
1:36:49 S
1:36:51 vud Peter
1:36:55 carried Source Camp great to see
1:37:00 [Music]
1:37:02 you pavin in the
1:37:04 house Mina in the
1:37:07 house all right I'm out tomorrow's
1:37:10 Wednesday it'll be a normal time I will
1:37:11 see you 8:00 p.m. mountain time here be
1:37:15 here or don't be here those are your
1:37:18 options peace have a good night