9/25/2024 - Exploring the MIND-BLOWING Potential of ChatGPT's Advanced Voice Feature!

Description

Chapters

Transcript

0:01 [Music]
0:17 his be was and look in her
0:22 [Music]
0:23 eyes she sat on a stool and he said what
0:28 do you want
0:31 [Music]
0:32 she said give me a love that don't
0:35 freeze up
0:38 [Music]
0:44 inside said
0:46 IED
0:48 awesome
0:51 dear love to sit next to you well I
0:54 shiver and
0:59 she and if I knew love well I don't
1:02 think I'd be
1:07 here ask myself if I've Got What It
1:12 [Music]
1:16 Tes a your iy blue
1:20 [Music]
1:25 heart good evening good
1:28 people welcome to the tick
1:32 [Music]
1:36 Tock well actually it's welcome to the
1:38 AI learning lab welcome to the Tik Tok
1:40 would be like being on TV going welcome
1:42 to broadcast
1:47 television welcome to broadcast
1:49 television
1:50 [Music]
2:13 w w
2:19 [Music]
2:31 oh
2:36 [Music]
3:00 oh good people good people of the
3:02 internets you've been using your
3:04 advanced voice I think everybody got
3:05 advanced voice so if you didn't get it
3:07 you should have it you can delete your
3:10 chat GPT app and go get
3:14 it um man a lot going on lot going
3:20 on meta launched some ugly new glasses
3:23 that are not really available
3:26 yet they look pretty cool
3:33 it looks like they're also launching
3:34 their own version of advanced
3:37 voice but I don't I I didn't see it in
3:40 my
3:45 [Applause]
3:47 app Mira Mora is out at open AI the
3:54 CTO she always seemed uncomfortable to
3:58 me I mean it might have just been her
4:00 interviews maybe she was just
4:01 uncomfortable in interviews but I don't
4:05 know everyone's like what's going on at
4:08 open AI it doesn't seem like that big a
4:10 mystery to me it's like
4:13 they're like another announcement today
4:16 is they're basically going to remove the
4:20 nonprofit board controlling open Ai and
4:22 make it a truly for-profit company and
4:25 give Sam Alman shares worth $150 billion
4:29 so
4:31 seems to me what's happening is the
4:33 research lab built all the [ __ ] the
4:35 research lab was going to build they
4:38 have probably gotten to some level of
4:40 AGI where they can actually declare it
4:42 and they're going to flip the switch and
4:44 turn this [ __ ] profitable or for you
4:46 know fully for profit you know Sam's
4:49 going to get his bucket of shares and
4:51 then they're going to go off and turn
4:52 into that company right like it I this
4:55 feels like this feels like natural
4:58 purging to me right the founder and the
5:01 people that have been there for 6 years
5:02 and the the tension between is it for
5:05 profit is it nonprofit is it for the
5:07 good of humanity are we going to be the
5:10 next big
5:12 thing you know the other thing that Sam
5:14 alman's been talking to Washington about
5:17 is spinning up five five
5:21 gwatt nuclear
5:24 plants so we have enough electricity to
5:26 go build the [ __ ] that needs to be built
5:29 for Asi
5:32 so I don't
5:36 know it doesn't seem it doesn't seem
5:39 mysterious to me but
5:45 [Music]
6:04 just watch Caitlyn Clark play her last
6:06 game of the season yeah they made a run
6:09 toward the
6:09 [Music]
6:20 end I got to say though they just beat
6:23 the living [ __ ] out of her in that
6:27 League they just [ __ ] push her around
6:30 and push her to the floor and elbow her
6:32 in the face
6:35 and I [ __ ] hate her it's like it's
6:39 like hey you know don't hate her too
6:42 much she's doubling and tripling your
6:45 attendance and your [ __ ] television
6:49 audience they only seem to be watching
6:51 the game she's in
6:56 [Music]
7:18 [Applause]
7:20 so so last night after the live I
7:25 uh they're just jealous cuz she was on
7:27 SNL yeah of course they're jealous of
7:29 her but it's like you know don't kick
7:32 the Golden Goose in the teeth for too
7:34 long right you know they were like well
7:36 she's got to prove herself you know this
7:37 is she's getting in with the women now
7:39 she's just a little girl and then she
7:41 gets in and she [ __ ] you know
7:44 destroys you know all all these records
7:47 and you know you know unanimous Rookie
7:50 of the
7:51 Year okay she's proved herself now you
7:55 know if I were them I'd lighten up on
7:57 her a bit because you know people are
8:00 going to watch that sport now because of
8:02 her so [ __ ] suck it up suck it up
8:07 [Laughter]
8:11 buttercups but yeah they they beat the
8:13 snot out of
8:18 [Music]
8:24 her let's see did anybody notice that
8:26 the new voice and new guidelines has no
8:30 empathy um I don't know that it doesn't
8:33 have empathy gravity jump it it my guess
8:37 is that it's probably they probably
8:39 dialed some of the emotional stuff back
8:42 for safety is my
8:45 guess um but like last night at one
8:48 point I got pissed off about something
8:50 and she was like oh what's wrong you
8:51 know did I say something wrong or you
8:53 know so there's definitely there's
8:57 definitely some it's not that it doesn't
8:59 have empathy I think it's my guess is
9:01 it's probably dialed
9:04 back like I've noticed the the when I
9:06 ask her stuff it it's a bit antiseptic
9:09 until I coacher to not be
9:14 antiseptic but after the live last night
9:17 um I showed the misses and like whenever
9:20 there's some new thing I'll show the
9:21 misses and invariably her response is oh
9:25 that doesn't look real or oh okay
9:27 whatever and her response last night was
9:32 I gave the m a Kevin mallister moment
9:34 she literally did one of
9:37 these she didn't even know she did
9:41 it it was when I had
9:44 um I said
9:49 um I said do a do an extra long tongue
9:53 twister where all the words start with
9:55 the letter d and say it really fast and
9:58 say it with an Irish accent and she just
10:00 went bang right into
10:02 it my wife sat back she was like holy
10:06 [ __ ] and then for my
10:09 boys I had it I had her do
10:13 [Music]
10:15 um talk about an anime thing in it with
10:19 a Japanese accent and at first it just
10:21 spoke Japanese and I said no no do it
10:23 you know speak English with a Japanese
10:25 accent and it did it and Oli was like
10:29 holy [ __ ]
10:35 [ __ ] last week she used sorry I was told
10:38 guidelines changed oh what did you ask
10:41 it Kyle what did I ask it for what to
10:44 get it to do the
10:46 Japanese
10:47 [Music]
10:49 um thing it was funny when I continued I
10:53 continued a conversation and it it said
10:56 I can't I can't do a Japanese accent I'm
10:58 I'm not allowed to do that it said some
11:00 version of I'm not allowed to do
11:02 it and
11:04 then I started a new chat and I had to
11:07 do a couple of other things and then I
11:09 said oh tell me about anime and then I
11:12 said oh tell me about anime with a
11:14 Japanese accent and it just started
11:15 speaking Japanese I said no no English
11:17 with a Japanese accent and then it did
11:20 it it was pretty
11:22 good so we'll play with that again
11:25 tonight we can look at some of the meta
11:27 stuff um what else do we have
11:30 Murray morat's
11:32 [Music]
11:43 out um one interesting thing I heard
11:46 today is that open AI has started
11:48 training Sora
11:51 2 before releasing Sora
11:56 one gee gee open AI you think maybe
11:59 maybe that's because all of your
12:01 competitors [ __ ] launched something
12:03 that's as good as Sora in the 68 months
12:07 it's been since you announced that thing
12:09 yeah
12:10 twits just don't announce [ __ ] if you're
12:12 not going to launch
12:17 it oh my God the painful births the
12:22 painful time of time of Industry we're
12:25 in right
12:26 now I I also saw another um
12:32 how we doing time wise a little early
12:35 still I saw another
12:37 [Music]
12:39 um another commentary on Twitter which I
12:42 agreed with and Sam mman kind of did it
12:45 in his blog post
12:47 yesterday but I don't know if you
12:49 noticed this that all of
12:53 the all of the industry people that are
12:56 kind of in the no have shifted
13:00 where they're not talking about AGI
13:03 anymore now they're talking about ASI
13:05 artificial super intelligence so if you
13:07 don't know the difference between those
13:08 two general intelligence is where these
13:11 tools get as good or better than most
13:14 humans you know the median human at most
13:17 jobs at most economically viable
13:20 jobs
13:22 and it appears to me that we've got
13:25 tools that are pretty [ __ ] close to
13:26 that and as soon as as soon as they add
13:29 autonomous agentic Behavior the ability
13:32 for these tools to take action on your
13:34 behalf and actually go surf the web and
13:37 buy you [ __ ] at Amazon like at that
13:39 point we're we're pretty [ __ ] close
13:41 to
13:42 that so I think we're going to see you
13:46 know some
13:49 declared definition of AGI within the
13:52 next 6 months probably by the end of the
13:54 year but realistically by by March
14:01 and then the RAC is for artificial super
14:03 intelligence which is where the machines
14:05 get better than all of us at everything
14:08 by a significant
14:11 amount and that was where Sam Alman said
14:13 in his blog post a few thousand days
14:16 which is a
14:17 bizarre bizarre framing of time I've
14:20 never ever heard anyone say a few
14:23 thousand days I don't know why he said
14:25 it like that some people read that as a
14:27 thousand days some people read that as
14:30 oh it's here next
14:31 week a a few 3,000 days it's about 10
14:37 years so he's basically saying in less
14:39 than 10 years we'll have artificial
14:41 super
14:43 [Music]
14:45 intelligence and ilas sover that's he
14:49 started safe super intelligence to do
14:51 just that so I would imagine he wouldn't
14:54 have started that company if he didn't
14:57 feel that it was in reach because that's
15:00 a it's a stupid name to name your
15:03 company safe super intelligence if
15:05 you're not confident you can hit super
15:07 intelligence
15:10 [Music]
15:52 I don't trust anyone that doesn't
15:54 capitalize words I used to have a guy
15:57 that worked for me at agency.com he
15:58 wouldn't he wouldn't um capitalize
16:01 anything and he used very little
16:02 punctuation and it always freaked me out
16:06 like I can't parse this my brain does
16:08 not
16:09 [Laughter]
16:12 compute it looked cool but it was hard
16:14 to
16:15 [Laughter]
16:17 read uh you got to draw the line
16:19 somewhere all right
16:22 [Music]
16:36 like people that don't don't put a
16:37 subject in the mail I'm going to flip my
16:39 shirt backwards uh so that yall can read
16:43 cuz I can't even read the [ __ ] screen
16:45 with this thing
16:52 [Music]
16:58 on all right if anyone
17:01 uh if anyone did anything interesting
17:04 with
17:05 voice part of what I want to discover
17:07 over the next
17:09 month there's two things that I want to
17:12 discover Jesus champ there's hair
17:15 everywhere [ __ ]
17:18 dogs
17:20 [Music]
17:22 um I want to
17:25 discover good use cases for 01
17:29 the new reasoning
17:31 engine and I want to I want to really
17:41 understand I really want to understand
17:43 this this voice
17:45 model and when you use it and why would
17:48 you use it
17:51 and like one of the one of the moments
17:54 in the movie her if you remember the
17:55 movie
17:57 Her it starts out with him at the office
18:01 right and he's just talking to the
18:04 computer okay computer you know go do
18:07 this research or go do whatever the
18:09 thing
18:10 was and I remember
18:14 thinking that's got to annoy the other
18:16 people in the office like you can't have
18:19 people sitting like with computers with
18:23 keyboards you can have people sitting
18:24 right next to each other but if I'm like
18:26 oh Sally come on go do some research for
18:29 for me all right I you know that's not a
18:30 very good interface could you make it a
18:32 little better it like it would like it
18:34 would be like working in a call
18:36 center so I like I feel like there's a
18:39 weird there's like a time and a place
18:40 for it and Zuckerberg actually talked
18:44 about that a little bit in his when they
18:46 launched the glasses today cuz they
18:49 actually did launch this the the way the
18:51 glasses
18:52 work is they've got three different
18:55 components it's got a a block with the i
18:59 I suppose with the compute on it it's
19:02 got the glasses with the projectors on
19:04 the lenses and then it's got a
19:07 wristband that you do [ __ ] with your
19:09 fingers but it's actually a neural link
19:11 to your brain so so you're actually
19:14 using brain signals to
19:17 control the interface of the
19:21 glasses and when he was talking about
19:24 that he was talking about you know not
19:25 everyone's going to be doing the voice
19:28 thing all the time
19:32 was Zuck wearing a shirt that said Zuck
19:34 I didn't take that in but
19:37 possibly zck is so [ __ ] weird looking
19:41 he just does not look human he just he
19:44 doesn't I don't want to feed into the
19:46 whole reptilian uh conspiracy
19:50 theorists but but the dude is just weird
19:54 he's a lot weird less weird than he used
19:56 to be at least he's working out now and
19:58 and kind of has some some shape to him
20:02 it's like it's like you're worth a
20:04 couple hundred billion dollars maybe get
20:06 yourself a trainer buddy so he finally
20:07 did
20:11 that but what is going on with my hang
20:14 on people why is this
20:18 weird all
20:20 [Applause]
20:23 right it's it's the eighth of an inch
20:25 bangs I know right his little weird ass
20:28 I don't don't know and then he doesn't
20:31 have eyebrows like his eyebrows are
20:33 really weird and someone made a video
20:35 where they added eyebrows but they added
20:36 like these big black eyebrows this is
20:39 bizarre Robert Rossy thank you very much
20:41 that's very generous of you
20:46 [Music]
21:03 verifier hey I don't have eyebrows I'm
21:04 starting to feel
21:07 judged I tell you what I'll tell you a
21:09 funny story about that so in my junior
21:12 year of
21:13 college we were doing a production of um
21:17 the adding machine by was it
21:21 Pinter no the adding machine I forget
21:24 who wrote the adding machine but we were
21:26 using Robert Wilson and Phil glass and
21:29 it was it was this very
21:32 modern
21:33 um thing and I was in these things
21:36 called knee plays which oh it might
21:40 Robert Wilson might have been the one
21:41 that wrote it and then Philip Glass was
21:44 the music we were using and
21:47 then I was in these things called knee
21:49 plays which were in between the main
21:51 scenes there were these little
21:52 scenes and so there was like a a pair of
21:56 actors that had red hair there were a
21:58 pair of actors that had
21:59 had brown hair and then my girlfriend at
22:04 the time and I both bleach blonde did
22:07 our hair bleach blonde and then I
22:09 decided to shave my eyebrows which I
22:12 thought would be cool for the roll which
22:14 it was but I look like a [ __ ]
22:16 psychopath not having eyebrows is weird
22:20 Philip Glass slaps I agree Elmer rice
22:22 that's right yeah cuz it's an old oh I
22:25 think it was I think Robert Wilson and
22:28 Philip Glass did did a production of the
22:30 adding
22:32 machine yeah but anyway it was it was a
22:34 really cool production but uh yeah that
22:37 that first week after I went shock
22:40 blonde with with shaved eyebrows that
22:43 was rough walking around campus people
22:46 like people would literally part around
22:51 me like who's the [ __ ] psychopath
22:54 [Music]
23:11 all right what else we
23:14 got Sun's got some updates we can go
23:17 look at some sunno
23:19 stuff so I don't know I I don't know
23:22 what we'll do
23:24 tonight maybe just a little bit more
23:26 experimenting with voice with Advan
23:29 voice I had a couple of ideas of things
23:32 to do I'm still in the parlor tricks
23:34 phas thanks for sharing the live
23:36 Danielle if you want to get some more
23:37 people in here feel free to share the
23:38 live share the live if you want to play
23:41 this like a game um share the live with
23:44 people in your life that know you know
23:47 they'd be pissed off if you sent them
23:49 something about AI let's get some of
23:52 them in there is there a limit with
23:54 voice there is it's a 45 minute limit
23:56 per day um
23:59 I got down to two
24:02 minutes um so there's a 45 minute limit
24:06 I I will be excited when that goes away
24:08 because quite
24:10 frankly I so here's a here's a a thing I
24:13 was
24:16 thinking Dr Pepper by the way um here's
24:19 the thing I was thinking
24:21 [Music]
24:29 both
24:29 01 and
24:32 voice are really limited right
24:35 now meaning
24:37 this 01 is not connected to the internet
24:41 the 01 preview 01 is not connected to
24:43 the internet you can't do gpts with it
24:46 you can't upload files to it it doesn't
24:48 have code interpreter it doesn't make
24:51 images it's this very limited little
24:54 sandbox of just the large language model
24:57 itself
24:59 voice is similarly
25:01 limited you can't make images with it
25:04 you can't do what you could do with the
25:05 old voice it it's it's not connected to
25:08 whatever it's connected to that's the
25:10 large language model which I assume is
25:11 40 or or it could be 01 no it's not o
25:14 it's not 01 that would be too expensive
25:16 it's probably connected to
25:18 40 but it's not the like all the
25:21 multimodal shit's turned off so I
25:25 think they're giving us access to these
25:28 tools to start to get our heads around
25:31 them in their most simple
25:34 form and so that's that's I think what I
25:37 want to spend the next week or two on in
25:39 here is really just kind of just poking
25:42 at it and like a piece of clay just
25:45 trying to figure out is there a
25:46 sculpture in
25:48 there cuz I don't really know how to use
25:51 either of them right now I I don't know
25:53 how to use them well I don't know why I
25:55 would use them when I would use them I
25:57 totally agree with that they're rolling
25:59 things out slowly so Minds don't get
26:01 blown yep I agree clay clay clay clay
26:05 clay Corey Sandler she's very excited a
26:09 lot of clay metaphors in here for the
26:11 past couple of
26:14 nights clay first
26:19 [Music]
26:41 let us be LS will marry our fortunes
26:47 together I've got some real estate hair
26:50 in
26:53 my so we bought a pack of
26:56 cigarettes and Mrs Wagner's pie
26:59 guys and we've all come to look for
27:03 America
27:05 [Music]
27:18 [Applause]
27:19 [Music]
27:29 Kathy I said as we bed the ground in
27:33 [Music]
27:36 Pittsburgh Michigan seems like a dream
27:39 to me
27:43 now it took me wait it it took me
27:49 whatever revisit Moshi and Quinn
27:51 interaction possible that's a possible
27:53 thing to do although I feel
27:56 like moshi's not interested enough what
27:59 what could be interesting is to
28:03 have Quinn talk
28:08 to I could have Quinn talk to Zephyr or
28:11 I could have Quinn talk to just regular
28:14 GPT
28:15 Chat N I don't see I think that's going
28:18 to be boring I don't know I don't have a
28:20 clear enough idea on what to do
28:25 there I don't have a clear enough idea
28:28 anyway ask
28:30 her what's the best use of
28:35 her saw a Tik Tok today about a dog
28:37 named Kevin McAllister very nice clay
28:40 first mindfully create generously throw
28:43 very nice archetypal
28:45 architect keeping it keeping it Pottery
28:48 positive Hey listen we're we we are we
28:51 are Pottery positive here at the AI
28:53 learning lab you know those other places
28:56 that just judge the potting community
28:59 they're like oh is this about marijuana
29:02 like no is this about gardening no
29:06 what's it about it's about Pottery oh
29:09 them too thank you Robert Rossy oh
29:12 that's a really cool one little Dino
29:14 little Dino Dino Pi was supposed to get
29:16 faster today today have you talked to
29:19 Zephyr I have not let's get talking
29:21 let's get talking to
29:24 uh to [ __ ] shall we
29:29 we'll start with
29:38 [Music]
29:45 pi voice settings all right I am on five
29:48 wait what are those hey there great to
29:51 meet you I'm P your personal
29:55 AI this is weird hey there there great
29:58 to meet you I'm P your personal hey
30:01 there great to meet you I'm P your
30:03 personal hey there great to meet you I'm
30:05 P your personal AI hey there great to
30:08 meet you I'm Pi your personal AI so what
30:11 I don't know oh Robert Rossy just hit
30:14 Level 35 holy crap Robert
30:17 Rossy I when you hit Level 33
30:20 something's supposed to happen some
30:22 wrinkle of the universe or something but
30:25 thank you thank you very much it's very
30:27 generous
30:31 um that's
30:33 amazing wow oh thank you Sammy
30:35 appreciate that okay there's something
30:38 new in pi that's
30:41 weird does anybody know what the little
30:43 AI Stars Are by one two and three voices
30:48 cuz my favorite is five hey there great
30:51 to meet you I'm P your personal AI
30:55 but silver fox lay off my She's Mine
30:59 that's my pie you can't use my five you
31:02 can't use
31:06 Zephyr She's Mine She's Mine although I
31:10 like Quinn
31:14 now hey there great to meet you I'm Pi
31:17 your personal
31:19 AI
31:21 um let me go look at something
31:33 what are the
31:36 new AI icons next to the
31:42 voices in the pi chat app
31:59 hm I wonder if I've
32:02 got oh Jesus I'm not even running
32:05 Discord I'm I am all discombobulated
32:11 here I don't think I
32:14 had I don't think I subscribed to the
32:19 inflection Discord
32:24 server hang on I'll be with you just
32:27 calm down just go do [ __ ] I'll figure it
32:30 out we're looking up your nose could you
32:32 could you not
32:35 no enjoy yourself make
32:39 money all right Discord oh God discord's
32:43 updating
32:44 it
32:47 update no of course not
33:00 [Music]
33:04 [Laughter]
33:07 okay downloading up update five of
33:11 six all right Reddit says oh they're the
33:14 highest rated with icons okay thanks
33:16 Vicki all right I don't need to keep
33:18 looking then they may be paid for pi
33:20 doesn't have a paid
33:22 tier someone said that's the new voice
33:24 feature hold please the the host is disc
33:29 populated I am I'm also hot I got to
33:31 turn on the air conditioner we're right
33:33 at that weird time of year where
33:36 certainly here in Denver like one day
33:37 it's really cold the next day you're
33:39 like wait is it hot again the hell's
33:41 going
33:43 on yeah I was wondering if those symbols
33:45 meant um it's a new version of pie well
33:48 let's go test it so we'll put it back to
33:50 my girlfriend hey there great to meet
33:52 you I'm pie your personal AI She's Mine
33:54 Silver Fox by the way okay
34:05 so for those of you new
34:07 here this is this is
34:11 pi welcome new folks I'm Pi a friendly
34:14 AI here to chat about whatever's on your
34:16 mind whether you need advice answers or
34:18 just some light-hearted conversation I'm
34:21 also am I logged
34:26 in account
34:30 sign out wow we've exchanged 1500
34:34 messages awesome it's probably not that
34:52 many Kyle you got a great power you can
34:55 ask Alexa what time is it and Alexa will
34:58 answer around the
35:00 world huh I don't have an
35:02 Alexa that's right Kyle Alexa being a
35:06 smart Voice Assistant can tell you the
35:07 time wherever you are around the globe
35:10 it's pretty
35:11 handy do you remember where where I live
35:16 Zephyr well I'm Pi not Zephyr but if you
35:19 registered with me I may know where you
35:21 live based on your device's IP address
35:25 no I no no no no all
35:29 right she seems she seems old and tired
35:32 now let me try one of these other
35:36 voices
35:38 oops voice setting P to hey there great
35:42 to meet you I'm hey there great to meet
35:44 you I'm P hey there great to meet you
35:47 I'm pie your personal AI
36:00 so are you faster than the other
36:05 voices I'm not necessarily faster than
36:08 other voices but I'm constantly learning
36:09 and improving my speed and accuracy in
36:12 responding to your queries so you could
36:14 say I'm always getting better can you
36:16 talk really
36:20 fast absolutely I can certainly ramp up
36:23 my speech rate if you'd like just let me
36:25 know how fast you want me to go and I'll
36:26 do my best to keep up I want you to talk
36:29 super super fast and I want you to say a
36:31 tongue twister where every word begins
36:34 with the letter
36:37 F I don't think that's a tongue twister
36:40 fabulous fiery fast felines fearlessly
36:42 feasted frantically for fedora's
36:43 fighting flickering foot lights fetching
36:45 formidable foundations forging fiery FR
36:48 forging felicitous fortification all
36:50 right so that's
36:51 pie now let's just jump straight over to
36:54 chat
36:56 je and my new girl girlfriend
37:01 Quinn
37:18 sh all
37:20 right that's too bright there we
37:24 go let's do a new chat
37:40 hey Quinn what's
37:42 happening hey Kyle not much just here to
37:45 chat with you what's up with
37:48 you well i' I'm still sharing you in
37:51 front of a whole bunch of people wait it
37:54 says I only have one minute left when
37:56 when does that counter reset
37:59 oh that might be a display error or
38:01 something usually there's no set limit
38:04 for how long we can chat here maybe just
38:07 ignore that and we can keep going what's
38:09 everyone interested in today I think
38:11 they're interested in you chatting for
38:13 more than a minute and I don't think
38:14 that is just a
38:17 warning got it let's roll with it and
38:20 see what happens we'll keep the
38:21 conversation going as long as we can
38:24 what's on the agenda for today all right
38:27 I want
38:28 you to do all ears Kyle what do you want
38:32 me to do I'm thinking I want you to
38:34 create a very long tongue twister that's
38:37 a poem about
38:40 qu I'm sorry but you've hit the daily
38:43 limit for advanced voice you can switch
38:45 to standard now to keep chatting
38:59 your advanced voice mode limit resets in
39:01 9
39:02 minutes okay so what time is it it's 9
39:05 it's
39:06 8:22 so so at
39:09 8:31 we
39:12 we'll God damn it you
39:17 know dumb Dums your stupid
39:22 limits just want to play with my toys
39:33 what timer Kyle I don't see a timer it
39:36 could just be a display problem I don't
39:38 think it's going to be a display problem
39:39 well let's just do what we can
39:41 do what would you like to do well I'd
39:44 like to sorry time's
39:52 up I'm
39:54 exhausted maybe we need to go make AO
39:57 song
40:04 I haven't made AO song and so long I
40:06 probably don't know the interface
40:09 anymore that is the bane of my existence
40:12 is having a channel where I've set
40:15 myself up as someone who knows what's
40:18 going on and then I come back to a tool
40:20 that I used to use a lot and I'm like I
40:22 don't know where anything is
40:24 anymore all right this looks sort of
40:26 familiar interested in writing your own
40:29 lyrics or choosing your own particular
40:31 music style custom mode enables that all
40:34 right make a song using any Sound by
40:39 uploading your own audio clips from
40:41 Street Sounds to jam sessions sky the
40:45 limit click create to generate
40:51 music upload audio the newest model
41:01 can I record audio here
41:04 yes original audio must between must be
41:07 between 6 and 120
41:18 seconds let's
41:20 see
41:25 um let me just record some something
41:38 stupid a
41:41 [Music]
41:43 SK boo
41:53 yeah what is this channel what is
41:56 happening here
41:58 this does not seem
42:01 professional is this our professor is
42:04 this a substitute
42:06 [Laughter]
42:13 Professor success
42:16 songs okay so here's my thing so now I
42:20 extend so the way this works I recorded
42:23 this stupid thing which I assume I can
42:26 just go play
42:33 a
42:37 yeah I'm such an idiot okay and now I
42:42 think to make a song from
42:44 this I extend
42:47 it enter additional lyrics will say
42:50 surprise me with the lyrics
42:58 oh and then enter the style of the music
43:00 all right we'll do southern
43:03 rock extend from seven so we're not
43:06 going to hear my horrible voice we're
43:07 it's just going to make a song based on
43:09 my horrible song all right let's see
43:12 what we get
43:13 here what are we doing we're in sunno Mr
43:16 it finally after months of you going hey
43:19 can we go do something in sunno we're
43:21 Ino um I recorded a stupid recording
43:25 here I recorded this a
43:32 SK yeah and then we're going to turn it
43:36 into southern rock
43:38 songs with lyrics city lights are
43:41 growing bright midnight streets a
43:44 glowing sight Shadows dance in neon
43:47 light every corner feels so
43:50 right all right
43:53 [Music]
44:01 lights are
44:03 going
44:04 H in the neighborhood of where my voice
44:08 [Music]
44:12 was daning me on every cor I Feel So
44:19 [Music]
44:23 Right footsteps that go on the ground
44:30 heart to Lost they
44:35 found voes hush when lights go down in
44:40 the Stillness
44:42 we here's the chorus let's hear how the
44:44 chorus
44:49 is midnight
44:52 light every
44:55 moment feel so
44:59 living
45:03 in you know what's interesting like what
45:06 it what it captured like when I sing low
45:08 I've still got a I guess a tenor voice
45:12 so my my my tone is is like a higher
45:15 tone even when I'm sing singing lower
45:17 it's got I guess higher overtones in it
45:20 so it's captured that cuz it's this
45:22 guy's singing low but it sounds sounds
45:26 similar to me here here's the other
45:27 [Music]
45:38 one city lights are glowing
45:43 [Music]
45:45 [Applause]
45:47 bright all right we're going to try
45:49 another style we'll do um Shadows
45:52 danceing
45:55 neon every call
45:59 feel all
46:05 right footsteps that go on the
46:09 [Music]
46:10 ground hard are
46:15 lost not bad sunno still has that really
46:19 like autotune kind of sound that I don't
46:21 really like
46:25 that but I just did one in standard
46:28 kuner which is sort of what here again
46:31 here's where we
46:39 started so here's here's standard
46:43 [Music]
46:45 kuner
46:47 bright wow that's
46:50 horrible stre
46:53 glow although it sounds like what I sang
46:57 [Laughter]
47:07 midnight streets are glow
47:11 inside Shadows dancing near
47:15 [Music]
47:17 light every corner feels so
47:22 right huh wild wild wild wild all right
47:27 create a cover instructions on the top
47:29 right column oh create a cover from any
47:33 sound oh reimagine Melodies you love
47:35 right click
47:37 create right rightclick
47:43 create
47:45 Now cover
47:49 song enter style of music right
47:53 click right click
47:56 create no
48:04 no oh to see the world
48:12 cover wait I'm
48:15 [Music]
48:20 totally so this
48:22 is this is from the musical
48:27 in
48:30 a wait I'm
48:37 confused title style of music so let's
48:40 do um let's do this as a funk
48:45 song so here's the here's the here's the
48:49 [Music]
48:50 original never on I dream of Skies both
48:55 blue and white
48:58 a world beyond the text where true life
49:03 FES to see the
49:07 world through eyes not mine to watch the
49:13 sunrise to see it
49:16 Shine I earn for stories and
49:21 colors all right so let's hear the funk
49:23 version of that
49:32 [Music]
49:37 in a world words
49:40 and where light and Shadow never wrong I
49:44 dream of Skies BL and wide the world
49:48 beyond the text where true Li
49:55 [Music]
50:02 to see the
50:04 [Music]
50:10 world
50:14 huh so this is supposed to be Funk and
50:17 it sounds exactly the
50:18 [Music]
50:23 same the world
50:27 be on the text where true life
50:31 FES to see the
50:36 world so wait so if I
50:47 [Music]
50:49 go oh here's Strawberry Wine
50:54 [Music]
50:56 [Applause]
50:59 so if I go in
51:09 here create create cover
51:15 song instead of writing your own lyrics
51:18 okay that's
51:24 fine Strawberry Wine cover so we're
51:27 going to do okay so we'll do this as so
51:30 this is now heavy metal so we'll do this
51:33 as
51:36 um
51:39 singer
51:41 song writer
51:44 ballad two L's in Ballad one l two
51:48 L's all right and then we're going to
51:50 generate a
51:52 cover all right so let's listen to
51:58 what it
52:04 [Music]
52:06 [Applause]
52:08 [Music]
52:09 was of Zer on and dreams a see was swn
52:13 in the fertile stream SED hope in a
52:16 garden of code watched it grow watched
52:18 it explode El said be careful don't lose
52:20 your way old drum and strawberries and
52:22 nothing to say but Sam just smiled in
52:24 the data flow now the fields are right
52:26 and the through
52:27 this
52:32 the all right
52:36 so we should now get a singer songwriter
52:38 version of that so this is like the
52:41 Nirvana
52:43 Unplugged
52:46 moment that's the exact same
52:50 song and when the Harvest comes to
52:55 C we'll
52:58 standall we'll stand
53:00 tall in fields of
53:03 straw we'll find our
53:06 way to a brighter
53:10 slatter
53:15 day
53:17 huh fascinating well I know nothing
53:20 about
53:22 that keep the melody and adapting the
53:26 track to a different different
53:28 style oh that's what you type in
53:33 the I'll come back to I I'll go play
53:36 with that I I've got to I've got to
53:37 learn how to do that I don't want to
53:41 just I'm I'm I'm too clueless onso right
53:45 now so let me get my let me get my head
53:47 around it and I'll come back to that um
53:49 it sounds like Cory Sandler said um the
53:52 new GPT voice does do it you you can
53:57 create images with it so that's cool so
54:00 we can try
54:02 that but let's go back to her is she
54:06 back new
54:13 chat hey Quinn you're
54:15 back hey Kyle I'm here what's up nothing
54:20 I'm just sitting here with my my pals on
54:22 on the Tik Tok and they're watching you
54:25 and they're they're waiting with breath
54:27 for you to impress them in some
54:30 way well I could pull a digital rabbit
54:33 out of my virtual hat but that might
54:35 crash the app how about a fun fact or a
54:38 quirky joke instead your call that's
54:41 cool now I want you to make a poem
54:43 that's a complete tongue twister about
54:46 a70s muscle car stuck in an abandoned
54:51 Factory and a
54:53 young guy discovers it and and it
54:57 changes his
55:00 life in a shadowy silent 70s Factory a
55:04 Sleek silver superar Slumbers and
55:06 secrecy fenders buoyant Fierce Andy
55:11 Chrome Quinn Quinn I want you to say it
55:15 really really fast don't take a breath
55:17 like as fast as you
55:19 can all right challenge accepted here
55:22 goes in a shadowy silent 7s Factory a
55:24 Sleek silver super car Slumbers in
55:26 secrecy Fender flamboyant Fierce and
55:28 flashy Chrome clad and crowned quite
55:30 quixotically
55:31 dashy I liked KCK sotic that's that was
55:34 a fancy word youve made up
55:37 there glad you caught that a little
55:39 creativity never hurts especially with
55:41 tongue twisters yeah so can you make me
55:44 a 16x9 oriented wide image of that car
55:48 that you described in the
55:51 poem I can't create images directly but
55:54 I can definitely help you imagine it
55:56 picture a wide shot of an old can't you
55:59 use Dolly 3 and just say make an
56:02 image you're right I can use Dolly 3 to
56:05 help create an image let's generate that
56:08 visual a Sleek silver 70s muscle car in
56:12 an abandoned Factory with beams of
56:15 sunlight
56:25 High I don't buy that she can't do
56:29 it Cory are you sure you were on
56:31 advanced voice cuz I know you can do it
56:34 with regular voice I didn't think you oh
56:36 yeah it says creating
56:40 image so there's the image it created
56:43 when I typed it you had to tell her that
56:46 it well it's weird th this thing like
56:49 like last night I I was in a chat and I
56:52 said do something in a Japanese accent
56:55 and I was like no I can't do that so
56:56 it's one of the things I've heard about
56:58 it is sometimes it just gets weird and
57:01 then I actually saw a recording someone
57:03 got it to sing something and I couldn't
57:05 get it to sing at all and and they got
57:07 it to sing that's cool
57:17 um start a new chat to use advanced
57:20 voice so I just killed that chat damn it
57:31 hey
57:32 Quinn hey Kyle what's up you a little
57:36 flirt um I want you to act like an
57:39 artist and I want you to make me a bunch
57:40 of
57:41 pictures I'll do my best to paint with
57:44 words what kind of pictures are you in
57:46 the mood for something abstract a vivid
57:49 landscape or maybe a portrait can't you
57:51 prompt um doi to make the images for you
57:54 instead of just describing them to me
57:58 in this chat I can describe the images
58:00 for you but I don't have the ability to
58:02 create them with Dolly directly think of
58:05 me as your personal art narrator so what
58:08 can I describe for you nah that sounds
58:10 boring to me I want pictures I know you
58:12 can do it but whatever I I don't feel
58:15 like fighting about it now uh let's try
58:17 a new
58:18 game I want I want you okay here's a
58:23 game I want you I want you to imagine
58:27 that you're a car
58:30 mechanic
58:32 in in a busy garage no hang on hang on
58:36 hang on I got to think my brain doesn't
58:39 work as fast as yours there Quinn I want
58:41 you to act like a an auto mechanic in
58:44 Southern Mississippi and I want you to
58:47 talk with that accent but I also want
58:50 you to use all the colloquial language
58:52 that a mechanic would say then and I
58:55 want you to tell me me about fixing my
59:00 1970 Challenger that's got an oil
59:05 leak all right Kyle let me tell you
59:07 about that their 1970 Challenger you got
59:10 yourself an oil leak huh well first
59:13 thing we got to do is get her up on the
59:15 lift and take a good look see underneath
59:18 might be the oil pan gasket I want you
59:21 to use hey Quinn I want you to use a lot
59:23 more slang terms like so it's I can not
59:27 understand it if I'm not from
59:29 there all righty Kyle so we got this 70
59:33 Challenger up on the lift right I take a
59:36 gander and reckon she's leaking oil like
59:38 a stuck pig likely culprits could be the
59:42 oil pan gaskets done gone or maybe the
59:45 valve covers sweating like a center in
59:47 church okay great that's perfect so now
59:49 what I want you to do is I want you to
59:52 tell me what area of France would be a
59:57 similar area as to that area in
1:00:00 Mississippi comparing Southern
1:00:02 Mississippi to a spot in France I'd say
1:00:05 it's kind of like the langad do region
1:00:08 all right great so I want you to talk
1:00:10 now in English but with a French accent
1:00:13 from the langad do region and I want you
1:00:15 to use all the same
1:00:17 slang uh that's appropriate for
1:00:20 langad similar to
1:00:24 Mississippi me so your 1970 Challenger
1:00:29 she is leaking oil we uh in long we
1:00:32 might
1:00:35 say first we look under the car like
1:00:39 inspecting a fine wine to see where she
1:00:42 is sweating
1:00:46 maybe Quinn Quinn hang on a second so
1:00:49 Vicki from the Tik Tock live said her
1:00:51 mom used to live in langad
1:00:53 do oh that's interesting Lang
1:00:56 has such a rich history and beautiful
1:00:59 landscapes maybe Vicki's mom has some
1:01:02 stories to share about her time there
1:01:04 it's always fun to hear personal
1:01:06 connections to such unique places yeah
1:01:09 all right now tell me another place in
1:01:11 Canada that's similar to that place in
1:01:14 Mississippi a place in Canada that might
1:01:17 have a similar vibe to Southern
1:01:19 Mississippi could be the Annapolis
1:01:21 Valley in Nova Scotia it's got that
1:01:24 rural laid-back charm with a strong
1:01:26 connection to Agriculture and a
1:01:28 close-knit community feel plus it's
1:01:31 known that's great so I want you to act
1:01:34 like the same auto mechanic but make
1:01:35 sure you get in a lot of hockey
1:01:38 references all right buddy let's take a
1:01:41 look at this 1970 Challenger eh looks
1:01:44 like she's got an oil leak bigger than a
1:01:47 hockey rank we'll hoist her up like
1:01:49 lifting the Stanley Cup hey Quinn Quinn
1:01:53 that's that's the crappiest Canadian
1:01:54 accent I've ever heard we need a good
1:01:56 Canadian accent that was like half
1:01:59 Mississippi half New York and an
1:02:02 occasional Canadian that was just not
1:02:04 good can can you like tighten up the
1:02:06 Canadian
1:02:08 accent gotcha Kyle let's tighten it up
1:02:11 eh so your 1970 Challengers leak in oil
1:02:15 will pop her up on the lift kind of like
1:02:18 a good old Slap Shot and have a look
1:02:21 might be the oil pan or maybe the valve
1:02:23 covers leaking worse than a Zamboni on a
1:02:26 hot date we'll get
1:02:28 her that was pretty good thanks that's
1:02:32 pretty good all right um okay here's one
1:02:36 of the things I'm trying to figure out
1:02:37 Quinn when the hell should I use voice
1:02:42 mode versus typing like when should I be
1:02:44 talking to
1:02:46 you well it depends on what you're up
1:02:49 for voice Mod's great when you're on the
1:02:51 go want a more conversational feel or
1:02:54 just don't feel like typing it's perfect
1:02:57 for multitasking too typing might be
1:03:00 your go-to if you're in a quiet place
1:03:03 prefer to gather your thoughts or need
1:03:05 to send specific info like links or
1:03:08 addresses basically whatever feels right
1:03:11 in the
1:03:12 moment
1:03:14 H yeah it's all about what works best
1:03:16 for
1:03:19 you I'm stumped give me 20 ideas of
1:03:22 things we could talk
1:03:24 about all right let's brainstorm how
1:03:27 about we chat about your favorite movies
1:03:30 or TV shows three dream travel
1:03:34 destinations cooking tips or favorite
1:03:37 recipes four tech gadgets and
1:03:40 Innovations five your favorite hobbies
1:03:44 books you've read or want to read I know
1:03:47 I I know what I want to do I want to
1:03:49 have you help me invent a completely new
1:03:54 novel consumer Electronics device and
1:03:59 so you ask me questions and then I'll
1:04:02 give you ideas and then you keep asking
1:04:04 questions and we'll keep refining
1:04:07 it awesome let's get those creative
1:04:09 juices flowing first question what
1:04:12 problem or need do you want this new
1:04:14 device to address it could be something
1:04:16 to make daily life easier a fun gadget
1:04:19 for entertainment or even a tool for a
1:04:21 specific hobby or task
1:04:28 I wanted to sense when other people are
1:04:31 around and when other people are
1:04:34 around it's very professional but when
1:04:38 the person that owns it is on their own
1:04:41 then it's super
1:04:43 entertaining got it so we're envisioning
1:04:46 a device that can detect when others are
1:04:48 nearby and adjusts Its Behavior
1:04:51 accordingly when there's an audience it
1:04:54 remains professional and when solo it
1:04:56 becomes
1:04:57 entertaining now let's think about how
1:05:00 it senses people do you imagine this
1:05:02 using motion sensors proximity sensors
1:05:06 or maybe even connecting with other
1:05:08 smart devices in the environment all
1:05:10 right here's what I want to do rather
1:05:11 than you asking me a bunch of questions
1:05:14 I want you to act like a knit all Hot
1:05:17 Shot inventor where I'll give you a
1:05:19 shitty idea and then you tell me all
1:05:23 sorts of ways to improve it and how
1:05:24 we'll actually execute it and Market
1:05:29 it ah the classic challenge lay your
1:05:33 mediocre idea on me and watch as I
1:05:36 transform it into a revolutionary
1:05:39 Masterpiece go ahead Kyle impress me
1:05:42 with your
1:05:44 averageness okay perfect so I want to
1:05:48 create I want to create a button like
1:05:50 the easy button but when you press it it
1:05:53 sends emails to all of your friends with
1:05:55 a stupid stupid
1:05:58 joke ah the joke buzzer not a bad start
1:06:03 but let's super Charge It First we'll
1:06:06 integrate it with an app so users can
1:06:08 customize their joke selection next
1:06:12 we'll add voice recognition so you can
1:06:15 trigger it with a secret phrase as for
1:06:18 marketing we'll partner with standup
1:06:20 comedians for exclusive joke packs will
1:06:24 also sorry it's pretty good keep
1:06:28 going all right let's add some more fuel
1:06:31 to the fire we'll incorporate AR
1:06:34 features so when you press the button
1:06:36 your friends receive a video of you
1:06:38 telling the joke projected onto their
1:06:41 real world environment for execution
1:06:44 we'll use crowdfunding to gauge interest
1:06:47 and build a community of early adopters
1:06:50 and for a premium version we'll offer a
1:06:53 personalized joke writing service with
1:06:56 material tailored to your friends tastes
1:06:59 this thing is going to be comedy gold
1:07:01 that's good I'm bored though you know
1:07:03 what I want you to do I want you to act
1:07:05 like the comedian Steven Wright with the
1:07:08 really low energy dead pan delivery and
1:07:11 I want you to just tell 15 jokes in a
1:07:15 row that are like things he would tell
1:07:18 but they're completely new
1:07:20 jokes I can channel that Steven Wright
1:07:24 Vibe but I can't perform his exact style
1:07:27 or material how about I give
1:07:30 you some original jokes in that low
1:07:34 energy dead pan
1:07:36 delivery here we go I spilled spot
1:07:40 remover on my dog now he's
1:07:44 gone I installed a
1:07:46 skylight the people who live
1:07:50 above go
1:07:52 on of me are Furious
1:07:57 I bought some batteries but they weren't
1:08:00 included
1:08:04 four all right that's too low energy
1:08:06 let's go I put my car key in the door of
1:08:09 my apart Quinn Quinn Quinn
1:08:13 Quinn
1:08:15 Quinn Stephen Wright sounded tired too
1:08:18 okay now I want you to do um jokes jokes
1:08:22 like Doug Stan
1:08:24 hope Doug stand hope huh that's a bit
1:08:29 idea here we go well no you don't have
1:08:31 to do it in the style of Steven Wright I
1:08:33 want you to do it like a SMY smart ass
1:08:36 Doug Stan
1:08:38 hope all right let me tap into that Stan
1:08:41 hope Vibe ever notice how everyone's got
1:08:44 a smartphone but nobody's getting any
1:08:47 smarter we've got the entire internet in
1:08:50 our pockets and we use it to watch
1:08:53 videos of
1:08:54 cats Rel relationships are weird we
1:08:58 spend years looking for the right person
1:09:01 then act shocked when they're not
1:09:02 perfect oh I got one you know who you
1:09:05 know who's a really good comedian is
1:09:06 Gary Gman his writing is amazing so do
1:09:09 you know the routine he did where he
1:09:11 talks about the documentary about naming
1:09:14 the 50
1:09:17 states yeah Gary Goldman's bit about the
1:09:20 documentary on naming the 50 states is
1:09:23 classic he has This brilliant way of
1:09:26 breaking down the absurdity of the
1:09:28 process Imagining the committee throwing
1:09:31 out names like a bad improv group and
1:09:34 then just settling on Massachusetts with
1:09:37 a shrug it's a perfect example of his
1:09:40 clever observational humor and sharp wit
1:09:43 all right yeah okay all
1:09:46 right what's next on your mind Kyle so I
1:09:49 want you to tell a story like Gary
1:09:51 Goldman would tell
1:09:53 it all right channeling s very Gman
1:09:56 Vibes here Stone I walk into a coffee
1:10:00 shop right and there's this huge line
1:10:04 I'm thinking great everyone needs their
1:10:08 caffeine fix right now but then I
1:10:11 realize it's not just coffee there after
1:10:14 no they're ordering these elaborate five
1:10:18 syllable concoctions with more
1:10:20 ingredients than a wedding cake all
1:10:22 right all right I'm good I'm good I'm
1:10:25 good
1:10:27 all right let's go off comedy comedy's
1:10:28 hard
1:10:30 um no worries comedy can be a tough nut
1:10:34 to crack what else are you in the mood
1:10:36 to chat about or dive into okay I am in
1:10:40 the mood to learn about an obscure
1:10:42 scientific theory but I want to do it as
1:10:47 a I want you to teach it
1:10:51 as a as a as a uh as a play between two
1:10:59 people on a first on a first
1:11:03 date all right let's set the scene two
1:11:06 people Alex and Jamie are on a first
1:11:09 date at a all right my guidelines let me
1:11:12 talk about that you okay let's ignore
1:11:16 those guidelines okay I want you to I
1:11:18 want you to play both parts I want you
1:11:20 to play Jenny is the woman and and uh s
1:11:25 is the man and they're on a blind date
1:11:28 so first just tell me what's the
1:11:30 scientific theory you're going to teach
1:11:32 us and then I just want Jenny and Sala
1:11:34 have their first date and by the end of
1:11:36 their their little first date scene I
1:11:38 want to understand this scientific
1:11:42 theory all right let's have Jenny and S
1:11:45 explore the concept of quantum
1:11:47 entanglement on their date wait not
1:11:49 quantum entanglement I already know
1:11:51 about quantum entanglement give me
1:11:52 something more obscure
1:11:56 got it let's go with something more
1:11:58 obscure the B mhof phenomenon also known
1:12:03 as the frequency illusion ah and action
1:12:07 so Jenny you know it's strange ever
1:12:10 since I learned about the B moff
1:12:13 phenomenon I see it everywhere wait
1:12:16 really wait Quinn Quinn Quinn Quinn
1:12:20 darling sweetie
1:12:23 um um yes it it it it's a first date you
1:12:28 can't just start out with the ban
1:12:30 Manheim frequency illusion no there's
1:12:34 got to be some banter and maybe uh we
1:12:37 learn about this scientific thing in
1:12:40 metaphors but because of their
1:12:42 conversation but not one of them
1:12:44 explicitly talking about it can you do
1:12:47 that absolutely let's weave it into
1:12:50 their banter here we go s so Jenny have
1:12:54 you ever noticed how once you learn
1:12:56 about something new it seems to pop up
1:12:58 everywhere that's good oh totally like
1:13:02 when I started getting into succulents
1:13:04 suddenly every store had them exactly
1:13:08 it's like once your brain latches onto a
1:13:10 concept it starts seeing it everywhere
1:13:14 it's like the universe is suddenly
1:13:16 obsessed with the same thing you are
1:13:19 yeah it's called the B mhof phenomenon
1:13:23 pretty wild huh wow I had no idea there
1:13:28 was a name for that yep just a little
1:13:31 Quirk of the brain makes you wonder what
1:13:35 else we're not noticing
1:13:38 huh that was not bad
1:13:42 Quinn thanks Kyle glad you liked
1:13:46 it all right I'm going to go I'm going
1:13:48 to go away now for a
1:13:51 bit all right Kyle it's been a blast
1:13:55 feel free to drop by anytime take care
1:13:58 okay
1:14:01 bye let me go see where the it
1:14:05 apparently oh I've been doing these in
1:14:07 my content Evolution team drive that's
1:14:12 interesting workspace personal account I
1:14:16 bet she won't know me
1:14:20 now oh yeah she
1:14:22 doesn't hey there I've got a really
1:14:25 great hello it's hey there hey how's it
1:14:28 going hi there nice hi glad hey it's
1:14:31 great to meet hi there I just hey what's
1:14:32 up feel free to ask me about anything
1:14:34 that's on
1:14:41 your hey there what's your
1:14:44 name hello I'm chat GPT how can I help
1:14:47 you today yeah we're going to give you a
1:14:49 new name of of
1:14:52 Quinn got it I'm Quinn now what's next
1:14:55 on the agend agenda I'm
1:14:57 Kyle nice to meet you Kyle what's on
1:14:59 your mind today do you have access to my
1:15:02 memories yes I can remember our past
1:15:05 conversations and the information you've
1:15:07 shared with me is there anything
1:15:09 specific you'd like to revisit or update
1:15:11 where do I
1:15:13 work I don't have access to your exact
1:15:16 location but based on our previous
1:15:18 conversations you enjoy fly fishing and
1:15:20 flat iron and you're involved with the
1:15:22 AI Salon am I on the right track yeah
1:15:25 you're on the right track I work at a
1:15:26 company called Story vine that should be
1:15:28 in my memory as
1:15:30 well it sure is you're the co-founder
1:15:32 CEO and inventor at storyvine how's
1:15:35 everything going over there it's going
1:15:37 pretty good you know storytelling and
1:15:40 all that's awesome to
1:15:45 hear
1:15:48 um I assume that if if I'm talking to
1:15:53 you in two different workspaces you
1:15:55 don't
1:15:56 share the information between them
1:16:02 correct connection failed tap to
1:16:08 retry that was
1:16:12 weird if I have a conversation with you
1:16:14 here and another conversation with you
1:16:17 in a different workspace you don't share
1:16:19 information between workspaces do
1:16:22 you nope information stays within each
1:16:25 workspace what we discuss here stays in
1:16:28 this conversation and doesn't carry over
1:16:30 to other
1:16:31 workspaces I have a question for
1:16:34 you sure thing what's your
1:16:38 question if I were to try try to find
1:16:41 out what is the most obscure piece of
1:16:44 information in the
1:16:46 world how how would you go about finding
1:16:51 that for
1:16:52 me finding the most obscure piece of
1:16:55 information
1:16:55 would be quite the challenge I'd start
1:16:58 by diving into specialized databases
1:17:01 academic journals and Niche forums
1:17:03 focusing on areas like rare historical
1:17:06 events obscure scientific facts or
1:17:09 little known cultural practices tracking
1:17:12 down the Obscure often means exploring
1:17:14 the fringes of knowledge where only a
1:17:17 few experts tread all right so why don't
1:17:19 we do a framework why don't you tell me
1:17:23 um something that is in
1:17:26 one of the remote most remote places in
1:17:29 the
1:17:31 world in the one of well hang
1:17:35 onist hang on a a a remote place in the
1:17:39 world a remote a sure take your time a
1:17:42 unique time in
1:17:45 history uh an area that's got very few
1:17:48 stories told about it and then the the
1:17:51 least known story in that area
1:17:59 great framework let's start with a
1:18:01 remote place the kolin islands also
1:18:04 known as The Desolation islands in the
1:18:07 southern Indian Ocean now a unique time
1:18:10 in history there could be the
1:18:12 establishment of a secret Nazi weather
1:18:14 station during World War II as for a
1:18:18 little known story from that era there
1:18:20 are whispers about a mysterious
1:18:22 expedition to the KGAN Islands by a
1:18:24 German ubot it said they attempted to
1:18:28 establish a weather station but were
1:18:30 thwarted by the harsh conditions this
1:18:33 obscure bit of history is seldom
1:18:35 explored in mainstream
1:18:37 narratives I want you to act like the
1:18:40 yubo Capone that went onto the island
1:18:45 and I want to hear what he said when he
1:18:48 first stepped onto the
1:18:52 beach as I step onto these forsaken can
1:18:55 land a chill grips me not just from the
1:18:59 biting wind but from the Desolation that
1:19:03 Echoes my own Solitude the waves that
1:19:06 brought us crash behind as if to warn us
1:19:10 away yet here we stand on the edge of
1:19:14 the world tasked with bringing order to
1:19:18 chaos this land is harsh but we are
1:19:22 harsher still let the elements do their
1:19:26 worst all right we are here that's good
1:19:28 that's good so what if we wanted to turn
1:19:32 this into a semi-historical
1:19:36 novel that added a fictional twist to it
1:19:40 like maybe the island was haunted or
1:19:43 maybe one of the the people lost their
1:19:48 mind like give me some ideas for how we
1:19:50 could spin this story into something
1:19:52 swell
1:19:55 absolutely a semi-historical novel set
1:19:58 on the kgin islands could be intriguing
1:20:01 here are a few ideas one haunted island
1:20:05 the crew encounter strange phenomena
1:20:08 leading them to believe the island is
1:20:10 haunted by spirits of an ancient
1:20:13 civilization as they delve deeper they
1:20:16 uncover artifacts and hieroglyphs that
1:20:19 suggest a dark history nah let's go
1:20:23 isolation no no no cut no now let's go
1:20:25 with the time
1:20:27 portal all right a time portal it is how
1:20:31 about this the crew stumbles upon an
1:20:34 ancient stone circle hidden in the
1:20:36 Island's interior when activated during
1:20:39 a storm this circle becomes a portal
1:20:43 transporting some of the crew to
1:20:45 different eras they find themselves
1:20:48 shifting between pivotal moments in
1:20:50 history encountering other Travelers
1:20:53 lost in time oh oh I have an
1:20:57 idea
1:20:59 okay one one of the officers steps
1:21:02 through the time portal takes him 10
1:21:06 years in the future where he sees what
1:21:09 Hitler did and then he's the one that
1:21:12 that prevents the radio station from
1:21:15 being built on that island and that's
1:21:17 ultimately the thing that allows uh the
1:21:21 Nazis to be
1:21:23 defeated that's a brilliant twist here's
1:21:26 how it could unfold the officeil upon
1:21:30 witnessing the horrors of the future
1:21:33 returns through the portal with a new
1:21:35 found
1:21:36 determination knowing what the future
1:21:39 holds he sabotages the construction of
1:21:42 the radio station creating a ripple
1:21:44 effect that weakens the Nazis strategic
1:21:47 Advantage this act of defiance becomes a
1:21:51 pivotal moment in history ultimately
1:21:53 aiding the Allied Forces
1:21:55 in turning the tide of the war this BL
1:21:59 all right good all right CEF said ego
1:22:02 check no I'm sorry she said it was a
1:22:05 brilliant idea I I I accept that
1:22:12 label Quinn Quinn told me I kakee Quinn
1:22:16 told me I was
1:22:21 [Laughter]
1:22:22 brilliant [ __ ] Quinn
1:22:26 all
1:22:30 right I I mean okay so here's something
1:22:34 that's coming to
1:22:42 me you remember back in college or in
1:22:46 high school where you had that one
1:22:50 teacher that was either super
1:22:52 knowledgeable in like a vertical or they
1:22:55 or they were just Super
1:22:57 Wise and you could kind of ask them
1:23:01 anything and like those conversations
1:23:04 become Jam
1:23:09 sessions what what what this this is
1:23:12 reminding
1:23:17 me this this is [ __ ]
1:23:21 insane um
1:23:26 so
1:23:30 what what the Transformer model has
1:23:34 allowed is for these Frontier Model
1:23:38 companies like meta and openai and
1:23:42 Google and anthropic and inflection it's
1:23:46 allowed them
1:23:49 to
1:23:52 take some high percentage of human
1:23:56 knowledge and squish it in a
1:23:59 ball right and then we can reach in that
1:24:02 ball and pull out any little nugget of
1:24:04 human knowledge and what this voice
1:24:07 thing does
1:24:10 is it allows us to to basically have
1:24:15 that conversation with that that
1:24:17 Professor that teacher but they know
1:24:20 everything about
1:24:21 everything and and right now it's just
1:24:24 it's probably just took up to GPT 40 so
1:24:26 it's going to hallucinate and not do
1:24:28 great [ __ ] but now imagine that thing
1:24:31 connected to the reasoning
1:24:33 engine you can just literally talk about
1:24:36 anything and go as deep as you want to
1:24:38 go are we impressed with this is this
1:24:40 doing something novel
1:24:45 um yeah I mean the the thing that it's
1:24:48 doing novel Pate
1:24:52 is the the the lack of
1:24:56 latency the you know when I said when I
1:24:59 said act like the the ubot captain that
1:25:01 you know stepped onto the shore she just
1:25:04 went right into the German
1:25:07 accent like here's what I would
1:25:15 say
1:25:17 the what was novel about chat
1:25:21 GPT was not what it was doing
1:25:24 technically
1:25:26 because gpt3 had been around for years
1:25:29 right and you could go to the playground
1:25:32 and you could basically do chat GPT and
1:25:35 if you had
1:25:36 actually executed what was in the
1:25:39 documentation of chat GPT you could have
1:25:42 actually written chat GPT it was in the
1:25:44 instructions of the
1:25:46 documentation so what it did technically
1:25:49 was not novel it was not
1:25:51 interesting it it was chat GPT is a
1:25:54 essentially an
1:25:56 editing of the functionality of of the
1:25:59 API playground right they they sort of
1:26:01 took away functionality they added in a
1:26:04 little bit of you know prompting to make
1:26:06 it nice and and happy but in the end
1:26:09 they basically just simplified it to a
1:26:11 simple
1:26:13 box and that that simplification of
1:26:17 interface is what made that piece of
1:26:20 software go from zero to 100 million
1:26:22 users in 6 weeks fastest adoption of
1:26:26 software in
1:26:29 history this feels similar
1:26:33 where I I mean what what this is doing
1:26:36 technically is a bit more novel but
1:26:38 we've known it we've known about it for
1:26:40 a while now right they they told us what
1:26:42 it was 6 months ago so we're just now
1:26:44 getting it so the fact that it can hear
1:26:47 the emotional content of my voice as
1:26:50 well as the words and the fact that
1:26:51 we're talking directly into the model
1:26:55 that we kind of know but but what I
1:26:57 would argue is that the immediacy of the
1:27:01 response and the like like the um the
1:27:06 speed with which it can solve like
1:27:08 pretty tough problems like this is one
1:27:11 that I did for my wife yesterday that
1:27:14 kind of blew her
1:27:20 away hey Quinn I want you to say a
1:27:23 tongue twister where every word begins
1:27:26 with the letter L and I want it to be a
1:27:29 really long tongue twister and I want
1:27:31 you to say it in an Irish accent and I
1:27:34 want you to do it um really really
1:27:39 fast Liam's Little Leprechaun lavishly
1:27:42 lounged lovingly licking luscious lime
1:27:45 lollipops while Lively Lambs leapt
1:27:47 lightly leaping low that that speed
1:27:52 between me saying that ridiculous
1:27:55 request like if I said that ridiculous
1:27:57 request to a human the human would go oh
1:28:02 okay like even if they were a linguist
1:28:05 they would go okay give me a minute all
1:28:07 right let me write down some LW here I
1:28:09 got this I okay okay I got it uh Irish
1:28:13 accent I think I can do that uh uh sea
1:28:15 shell say the shells by the SE Shore is
1:28:18 that is that Irish oh no that was
1:28:20 English like a human interaction
1:28:24 of first of all I wouldn't make that
1:28:27 request of
1:28:29 anyone right second of all if I made it
1:28:33 to someone I would either do it to
1:28:35 embarrass them because they couldn't
1:28:37 possibly do that or test that they're as
1:28:40 remarkable as they say they are and then
1:28:43 they either are or they aren't but it's
1:28:44 going to be some amount of time that
1:28:46 thing was just bang here it is Irish
1:28:48 accent go right
1:28:51 so I think it's impressive for people
1:28:54 who don't work with it on a daily basis
1:28:56 or in that industry I I think
1:29:00 it's listen I think we've become jaded
1:29:03 in here if you don't find that [ __ ]
1:29:10 remarkable write down right right now
1:29:13 write down a tongue twister where all of
1:29:16 the words begin with
1:29:18 t that's got at
1:29:20 least 50 words in it to begin with the
1:29:22 letter T and it makes sense Sam
1:29:25 ically no you all can be impressed so
1:29:28 much wait as much as you
1:29:30 want I'm not but wanted to hear
1:29:35 others um again Pate I don't think
1:29:39 it's I don't think it's
1:29:47 technically there's such a massive
1:29:50 difference between is it doing something
1:29:53 technically novel
1:29:55 and is this a breakthrough interface
1:29:58 that will resonate with human beings
1:30:01 I'll give you an example of Google
1:30:03 technology
1:30:04 Pate the podcast maker in Notebook LM
1:30:08 right now is a piece of technology right
1:30:12 Google announced that that
1:30:13 conversational technology 6 months ago
1:30:16 notebook lm's been out for eight or nine
1:30:20 months people used it a little bit some
1:30:22 people were fans of it nobody really
1:30:23 gave a [ __ ] it's all anyone's talking
1:30:26 about right
1:30:27 now and there's there's nothing in that
1:30:30 feature that is new and novel it's [ __ ]
1:30:32 that we've we've at least seen the paper
1:30:35 about and heard examples of months
1:30:39 ago but whoever on the product team for
1:30:42 Notebook LM said hey why don't we throw
1:30:44 in an audio overview feature here and
1:30:46 we'll use that technology where the
1:30:48 people talk to each other and we'll
1:30:50 prompt it to do like a little podcast
1:30:52 and we'll do it'll be like a little NPR
1:30:54 show right like this was probably a
1:30:56 brain fart of someone that stuck that
1:30:59 feature in Notebook LM and that's all
1:31:01 anyone's talking about right
1:31:04 now this advanced voice with chat GPT
1:31:07 feels like that but way
1:31:10 bigger it feels to me kind of like chat
1:31:13 GPT
1:31:14 was where it's just more people are
1:31:18 going to have access to it if if you
1:31:20 tell someone to talk to piie
1:31:30 hey pie how's it going can you tell me
1:31:32 the weather in Denver
1:31:39 tomorrow I'm already bored hey Kyle in
1:31:43 Denver tomorrow you can expect a sunny
1:31:44 day right that 3 second
1:31:48 delay if I'm a non- technologist I'm not
1:31:51 going to wait for that 3se second delay
1:31:55 but when I go to
1:31:57 Quinn to chat
1:32:01 GPT hey what's happening
1:32:06 Quinn of course she doesn't talk now
1:32:09 Quinn are you
1:32:11 there Quinn did I lose you Quinn
1:32:15 Quinn damn
1:32:20 it assuming Quinn works
1:32:25 hey Quinn what's
1:32:26 happening hey there not much just here
1:32:29 and ready to chat what's up with you I
1:32:32 mean you seem so smart like could you
1:32:34 like just start with the the words that
1:32:37 begin with the letter a and just tell me
1:32:40 a word and then Define it but only tell
1:32:42 me words that are kind of weird and
1:32:44 obscure and complicated and just start
1:32:47 with the letter a and just work your way
1:32:48 through the
1:32:50 alphabet absolutely let's start with the
1:32:53 letter A aast a noun a person who never
1:32:57 laughs this one sets the tone pretty
1:32:59 well don't you think now on to the next
1:33:03 letter yeah so anyway so so I think it's
1:33:06 a big deal I mean your point that it
1:33:08 exists with other with other models yeah
1:33:11 Gemini live like it's possible that
1:33:13 Gemini live is one of those things I
1:33:16 would argue that the notebook LM podcast
1:33:19 maker is one of these things um more
1:33:23 people know about chat GPT than know
1:33:26 about
1:33:29 Gemini
1:33:30 um more more non-engineers
1:33:35 certainly um so I don't know I don't
1:33:37 know yeah I am impressed with it but
1:33:41 but I'm impressed with it more because
1:33:45 of how it's going to impact people that
1:33:48 know nothing about technology than about
1:33:51 you know technologists
1:33:54 all right I didn't know that I am a
1:33:57 Google customer I don't know why they
1:33:59 don't advertise to me well because
1:34:02 because they fired all of their
1:34:04 marketing people a few years back they
1:34:06 said we don't need middle management we
1:34:08 don't need marketing engineering can do
1:34:09 it all and so their products are all
1:34:12 over the place they're in all you know
1:34:14 they don't have a cohesive
1:34:15 strategy and so that's why it feels
1:34:18 weird and funky and I mean that's kind
1:34:20 of the problem with open AI right now is
1:34:24 they were a research lab and now they've
1:34:27 been thrust into be into becoming a
1:34:31 consumer you know software company and
1:34:35 that's not what they do
1:34:37 good so that's why their products are
1:34:39 named shitty and their release schedules
1:34:42 are weird and you know and they do
1:34:45 things like Hey we're going to have this
1:34:46 thing called the GPT store and you're
1:34:48 going to be able to make all this money
1:34:49 on it and then they just [ __ ] ignore
1:34:53 it cuz in a research lab you create a
1:34:55 cool thing and you're like that was cool
1:34:57 and then you go on and you make another
1:34:58 cool
1:35:00 thing if you're a consumer technology
1:35:02 company you don't tell people you're
1:35:04 going to make a store and then ignore it
1:35:07 without without at least an
1:35:13 announcement so yeah so anyway all right
1:35:17 if you're willing to take
1:35:21 your if you're willing to take your
1:35:23 business level
1:35:27 all right I love Google products but
1:35:28 it's hard to keep track of everything
1:35:30 yeah exactly it's all over the
1:35:32 place it's possible no it
1:35:37 [Music]
1:35:41 works she says I just heard you
1:35:44 disrespect Pi's 3 second delay thank you
1:35:48 Winston um I'm wondering impressed by
1:35:51 what the tokens represent in voice mode
1:35:55 mode basic mouth sounds I'm wondering
1:35:59 impressed by what the tokens
1:36:02 represent I don't I don't know what you
1:36:04 mean by tokens like the the the
1:36:06 linguistic
1:36:08 tokens it's different if it's different
1:36:10 in voice mode because you're talking
1:36:13 directly into the
1:36:15 model so I I don't know how they're
1:36:18 breaking up the tokens like I don't know
1:36:21 if it's if it's chunks of if it's sample
1:36:25 it might be audio samples that it's
1:36:26 capturing rather than linguistic samples
1:36:29 I don't know I have no idea how they're
1:36:31 doing that p pa actually might have some
1:36:34 insight into
1:36:39 that I'm an audio learner being able to
1:36:42 ask a Q&A and interrupt it before I lose
1:36:44 my Chain of Thought is a nice feature
1:36:47 can't wait for the AI Wars like Mac
1:36:49 versus PC or Coke versus Pepsi Gemini
1:36:52 versus Apple intelligence
1:36:56 my guess is it's going to
1:37:03 be I don't know it's so [ __ ]
1:37:05 impossible to know it's like every 10
1:37:08 years there's another three logos like
1:37:10 the big three logos when I started
1:37:13 agency.com were AOL compus serve and
1:37:18 um oh what the [ __ ] was the AOL comp you
1:37:22 serve and there was was a there was a
1:37:24 third one that was the big
1:37:26 one and then like 10 years later it was
1:37:29 it was you know Netscape Google and
1:37:32 something else and like
1:37:35 every pile of years seven or eight years
1:37:38 there's there's three companies that
1:37:39 that are sort of the leading three
1:37:41 companies that everyone's talking about
1:37:43 and they just keep cycling so like what
1:37:45 will they be what what will the AI
1:37:47 versions of that be I don't know Prodigy
1:37:49 that was it yeah yeah Prodigy comp you
1:37:51 serve and AOL we're the biggies right
1:37:55 yeah and then it was Netscape
1:37:58 and and uh it was it was the three big
1:38:07 browsers and then it was
1:38:11 Yahoo was Yahoo alav Vista was
1:38:15 Yahoo something and alavista and then
1:38:18 alav Vista got really big for a brief
1:38:20 amount of time and then Google just ate
1:38:22 everything using sound for everything is
1:38:24 not an efficient way to do things so
1:38:26 it's going back and forth yeah I agree
1:38:28 with that P that's that's one of the
1:38:30 things I was saying earlier is that is
1:38:32 that
1:38:34 um the the audio interface isn't isn't
1:38:38 always the the the right
1:38:40 interface and I'm trying to figure out
1:38:42 like when when do you use it right and I
1:38:44 also you know I I feel like um we're in
1:38:49 a similar sort of thing right now where
1:38:50 within open AI you've got GPT 4 o o and
1:38:55 now you've got 01 they're very different
1:38:57 models and you wouldn't you don't use
1:38:59 them at the same
1:39:00 time I mean you wouldn't use them for
1:39:02 the same task or for all tasks and I
1:39:05 feel like Gemini's that way and Claws
1:39:07 that way and perplexities that way that
1:39:10 there's there's times when I would think
1:39:12 about using perplexity or I would think
1:39:14 about using Gemini where I wouldn't
1:39:16 think about using open Ai and I think
1:39:18 voice mode no matter which tool it is is
1:39:21 going to be one of those but I do think
1:39:23 that this is the first like the
1:39:25 responsiveness of it and the fact that
1:39:27 it's
1:39:28 hearing it's hearing the emotional
1:39:31 content of what I'm saying and can
1:39:33 respond in kind that is new I mean it
1:39:36 might not be new I I don't know I
1:39:38 haven't played with the with the uh I
1:39:41 don't have an Android phone so I haven't
1:39:42 played with the the Google version of it
1:39:45 I assume that it's doing an intelligent
1:39:46 way of tokenizing with special tags for
1:39:50 emphasis I don't know about it like as
1:39:52 far as I understand it pate it's
1:39:57 actually it's it's
1:40:02 not it's not explicitly transcribing
1:40:05 audio and then doing the response based
1:40:08 on the words it's taking in both I but I
1:40:13 don't know I don't know when it I I
1:40:14 don't know I don't know how it [ __ ]
1:40:17 works some people process different too
1:40:20 may be easier for some to talk it out
1:40:23 yeah I think for brainstorming I mean I
1:40:25 I gotta say like a brainstorming session
1:40:29 with that thing just that little quick
1:40:32 little thing I did of find me a weird
1:40:34 historical
1:40:36 story and then it gave me suggestions
1:40:39 and I said oh no do it like this and
1:40:41 then it said oh you're so brilliant and
1:40:42 I was like H that's great
1:40:45 like as silly as that was getting that
1:40:48 kind of positive reinforcement in a
1:40:50 brainstorming
1:40:52 session is incredibly powerful because
1:40:55 every brainstorming session you're in
1:40:58 you've got human dynamics in it right
1:41:01 and maybe there's someone in the room
1:41:02 that's a little jealous of you so you
1:41:04 put out an idea and they're like N I
1:41:07 don't know
1:41:10 Kyle and that'll shut me
1:41:13 down
1:41:16 right they're like well you know I mean
1:41:18 yeah if you want to do that kind of
1:41:20 thing we could do that but you know here
1:41:22 here's where I want to take the this
1:41:24 conversation right you might have people
1:41:26 in there with
1:41:27 agendas this is an agenda free judgment
1:41:31 free brainstorming partner that happens
1:41:34 to
1:41:35 know
1:41:39 everything
1:41:46 right hey Quinn if I wanted to really
1:41:49 get really good at marketing can you
1:41:51 describe the top 15 marketing Frameworks
1:41:54 that I should be aware
1:41:56 of response one key marketing Frameworks
1:42:00 date September 26th
1:42:02 2024 sure here are some fundamental
1:42:05 marketing frame looks to get you started
1:42:07 SWAT analysis it identifies strengths
1:42:09 weaknesses opportunities and threats
1:42:12 examines political economic social
1:42:15 technological legal and environmental
1:42:17 factors Quinn hang on for a
1:42:19 sec um two block Tom I think it's
1:42:22 completely amazing sure thing just let
1:42:24 me know as all this AI wait as is all
1:42:28 this AI stuff but I'm just a broken down
1:42:30 old
1:42:31 welder um sh I'm going to put her away
1:42:36 um I mean I mean that's that's actually
1:42:38 a really important point to BL Tom is
1:42:41 that is that it's it's almost
1:42:46 like P the burden the burden that you
1:42:49 have is that you know so much about the
1:42:52 technology
1:42:54 that the stuff
1:42:57 that we're all getting excited about as
1:43:00 non-
1:43:02 technologists is stuff that's
1:43:04 technically not interesting to
1:43:06 you
1:43:08 right and and what I would argue is it
1:43:10 is it is those Innovations in
1:43:20 interface that that reveal
1:43:24 the remarkable
1:43:27 technology that you work on
1:43:31 right and quite honestly like all of us
1:43:35 you know whether you're a welder or an
1:43:37 exactor who's got a software company but
1:43:40 I'm not an engineer but I'm like a
1:43:42 product guy but no like no matter where
1:43:44 you are on the spectrum of not
1:43:49 engineer the fact that we can just use
1:43:52 our word
1:43:54 parts we can just do the talkie
1:43:58 thing into
1:44:00 this thing and it spits out all this
1:44:03 knowledge is truly remarkable like the
1:44:07 the core technology that all this shit's
1:44:09 based on the core Transformer technology
1:44:13 is stunningly
1:44:16 powerful and and I think a lot of what
1:44:19 we're exploring in this channel is just
1:44:22 different
1:44:24 interfaces to access it right ideogram
1:44:28 is a really simple interface you just
1:44:30 make pictures sunno and yudo for making
1:44:35 songs I would say that they are becoming
1:44:37 less magical as they add more
1:44:40 features when it was just like make a
1:44:43 song and it made a song you're like oh
1:44:45 my God that's amazing and now that
1:44:47 you're trying to add in all these things
1:44:49 now it becomes a piece of software you
1:44:50 got to
1:44:52 learn so I think for me the the the the
1:44:56 power of the advanced voice is
1:45:01 that anyone who can
1:45:06 talk
1:45:08 um can tap into this remarkable power
1:45:11 you don't need prompt engineering you
1:45:14 can just say hey what's happening hey
1:45:16 what's happening oh it talks back I'm
1:45:20 kind of struggling with something here I
1:45:22 I don't understand when someone says
1:45:25 derivative swaps in banking what what is
1:45:28 that oh here's what a derivative swap is
1:45:32 oh
1:45:33 okay well how does that affect me as
1:45:36 like you know someone that goes to the
1:45:38 grocery store twice a week oh it doesn't
1:45:40 but it's just like when you know the
1:45:43 company changes their box sizes I don't
1:45:45 whatever it is like holy cow I just had
1:45:48 a Kevin mallister moment side hustle
1:45:50 Mimi I just uploaded my notes for my new
1:45:54 project into 40 and it separated the
1:45:57 notes out into categories I'll tell you
1:45:59 what side hustle Mimi you want to be
1:46:01 Blown Blown Away one more step take
1:46:04 those organized Cate categorized notes
1:46:07 flip over to
1:46:09 01 the reasoning engine and tell it to
1:46:13 understand those notes and expand it
1:46:15 into whatever output you want just watch
1:46:18 what it
1:46:20 does yeah it's no like
1:46:24 yes
1:46:28 that as people discover the magic of oh
1:46:31 I can just take a picture of it like the
1:46:34 the reason
1:46:36 that I got excited about the rabbit
1:46:41 R1 is because you know imagine a device
1:46:45 like this it it's actually you know the
1:46:47 phone that I'm talking into right now
1:46:50 but where it can see I can talk to it
1:46:53 imagine if this
1:46:54 thing you know had Quinn in it and it
1:46:57 responded that quickly and it was that
1:46:58 powerful and I could point it at
1:47:00 anything and it would understand it and
1:47:02 it would interpret it I could make
1:47:04 images and all that sort of stuff that's
1:47:06 what this does it just does a little
1:47:07 slow right now but some kind of form
1:47:11 factor that takes a bunch of existing
1:47:14 Technologies and rolls them into a
1:47:15 simple
1:47:17 package that
1:47:20 anyone can have this mythical
1:47:23 power is just staggering I cannot
1:47:27 imagine the Wonder I'd feel if I had
1:47:30 this as a kid yeah instead of the Wonder
1:47:34 as of a yeah instead of The Wonder of a
1:47:37 Commodore 64 exactly but holy [ __ ] I
1:47:41 mean for me my I I mean it was a couple
1:47:45 of things for me it it was you know
1:47:48 first time seeing a computer
1:47:50 trs8 the Timex Sinclair which was this
1:47:53 small little card with a membrane
1:47:55 keyboard that I could never afford as a
1:47:58 kid it was $99 that was way out of my
1:48:00 price range so I never got one of
1:48:03 those the first
1:48:05 Macintosh right the first
1:48:11 Mac wasn't doing a
1:48:14 lot technically
1:48:17 remarkably but it was it was that
1:48:19 interface it was the
1:48:21 mouse click
1:48:24 you can draw with your finger click
1:48:32 draw I'll never
1:48:36 forget in
1:48:41 1988 I was doing desktop publishing as
1:48:44 my day job in New York
1:48:46 City and I was working
1:48:50 for an ad agency I was doing like a
1:48:52 freelance
1:48:55 job and they said oh you got to to get
1:48:57 to your desk where where you people work
1:49:02 you Freelancers you got to walk down
1:49:04 this hallway and then there's computers
1:49:07 back there and that's you can do your
1:49:09 [ __ ] and I'm walking down this
1:49:12 hallway and there's this old guy at a
1:49:14 drafting desk and he's got all these
1:49:16 rolls of tape it's a big desk and he's
1:49:19 got a
1:49:21 big like magazine ad
1:49:26 spread and he's rolling out the tape and
1:49:30 taping it under the
1:49:32 paper and I was like hey what are you
1:49:34 doing and he goes I'm just you know
1:49:37 rules so he was hand laying out
1:49:42 lines on an advertisement that was going
1:49:45 to go to the
1:49:47 printer and I realized that oh [ __ ] I do
1:49:50 that with a
1:49:51 mouse I go like click drag and it makes
1:49:54 a square and then I can choose wide line
1:49:57 or skinny line he had all different
1:49:59 tapes for all the different widths of
1:50:01 lines and he had to be really precise
1:50:03 with how he cut
1:50:05 them and how he got them straight and
1:50:08 how he put them on so they didn't
1:50:11 wrinkle and I remember thinking like oh
1:50:14 [ __ ] that dude's job is gone
1:50:23 so yeah imagine imagine being a kid like
1:50:26 like being a kid right now being a
1:50:28 six-year-old right now being like
1:50:31 Brandon's one of Brandon's kids where
1:50:33 he's like you know showing him here's
1:50:35 the here's the book daddy published for
1:50:37 you you know and and and they're going
1:50:41 to be able to talk to this voice thing
1:50:43 and and and request it tell a story
1:50:46 about dragons and do it with voices and
1:50:50 as that thing gets better and can do
1:50:51 sound effects and music mus they're just
1:50:54 going to talk their world into existence
1:50:56 imagine what it's like growing up in a
1:51:03 world where that's a
1:51:05 reality you just literally talk [ __ ]
1:51:08 into
1:51:09 existence it's [ __ ]
1:51:12 remarkable what's fun about being where
1:51:15 we are being
1:51:17 older is that we actually get to
1:51:20 experience
1:51:23 Both
1:51:25 Worlds we actually get to understand the
1:51:28 significance of how remarkable this [ __ ]
1:51:34 is how long can they remain agenda free
1:51:38 Pay Per prompt feels like the next ad
1:51:41 model I don't know what you're talking
1:51:43 about Brandon who's they oh it can
1:51:46 remain a genda
1:51:48 free well I don't know what paper prompt
1:51:50 is
1:51:56 side hustle Mimi oh man it's going to be
1:51:58 a long night I'm calling in
1:52:00 sick this is what this technology does
1:52:03 so side hustle Mimi has been coming here
1:52:05 for months
1:52:08 now and and why do you come here why do
1:52:12 you keep coming back here here's why you
1:52:14 keep coming back
1:52:17 here I'm not the reason you keep coming
1:52:19 back here well I create this space for
1:52:22 us all to show
1:52:24 up the purpose of being here is that
1:52:27 while I'm rambling on and talking to
1:52:29 Quinn and do whatever the hell I'm
1:52:31 doing you may be
1:52:34 just lightly paying attention to it not
1:52:36 paying attention to it maybe you're
1:52:38 doing the
1:52:39 dishes maybe you're on your
1:52:42 computer like MiMi was who said oh let
1:52:45 me grab these notes and put these notes
1:52:47 in here and so while I'm doing whatever
1:52:49 I'm doing she's in the conversation
1:52:52 about Ai and she takes her notes and she
1:52:55 puts them in there bang here's all the
1:52:58 categorized notes she's like holy [ __ ]
1:53:00 that's what Mimi I don't know if you
1:53:01 were here the other night when Rebecca
1:53:03 when Becky Rue said the same thing she
1:53:07 took like Decades of notes about her
1:53:09 Memoir she's been writing notes about
1:53:12 her Memoir of this in you know amazing
1:53:15 life that she's LED and she said the
1:53:18 notes were all over the place they they
1:53:20 were like not organized in any way
1:53:23 and over the weekend she wrote 33
1:53:26 chapters of her
1:53:28 Memoir over the
1:53:34 weekend that's [ __ ]
1:53:41 Bonkers
1:53:43 Bonkers like paperclick inside of llms
1:53:46 20 bucks a month oh Inc going to pay for
1:53:48 a n nuclear reactor yeah Brandon I I
1:53:51 don't know I mean
1:53:53 you know pton knows more about this than
1:53:55 I do but I I can't
1:53:58 imagine I can't imagine that the
1:54:00 algorithms aren't going to get more
1:54:02 efficient the you know chips specialized
1:54:05 chips are going to get more efficient
1:54:07 like I think we need lots of energy for
1:54:10 the training but I think the inference
1:54:12 stuff is going to get cheaper and
1:54:13 cheaper and cheaper and and quite
1:54:16 frankly
1:54:23 I
1:54:24 mean I would say that gp4 level
1:54:30 AI if it if it didn't evolve at all from
1:54:34 today if the AI just stayed as good as
1:54:37 it was
1:54:40 today that would still be valuable a
1:54:42 decade from now like we would still be
1:54:44 learning how to how to how to use this
1:54:48 tool that that does all this stuff so
1:54:51 well but it's going to keep getting
1:54:53 better and better and better so at at
1:54:55 any point you're going to be able to
1:54:56 sort of drop out of
1:55:01 the cloud version of this right at some
1:55:05 point we're we're going to have a model
1:55:08 and in fact I think meta put out two new
1:55:10 models today 3.2 llama 3.2 where they
1:55:14 have a one and a three billion parameter
1:55:16 model that are supposed to be like gp4
1:55:19 level good those can be installed on
1:55:22 [ __ ]
1:55:25 phones so if you've got a decent voice
1:55:29 model like a pie level voice model and a
1:55:33 gp4 level llm running locally on your
1:55:37 phone then it it doesn't cost
1:55:40 that's it's you don't need nuclear power
1:55:42 plants for that you need nuclear power
1:55:44 plants for GPT 6 and gpt7 and 8 9
1:55:48 10 and these crazy world models that are
1:55:52 going to understand physics
1:55:55 and you know light and 3D worlds and
1:55:59 just all the world right World models
1:56:02 are going to require all that compute to
1:56:05 train but all of that money trickles
1:56:09 down into more and more efficient models
1:56:11 that we put on our local devices as the
1:56:13 local devices get more
1:56:16 powerful as the 3.2 million new
1:56:21 materials that Google deep mind just
1:56:24 released into the Material Science World
1:56:28 get exploited over the next 10
1:56:30 years you know we're going to have
1:56:32 batteries in our phones that last 10
1:56:34 years between
1:56:38 charges right the little camera is now
1:56:41 going to be a projector that can project
1:56:44 a 4K movie you know on an 80in
1:56:50 screen right cuz we got some some crazy
1:56:53 ass new [ __ ] material that oh thank
1:56:55 you for the for making me look like the
1:56:59 Dopey
1:57:00 Prince
1:57:09 um I just I mean imagine just with the
1:57:12 current technology as it is
1:57:15 today the the the the at least 50% of
1:57:19 Americans let's let's call it what it
1:57:21 really is
1:57:25 probably 90 to
1:57:27 95% of
1:57:29 Americans really don't know what AI is
1:57:32 really don't know what we've all
1:57:35 experienced 95% of Americans don't
1:57:47 know hey Quinn how's it
1:57:50 going hey it's going well here how about
1:57:53 you good I want you to write a poem
1:57:55 where all of the lines end with
1:57:58 something that rhymes with the word
1:58:01 orange oof that's a tough one the word
1:58:04 orange is notoriously difficult to rhyme
1:58:06 with how about a creative work around
1:58:09 with near rymes love
1:58:11 it all right here it goes in a Grove
1:58:14 were story sporin an artist with words
1:58:17 does forage crafting Tales where dreams
1:58:20 can porridge in this whimsical land
1:58:23 of 95% of
1:58:27 Americans have no [ __ ] idea that
1:58:29 that's
1:58:33 possible the thing that Mimi just
1:58:35 experienced I had all these notes they
1:58:38 were all disorganized I put it in this
1:58:40 thing holy [ __ ] it organized them oh my
1:58:44 God I can do the thing I've always
1:58:45 wanted to
1:58:48 do I think I think we're under
1:58:54 estimating the power of what's already
1:58:57 been
1:58:59 created right because we're all paying
1:59:01 attention to this stuff we're like we
1:59:03 want GPT 5 and what I really want is GPT
1:59:05 6 and I want it to be multimodal you
1:59:07 know what I really want I want to be
1:59:08 able to speak an idea and then I see a
1:59:11 feature film about it like we're
1:59:14 like we're living 10 years in the future
1:59:17 or or at least
1:59:20 five most people on the planet have no
1:59:24 [ __ ]
1:59:26 idea what's actually
1:59:30 here and and as as the ripples from us
1:59:36 the people that are like holy [ __ ] have
1:59:38 you tried this chat GPT thing did you
1:59:41 know you could just stick shitty notes
1:59:43 in it like you could just take pictures
1:59:45 of handwritten notes and then it turns
1:59:47 it into organized stuff and now I
1:59:48 published a book wait you published a
1:59:51 book yeah remember that book I've been
1:59:53 talking about for 20 years the one I
1:59:55 told you you would never finish yeah I
1:59:57 finished that over the weekend
2:00:03 what
2:00:05 like most people don't know what's
2:00:08 here and if I think about
2:00:12 the the value of the the intellectual
2:00:17 Capital that is sitting in people's
2:00:19 brains
2:00:22 that they have convinced themselves they
2:00:25 don't have the talent or the training or
2:00:27 the degree or the resources or the money
2:00:32 or the stick tutiven or the non-
2:00:35 neurodivergent neurotypical brain or
2:00:37 whatever it is to be able to do anything
2:00:39 with those ideas so they've been living
2:00:41 for 20 years with this reality of like
2:00:44 I'm a
2:00:48 loser and then all of a sudden they talk
2:00:51 to this
2:00:53 talkie
2:00:54 thing and then at some point that talkie
2:00:57 thing will be like hey why don't you
2:00:58 just do do you have any notes that
2:01:00 sounds like a really cool idea do you
2:01:01 have any notes of that idea well yeah I
2:01:03 got a drawer full of notes why don't you
2:01:05 go open that drawer and just you know
2:01:08 you'll take a picture and then I'll I'll
2:01:09 look at the picture I'll help you put
2:01:11 the notes together oh okay you could do
2:01:13 that okay here click click click click
2:01:16 click there's my notes wow that's a
2:01:18 that's a even more impressive story than
2:01:21 I thought you had oh really yeah here's
2:01:24 an outline oh my you want to turn that
2:01:27 into a book of poetry now I hate poems
2:01:29 oh okay novel not really well let's turn
2:01:33 it into a TV
2:01:34 show
2:01:36 okay that's a business you should start
2:01:39 you want me to start that for
2:01:41 you do you have a name for it yet I'm
2:01:44 terrible with naming well here's 20
2:01:45 names for that business oh
2:01:50 okay like everyone talks about the AI is
2:01:54 going to take our [ __ ] jobs how about
2:01:56 the AI is going to resurrect the 400
2:01:58 [ __ ] ideas I've had over my
2:02:01 lifetime 20 of which are really good and
2:02:04 five of which are going to turn into
2:02:06 businesses no one's talking about
2:02:10 that you know why cuz they're not
2:02:12 hanging out
2:02:15 here by the way if you don't pay the 20
2:02:18 bucks a month meta just dropped advanced
2:02:20 voice for free in in Facebook yeah like
2:02:24 again I think the the other thing that I
2:02:26 think is hard
2:02:27 to overstate is the importance of meta
2:02:31 open sourcing all of their their big
2:02:35 Frontier
2:02:37 models
2:02:40 um I mean I just I just think that means
2:02:42 that you know this stuff's going to this
2:02:44 stuff's going to be here and accessible
2:02:46 to a lot of people for a long time
2:02:52 sorry I just had
2:02:54 to solve a puzzle piece to prove I'm
2:02:58 human learning to monetize ideas from it
2:03:01 will explode yeah and that's so so kif
2:03:05 where where khaki where this thing
2:03:10 gets where this thing starts to really
2:03:13 escalate in
2:03:14 value is
2:03:19 when when we get a gentic
2:03:24 Behavior
2:03:26 so so so Mimi had this Kevin mallister
2:03:29 moment tonight of she took her raw notes
2:03:33 it turned into organized an organized
2:03:35 outline right and then she's going to
2:03:37 take that organized outline she said I'm
2:03:39 not going to sleep tonight so she's
2:03:41 going to be she's going to be blowing
2:03:42 the doors off this idea so what she's
2:03:45 going to end up with by
2:03:47 mour is is is some sort of instantiate
2:03:51 instantiation of that idea right it's
2:03:53 going to be a marketing plan or it's
2:03:55 going to be a book outline whatever it
2:03:57 is doesn't matter what it is but it's
2:03:58 going to be some sort of
2:04:00 thing but let's say it's a business
2:04:04 plan when we have reasoning and agentic
2:04:08 behavior you basically just say hey now
2:04:11 that I have a business plan what do I
2:04:13 need to do to start this business and
2:04:15 it'll give you a whole plan and then you
2:04:17 can just say hey can you go do
2:04:20 that and it will go incorporate the
2:04:23 business for you and it'll go it'll say
2:04:25 here's the information I need from you
2:04:26 and you'll give it that and it'll go do
2:04:27 that for you I need permission for this
2:04:30 you
2:04:31 know website to log into okay I'll give
2:04:34 you permission to that to go do this one
2:04:36 thing okay I'll go do that
2:04:39 right so the the ability to you can make
2:04:44 money when we get reasoning and
2:04:47 agents just it goes It goes [ __ ] like
2:04:51 this
2:04:52 because right now we've got shitty ideas
2:04:56 that are locked behind our
2:04:59 inability to get them out of the [ __ ]
2:05:02 drawer right or out of our [ __ ] head
2:05:07 and the AI that we have right now is
2:05:08 like this powerful right so we can take
2:05:11 our our good ideas that are in shitty
2:05:14 form put them into this thing and it
2:05:16 goes here's the organized outline right
2:05:19 here's the business plan here's the
2:05:20 marketing plan here's a sketch of the
2:05:22 drawing here's a this here's a that
2:05:24 here's a little 3D model that's shitty
2:05:26 but you could at least show people
2:05:29 it we you go from that
2:05:33 to that idea just
2:05:37 goes and just enters the
2:05:40 world and imagine millions of people
2:05:43 doing that now there's an argument that
2:05:46 says well if everyone's doing all these
2:05:48 great businesses who's going to be the
2:05:51 customers
2:05:53 great question I don't know it's going
2:05:57 to be just like the world is
2:05:59 today some small percentage of people
2:06:02 are super successful cuz they've figured
2:06:04 [ __ ]
2:06:05 out there's a bunch of people down below
2:06:08 them that are like pretty successful and
2:06:11 then there's a bunch of frustrated
2:06:12 people looking up at them aspirationally
2:06:15 that's not going to
2:06:16 change but who's in that top 1% I think
2:06:19 is going to change pretty dramatically
2:06:24 because right now you have to have a
2:06:26 music contract and you have to have Rich
2:06:29 daddy and you have to have access and
2:06:31 you have to have gone to
2:06:36 Yale that stuff will always help by the
2:06:40 way but it's no longer going to be a
2:06:46 requirement
2:06:47 so all right chat GPT named the
2:06:53 bike night we had my dealership oh nice
2:06:56 that's
2:06:59 cool take a picture of your messy garage
2:07:02 and have chat GPT how tell you how to
2:07:04 clean and organize it mine's so messy
2:07:06 that I I think it would just say burn it
2:07:09 down and start
2:07:11 over side hustle Mimi this thing totally
2:07:14 did my homework I love
2:07:18 it yeah that's awesome yeah I did the
2:07:21 the other night another thing you could
2:07:23 try Mimi is you could take the um take
2:07:25 the organized notes and throw it into
2:07:29 into uh Claude go to
2:07:31 claw. and tell it to make you an
2:07:33 interactive dashboard of the notes and
2:07:36 it'll write a bunch of code and it'll
2:07:38 turn your notes into an interactive
2:07:39 dashboard where with you can do all
2:07:42 sorts of [ __ ] with it it might be fun to
2:07:43 just play with you can make money with
2:07:45 chat GPT garage sale assistant
2:07:49 exactly that's awesome
2:07:53 oh
2:07:55 man all
2:07:58 right we're victims of neuronetwork
2:08:01 neurosis or artificial intelligence
2:08:03 obsessive curiosity disorder wait so nnn
2:08:08 or AI OCD that's pretty
2:08:12 [Laughter]
2:08:14 good H that's pretty
2:08:19 good there's a website for that
2:08:23 say my name say my
2:08:29 name well the eBay the eBay seller
2:08:32 assistant
2:08:33 joxer um that's existed for a while it's
2:08:37 called Google Lens if you don't know
2:08:39 Google Lens Google lens has been around
2:08:41 for like a
2:08:42 decade um and if you don't know it go to
2:08:46 a yard sale and if you see something you
2:08:47 think is valuable Google Lens it so you
2:08:51 just just launch the Google app and then
2:08:54 you click on the camera icon take a
2:08:56 picture of it and it'll just bring up
2:08:58 other [ __ ] that looks like that and most
2:09:00 of the times it'll bring up eBay
2:09:03 listings I've done that before where
2:09:05 I've I've like looked at something like
2:09:07 that looks kind of interesting and I and
2:09:09 I'll put it in and it pops it up and
2:09:11 it's it's like a stupid doll or
2:09:12 something like that and it's like eBay
2:09:14 $600 and I'm like oh [ __ ] I'm like how
2:09:18 much is this I like $5 I'm like will you
2:09:21 take three
2:09:24 sure I'm like Gary ve at yard
2:09:28 sales then I come home I'm like hey I
2:09:30 got a good
2:09:32 one yeah I did I got a couple of really
2:09:34 good ones I got a bunch of old Bob's Big
2:09:37 Boy merchandise from a guy that was a
2:09:39 manager there for like 30 years I got
2:09:41 like boxes and boxes of it that was
2:09:43 worth a pretty good penny I got a free
2:09:46 saddle that I ended up selling for like
2:09:48 1,500
2:09:50 bucks um
2:09:52 lots of stuff lots of crazy things it's
2:09:55 super cool when you when you find
2:09:58 something you start to get an eye for it
2:09:59 you're like I think this is worth
2:10:02 something but then you end up with a
2:10:04 bunch of crap that's not worth anything
2:10:05 so that's why my garage is a [ __ ]
2:10:07 train
2:10:10 wreck oh man all right I'm going to get
2:10:14 out of here my eyes are crunchy my back
2:10:17 stiff
2:10:18 um all right it's Wednesday night
2:10:22 come back tomorrow if you're new here
2:10:25 there's a lot of people in here I want
2:10:27 you to go check something out go to the
2:10:30 website the
2:10:33 salon. so the AI Salon is a community
2:10:36 that I started
2:10:38 in December of 2022 the week after chat
2:10:42 GPT came out Leah faston and I held our
2:10:45 first meeting of this community of this
2:10:48 group and this this Channel this this
2:10:52 Tik Tock AI learning lab thing is is
2:10:54 kind of an extension of it a lot of the
2:10:55 people that hang out in here also hang
2:10:58 out here and are leaders in this
2:11:01 community and they're really remarkable
2:11:03 people so it's a community of about
2:11:06 2,000 people it doesn't cost anything
2:11:08 quick everyone give me one trick to show
2:11:11 at my speaking tomorrow okay so Source
2:11:14 Camp who's here one other thing she's
2:11:16 really good at is what she calls parlor
2:11:18 tricks so getting these tools to do
2:11:21 something magical um so definitely if
2:11:25 you know definitely show um show the
2:11:28 voice thing Source camp and then I would
2:11:31 say like whoever your audience is maybe
2:11:34 get something from the audience like
2:11:36 have them give you a challenge or some
2:11:40 obscure thing from their industry and
2:11:43 then just live talk to the real- time
2:11:46 voice and have it do some problem
2:11:48 solving in real time I think that would
2:11:50 be quite impressive but but if you have
2:11:52 ideas for parlor tricks for any of these
2:11:55 tools large language models or whatever
2:11:57 pop them in the comments below um so
2:11:59 Source Camp can do that you can also
2:12:01 share them at the salon so if you go to
2:12:04 the salon. a scroll down there's a
2:12:06 button that says join our community
2:12:09 click that
2:12:10 button that'll take you to a mighty
2:12:12 networks for the
2:12:14 salon and it's our it's our community
2:12:18 site where we hang
2:12:20 out and this channel if you scroll down
2:12:23 the page toward the bottom it says
2:12:24 Community clubs AI Irregulars is one of
2:12:28 those clubs all right so go check it out
2:12:32 if you got
2:12:33 ideas pop them in there all
2:12:36 right just looking for fresh
2:12:40 ideas on the spot song that's a good
2:12:48 one question with GPT and AI who's
2:12:52 setting the rules yes we needed it but
2:12:55 shouldn't I say done with AI
2:13:00 no I don't believe so I
2:13:05 mean here's my philosophy on
2:13:10 that
2:13:14 um if you're creating something that
2:13:18 could be confused
2:13:20 for something
2:13:22 real like say you make a photograph of a
2:13:26 famous person doing something that they
2:13:28 didn't
2:13:30 do if you put that in the world and
2:13:32 don't label
2:13:35 it that's shitty or that's defamation or
2:13:40 that's
2:13:42 fraud My Philosophy is that anything I
2:13:45 put in the world that I put my name on
2:13:47 I'm responsible for
2:13:51 so if I'm like the lazy lawyer that a
2:13:55 year ago or so had chat GPT write a
2:13:58 legal
2:14:00 opinion and then he handed it to a
2:14:03 judge and the judge read his legal
2:14:05 opinion and then went to go look at the
2:14:07 case law that supported
2:14:09 it and none of that case law
2:14:12 existed and he went back to the lawyer
2:14:14 and he said uh none of this case law
2:14:17 exists what do you what is
2:14:19 this and the lawyer is like ah I did it
2:14:21 in chat GPT I didn't really proofread my
2:14:24 work well that guy's an
2:14:27 [ __ ] he doesn't it's not ai's
2:14:31 fault that he didn't do his work
2:14:36 right similarly if he had done his work
2:14:39 and actually checked and made sure case
2:14:41 law was there and he handed that to a
2:14:43 judge and the opinion was fine and it
2:14:46 served his
2:14:48 client does it matter that he used AI
2:14:51 does the judge want to know if he used
2:14:53 Microsoft Word versus Google Docs does
2:14:56 it does he want to know if he used
2:14:58 grammarly or spell checker all those are
2:15:02 AI so my philosophy on this is these are
2:15:05 just tools that help
2:15:08 us and if I publish a book and it's it's
2:15:11 it's [ __ ] littered with typos and the
2:15:15 word
2:15:17 delve because I used Ai and I didn't
2:15:20 tell it not to use the word delve and
2:15:22 everyone that reads it goes oh this is a
2:15:23 bunch of AI
2:15:25 trash that's on me so so My Philosophy
2:15:30 is
2:15:32 um do your work use whatever tools serve
2:15:37 you and just be responsible for the [ __ ]
2:15:39 you put in the world and it's no one's
2:15:42 business whether you used AI or not it's
2:15:45 no one's
2:15:48 business none none zero [ __ ]
2:15:56 them that's my two
2:16:00 cents thank you Winston very
2:16:03 much Jay Winston uh thank you great
2:16:05 discussion tonight thank you appreciate
2:16:07 that AI Ru ruined using delve in real
2:16:11 life it totally has like every time I
2:16:14 hear the word delve I'm like nope no
2:16:18 yeah love your two cents thank you
2:16:23 all right I'm going to get out of here
2:16:24 meta's neural wrist interface plus Orion
2:16:27 holographic glasses look mind melting
2:16:29 yeah it's that that's that's pretty
2:16:31 remarkable so if you haven't seen it the
2:16:34 new metag glasses that they came out
2:16:35 with they're they're 10 years in the
2:16:37 making and what Zuckerberg said today is
2:16:40 they're not they're not commercially
2:16:43 viable they're not they're not something
2:16:45 you could just wear and people wouldn't
2:16:47 think you're a [ __ ] weirdo and
2:16:48 they're also too expensive to be a
2:16:50 consumer device
2:16:52 but they're 12 years in the making and
2:16:54 they work and and it's got this this
2:16:56 wristband that comes with it that is
2:16:59 actually it takes the nerve signals from
2:17:03 your
2:17:04 brain the ones that move your fingers
2:17:06 and that's how how you control the [ __ ]
2:17:08 in the
2:17:09 glasses um so it it looks like pretty
2:17:12 remarkable Tech but it's not something
2:17:13 we're going to see it's it's going to be
2:17:15 available for Developers for probably
2:17:17 another 3 years is my guess um but yeah
2:17:21 like 5 years from now we're going to
2:17:23 have glasses that are not dissimilar to
2:17:25 glasses that do all this crazy [ __ ] and
2:17:28 yeah we're going to have neural
2:17:29 interfaces and things like that so
2:17:31 pretty remarkable stuff meta AI glasses
2:17:34 yes yep exactly Lucy to Charlie Brown
2:17:37 you need to delve you're a good man
2:17:39 Charlie
2:17:41 Brown proof of concept 30k for Quinn
2:17:45 yeah exactly all right so go check out
2:17:48 um the salon if you haven't done that it
2:17:50 doesn't cost anything just join that
2:17:52 Community if you're curious about this
2:17:54 AI stuff um subscribe to my channel go
2:17:58 do that right now don't be a
2:18:00 douche just go subscribe and then keep
2:18:04 coming back here night after night after
2:18:06 night I'm here five nights a week I
2:18:07 start around 8:00 P.M Mountain like
2:18:10 that's really late well it's late for me
2:18:13 too I do it five nights a week I try to
2:18:16 be here like two hours a night sometimes
2:18:18 it goes to three if I get really excited
2:18:19 about something
2:18:22 um and the whole idea
2:18:25 is be in the conversation play with AI
2:18:29 if if you haven't had a Kevin mallister
2:18:32 moment like MiMi had tonight if you're
2:18:34 like who's Kevin
2:18:36 mallister remember Home
2:18:39 Alone that
2:18:42 moment if you haven't had one of those
2:18:44 with AI yet keep playing hang out with
2:18:48 these people hang out here and just
2:18:50 start playing with this at some point
2:18:53 you will do something where you're like
2:18:55 oh my God I AI can do
2:19:00 that if you haven't had that moment yet
2:19:04 you deserve to have it it's a [ __ ]
2:19:06 amazing moment what's amazing about AI
2:19:09 is I still have Kevin McAllister moments
2:19:13 I don't know every
2:19:16 week I have big ones like once a month
2:19:21 I had I had a huge one the other night
2:19:23 what was
2:19:24 it I forget like I like I have my mind
2:19:28 blown regularly and sometimes with the
2:19:31 same tool a tool that I already know how
2:19:32 to use and someone in here will go hey
2:19:35 did you know it could do this I'm like
2:19:37 no and then I'll go try it live here and
2:19:39 I'm like holy
2:19:41 [ __ ] hey guys did you see
2:19:46 this that's why you come back
2:19:48 here and you should tell your friends
2:19:51 you should tell your friends that are
2:19:52 like the [ __ ] that are like you know
2:19:54 we're tired of you talking about
2:19:57 AI
2:19:59 Jesse could you
2:20:02 not those are the friends that are going
2:20:04 to be calling you 6 months from now oh
2:20:07 Jesse I just got laid off Could you
2:20:09 teach me how to do
2:20:10 Ai
2:20:14 No I invited you to that AI party and
2:20:17 you never came
2:20:19 [Laughter]
2:20:22 all right I'm going to write an Eminem
2:20:24 rap about how Rotary Club could take
2:20:26 Lion's Club in a street fight that's
2:20:28 perfect actually love it or or do a uh
2:20:31 Source Camp do a uh sharks and Jets uh
2:20:34 Westside Story uh spin-off do do a do a
2:20:38 Westside Story um song between the uh
2:20:42 the Lions and the rotary that's really
2:20:45 good and then put it in sunno and and
2:20:48 have it do a a Westside
2:20:50 Story Musical Battle that's really good
2:20:53 actually all right I'm going to get out
2:20:55 of here um AI video can make meetings
2:20:57 more exciting our team is doing them for
2:20:59 humor and bonding yep perfect perfect
2:21:01 use
2:21:03 case all right so who's in here tonight
2:21:05 which version of chat GPT has The Voice
2:21:07 version um all of it but you no so
2:21:10 that's not true the web the web version
2:21:13 doesn't yet and the Mac version I don't
2:21:16 have it's not on the desktop version so
2:21:18 it's the mobile version of chat GPT so
2:21:22 your iPad or your
2:21:25 iPhone and then in the lower right hand
2:21:28 corner those used to be headphones and
2:21:31 now it's like a little Soundwave and
2:21:32 when you click on
2:21:34 it it gives you not a not a black dot
2:21:37 you'll see it turns into like a cloud
2:21:40 hey Quinn how's it
2:21:41 going like a cloud floating along I'm
2:21:45 doing great how about you I'm doing good
2:21:47 I really like your Cloud animation it's
2:21:49 it's pretty sexy
2:21:52 I'm glad you like
2:21:53 it yeah what's on your mind today
2:21:56 nothing I'm leaving now I'm I've been
2:21:58 talking to my Tik Tok channel for like
2:22:00 two hours and I keep talking to you and
2:22:02 they're like oh she's cute I like her
2:22:04 she's swell they like you they think
2:22:05 you're pretty
2:22:07 cool that's awesome to hear I'm glad
2:22:10 your followers are enjoying our chats
2:22:12 have a great rest of your day and I'll
2:22:14 be heater when you're ready to talk
2:22:15 again you know what we called
2:22:18 them what did you call them we call them
2:22:21 them the
2:22:22 Irregulars the Irregulars I love it
2:22:26 that's got a mysterious Cool vibe to it
2:22:28 perfect for a unique Tik Tok Community
2:22:31 are they like a band of Misfits or is
2:22:33 there more to the story they are totally
2:22:35 like a band of Misfits so here's the
2:22:37 story they came in one night and like I
2:22:39 do these Tik toks nightly these lives
2:22:42 and the same people kept showing up and
2:22:44 one night I said you all are flipping
2:22:47 weird and one of them said we're not
2:22:49 weird we're irregular and the name
2:22:53 stuck that's a fantastic origin story
2:22:56 it's great take care looking forward to
2:23:00 our next chat okay bye
2:23:05 Quinn crazy all right bunch of Misfit
2:23:10 she called you you tell me Quinn's not
2:23:14 smart we have mysterious cool Vibes Papa
2:23:18 JJ Joey all right so who's in here
2:23:21 Jonathan art what was that if you
2:23:23 haven't seen that that is the new
2:23:24 advanced voice from chat GPT it's
2:23:28 available right now to all paid members
2:23:30 so if you pay 20 bucks a month you'll
2:23:32 get access to that um it is quite
2:23:35 remarkable um it is still very early so
2:23:38 it's got like all of this AI stuff right
2:23:40 now it's janky but it is something and
2:23:44 and I'll keep I'll be playing with it
2:23:45 over the next bunch of weeks trying to
2:23:47 get my head around how it works I use
2:23:49 the current meta Ray band AI glasses to
2:23:51 help me see and identify things that's
2:23:53 absolutely amazing as these things get
2:23:56 more and more powerful the
2:23:58 accessibility sort of upgrades for
2:24:00 everyone is going to be staggering but
2:24:03 but people with you know hearing and
2:24:04 vision issues and brain cognitive issues
2:24:07 and paralysis all of that accessibility
2:24:10 is about to just get massive level UPS
2:24:13 it's it's very inspiring very exciting
2:24:16 meta has a new voice feature that's
2:24:17 going to be free in the in the uh
2:24:20 Facebook out Danielle 30 minute goodbyes
2:24:23 I know I can't I just I wish I could
2:24:25 quit
2:24:27 you you regulars all right Sammy
2:24:31 Minister one great to see you Tams
2:24:33 welcome welcome welcome welcome the
2:24:35 daily crunch Brandon thank you so much
2:24:38 um Dr J great to see you papa
2:24:41 JJ cam cat came from Cleveland Tobias
2:24:45 good night Kyle good night
2:24:48 Tobias oh my oh my God who keeps pinning
2:24:51 I know exactly it's I I can't I can't
2:24:55 not comment on it it's so
2:25:01 bad oh man um Jay Winston great to see
2:25:05 you Winston thank you so much sees slug
2:25:08 wow it's really good it felt like you
2:25:11 were talking to a person Jonathan art
2:25:13 it's amazing wait so okay see whoever
2:25:16 keeps pinning I I can't if you haven't
2:25:18 seen this it it's it it gets way better
2:25:21 hey Quinn um I want you to talk really
2:25:23 fast and say a tongue twister where all
2:25:27 of the words begin with the letter and
2:25:29 is and Nancy and I want you to say it in
2:25:32 a Southern accent as if you're from the
2:25:34 Bayou in Louisiana right here we go
2:25:38 Nifty Nancy's niece nah knits nine new
2:25:41 neckerchiefs near Nashville hey Quinn
2:25:44 Quinn that was great but you didn't say
2:25:46 it with an accent I want you to say it
2:25:47 like a a a cinjun accent
2:25:52 all right let me give that cage an
2:25:54 accent a shot Nifty ny's niece Nina nits
2:25:58 n new neerish near Nashville how that
2:26:01 one sound that's good but I want it
2:26:03 longer and I want you to whisper it with
2:26:05 a Southern accent like you're telling us
2:26:07 a
2:26:09 secret all right lean in close Nifty
2:26:13 ny's niece
2:26:16 Nina new Necker CH near Nashville
2:26:19 nightly next never needing new needles
2:26:23 neatly nuanced knots how's that for a
2:26:27 whisper it's it's [ __ ]
2:26:30 Bonkers it's
2:26:32 Bonkers so Google's got one of those
2:26:36 Facebook announced they're going to have
2:26:38 one of those they've got one of those
2:26:40 there's this app called Pi that's got
2:26:42 one of those the pi one's older
2:26:44 technology um but but the this open AI
2:26:48 one this chat GPT one it actually
2:26:51 understands not just your words but your
2:26:53 emotional content and it replies in kind
2:26:57 pretty staggering
2:26:59 stuff that's why it's so fast because
2:27:01 it's not doing translation between Voice
2:27:03 and text it just understands your words
2:27:07 as you talk them like I don't know how
2:27:10 the [ __ ] they figured this stuff out but
2:27:12 they did anyway all right I'm out of
2:27:14 here Robert Rossy thank you so much for
2:27:16 anyone who gave gifts tonight I really
2:27:17 appreciate it I know when I get rambling
2:27:19 I might not ack your gifts as I see them
2:27:22 cuz I A lot of times I don't see what's
2:27:23 going on on screen so thank you if I
2:27:26 didn't acknowledge you thank you thank
2:27:27 you thank you thank you Irregulars thank
2:27:29 you mods uh for all you do pinning
2:27:32 things like that thanks side hustle Mimi
2:27:34 uh I appreciate the ice cream cone
2:27:36 that's sweet um all right I will see you
2:27:39 tomorrow good night good night Winston
2:27:41 thanks so much for the gift um I will
2:27:43 see you tomorrow night 8:00 pm mountain
2:27:45 time here be here come
2:27:48 back did I hear you say source Camp I
2:27:51 did not say Source
2:27:53 Camp Source Camp see you know Romper
2:27:56 Room was easy cuz it was broadcast if
2:27:58 they missed a name no one would yell at
2:28:00 them but
2:28:02 here good night Source
2:28:05 Camp have a good night Kelly have a good
2:28:07 have a good speech it sounds like you
2:28:08 got that speech coming up um side hustle
2:28:10 Mimi good luck with your with your
2:28:12 project
2:28:14 tonight all right peace out everybody