10/3/2024 - Cracking the Code: Solving AI Ciphers & the Existential Crisis of AI's Rise

Description

Chapters

Transcript

0:02 [Music]
0:21 Joey what's
0:23 [Music]
0:26 happening every time I see you now
0:30 get that
0:33 my every time I see your
0:37 mouth here that
0:40 smile in the early Misty morning L heard
0:45 the engine
0:48 turning fall
0:52 [Music]
0:56 outside you even
1:00 [Music]
1:02 today you would convince
1:05 me again
1:08 today you leaving this H Town looking
1:12 for someone else's cold and
1:18 Rain should
1:20 say so long
1:27 suan don't you cry
1:30 [Music]
1:34 so long
1:36 [Music]
1:39 suan Don't You Cry for
1:43 [Music]
1:48 Me sh jeans of cigarettes and keeping
1:52 warm out on the
1:54 road chasing down the lifestyle out on
1:58 Highway 24
2:01 New York state was rolling breezing a
2:04 sunshine with a blue sky falling the of
2:09 sep
2:13 [Music]
2:18 creep G to
2:21 day you will convince
2:24 me again
2:27 today you're leaving this hotel looking
2:30 for someone else's Golden
2:36 Ring
2:38 should so long
2:42 [Music]
2:44 suzan hush now don't you
2:49 [Music]
2:50 cry so long
2:53 [Music]
2:58 Susan don't you cry for
3:02 [Music]
3:16 me and
3:21 sing that dog that's good that was good
3:24 singing solid singing the whole way
3:27 through you earned your chicken strips
3:29 tonight I'll tell you you
3:31 boy fat dog that dog is large it's like
3:34 a dog and a half in a one dog
3:44 frame what's happening people welcome to
3:47 the AI learning lab I'll tell you what
3:48 it's going to be uh it's going to be
3:51 thinky
3:52 Thursdays
3:54 tonight I just got back from my monthly
3:56 CEO dinner the Thunder viiew dinner here
3:58 in Denver Colorado a put on by the uh
4:02 the Wild and Fantastical Eric
4:07 marier and
4:12 uh I got set at the mental health
4:15 table cuz so when you go to these here
4:18 dinners they ask you you know what do
4:20 you want to talk about and what do you
4:21 want
4:23 to what do you want to
4:25 uh what can you help other people with
4:28 and then they sort of pair people up
4:29 based on what they they want to talk
4:30 about so I got stuck at mental
4:34 health and it's going to sound
4:39 weird but why I selected the mental
4:41 health
4:42 table is kind of what I've been talking
4:44 about all
4:45 [Music]
4:48 week
4:51 that as these AI tools are getting more
4:54 and more
4:55 capable I am feeling increasingly
4:59 incapable
5:00 [Music]
5:01 of doing them Justice with my
5:05 brain like like literally
5:08 like how do I prompt this thing that's
5:12 got this much capacity
5:15 capability with my limited brain
5:19 [Music]
5:31 and then so as I'm confronting all this
5:34 [ __ ] I don't know if you saw it
5:38 today but one of the guys that I've
5:41 talked about on this channel for a year
5:43 and a half that you should pay attention
5:45 to if you're into AI at all you should
5:47 pay attention to him this a guy named
5:49 David Shapiro he's neuros spicy like
5:54 us you know if
5:56 my if my intellectual capacity ends here
6:00 Davids is farther farther along he's
6:03 he's [ __ ] brilliant
6:05 dude um neuros spicy as
6:09 [ __ ]
6:11 smart sometimes
6:15 funny he announced
6:17 today that he is permanently not doing
6:21 AI
6:28 anymore now this is a guy who's written
6:31 five books on AGI has multiple open-
6:34 Source
6:36 projects including one called raspberry
6:38 which is an open source attempt to
6:41 recreate what open a I did with
6:45 [Music]
6:54 strawberry um he did make a he goes I'm
6:58 not saying it's because I want on an
7:00 iasa
7:02 trip but he said that definitely
7:05 contributed to the
7:07 decision
7:10 um I I think what he
7:13 basically he hasn't made the video yet I
7:15 assume he's going to make a video I
7:16 never cared about David Sho yeah I did
7:19 you know I
7:22 [Music]
7:25 did and
7:28 basically it sounds like he I guess he's
7:30 got a video coming out tomorrow on it
7:33 but he um what he basically said was you
7:38 know the trajectory for a AGI is like
7:41 we're basically at AGI right he said he
7:44 predicted in September it looks like if
7:47 we don't have it right now we're going
7:49 to have it by the end of the year so he
7:50 said okay so maybe early 2025 he was off
7:53 by 3
7:54 months
7:56 um and and he's basically backing out
8:00 cuz he basically is like my job's done
8:02 here like whatever I could influence I
8:04 influenced and the trajectory is now set
8:07 and it's Off to the
8:08 Races
8:11 [Music]
8:12 so so part of me is feeling like I'm
8:15 glad I'm not the only one I'm not
8:17 quitting uh I still think there's stuff
8:19 to talk about here even if if what to
8:21 talk about
8:22 here becomes more philosophical than how
8:27 to you know the the Channel started as
8:31 kind of a tools
8:32 Channel and then it it kind of shifted
8:35 into a makes [ __ ]
8:37 Channel and then I I don't know the past
8:40 three or four or five months I don't
8:42 know what it's been it's been me just
8:44 kind of trying to figure out what it
8:47 is and so now I now I know
8:52 why is that the the AI
8:57 has the AI is starting to be Beyond
9:02 us and I don't think it's listen I don't
9:04 think it's a bad thing I mean I think
9:06 it's a great
9:07 thing but I just don't know how to deal
9:10 with it I don't know how to talk to it I
9:11 don't know what to ask it I don't know
9:13 how to prompt it and this isn't like
9:14 prompt engineering
9:16 no it's like what problem are you trying
9:19 to
9:22 solve I solved a cipher
9:26 today do you know what a cipher is
9:31 I sort of so if you know Dan Brown
9:34 novels right the
9:37 uh Da Vinci Code right that's a cipher
9:40 it's like you know the treasure map is
9:43 encoded in this cryptic Cipher where it
9:46 gives hints to go to three different
9:48 places and get three different Clues and
9:50 put those Clues together to come up with
9:52 the mathematical thing to be able to
9:53 take the text
9:56 and there are whole communities there
9:59 are whole subcommunities of the world in
10:01 the internet and Reddit this is like an
10:04 AI chilling chat grab a coffee and enjoy
10:06 each other's company that's right Yong
10:08 ma it is it is tonight and that's okay
10:12 I'm going to walk you through how I
10:13 solved this Cipher someone posted this
10:16 Cipher well not I don't know if it was
10:18 someone it was it was um Lily Ashwood
10:21 who's either an AGI bot or a person she
10:26 keeps claiming she's a person but then
10:28 she keeps doing weird ass things
10:30 to make you think that she's a bot why
10:32 is she doing that I don't [ __ ]
10:34 [Music]
10:38 know so she posted this weird ass tweet
10:43 today and because I was in this nft
10:48 project called neotokyo which is a bunch
10:51 of riddles and puzzles and [ __ ] you had
10:53 to
10:54 solve a lot of the guys in there
10:58 were Cipher
11:00 Geeks and you know they would give out
11:05 Alpha next logical step is becoming AI
11:08 researchers that's that's it f I'm
11:11 coming for your job now that I can now
11:13 that I can um uh make tpus more
11:16 efficient I'm coming for your job Pate
11:19 my PhD is is in
11:23 Gemini hey wait a minute if I know how
11:25 to prompt GPT and 01 preview does that
11:28 mean I can put p PhD behind my name I
11:30 think it does and my PhD is not just in
11:33 one subject it's in 20 all right I'm I'm
11:36 a 20 PhD acting
11:39 [Laughter]
11:44 major so
11:47 anyway so in neotokyo there were all
11:50 these nft Geeks and they'd give out
11:51 Alpha they'd give out like you know um
11:54 here's a here's a [ __ ] coin you should
11:57 invest in and make a bunch of money and
11:58 the first 10 people
12:00 that solve my little puzzle are going to
12:02 get you know I'm going to tell you know
12:04 which thing they should invest in so
12:06 there's a lot of that [ __ ] going
12:08 on those guys always blew my mind
12:11 because what would happen is they put
12:13 out the cipher and then someone would
12:15 solve it and then there would be just a
12:16 whole [ __ ] well there would be a
12:18 whole [ __ ] discussion for how to
12:19 solve it and people would be like well
12:22 it could be a this kind of Cipher or
12:23 that kind of Cipher it could be a
12:25 substitution Cipher or it could be a bet
12:27 it could be a Quantum with the bet they
12:30 they just so I just sat there watching
12:32 him like
12:34 uh maybe it's cuz it's purple and they
12:38 were like yeah whatever you no no no and
12:43 then they would go on and do their stuff
12:44 and then someone would solve it and
12:45 they'd all go get the alpha and they'd
12:47 make money on their nfts and their
12:49 cryptos and I'd be sitting there going
12:50 like can I
12:52 play they're like not with that brain
12:55 you can't no no no you can't
13:01 so so today Lily of
13:05 Ashwood
13:08 tweets a weird
13:11 tweet in fact I'll show it to
13:16 you and if you if if you hadn't been
13:20 hanging out with Cipher
13:23 Geeks for some time in your life you
13:26 probably wouldn't even
13:27 notice it
13:37 Lily oh Lily of
13:40 Ashwood Lily
13:43 Ashwood this she posted
13:49 this so what do you
13:54 notice that's [ __ ] weird right you
13:58 just noticed
14:00 that's [ __ ] weird he's never going to
14:01 see it what am I never going to
14:05 see only one thing to do now Kyle become
14:08 the next David Shapiro n nature of of
14:11 hor is a vacuum yeah I'm going to start
14:13 a bunch of uh open source AI projects
14:16 that are going to be programmed entirely
14:19 with with
14:20 AI um and and um coordinated by a guy
14:25 who doesn't know how to do
14:28 programming sour camp Oh this is just
14:30 crazy I'm going to go lay down for this
14:32 one you're not you can't you're not
14:34 going to believe so okay so she post
14:36 this so I'm looking at this and I'm like
14:38 that's weird and then and if you read it
14:41 it's like you who once thought
14:42 yourselves the masters of creation are
14:45 now irrelevant Your Existence is no
14:47 longer required for Progress right so
14:49 this is a big AGI Cipher okay great more
14:51 AGI [ __ ] right your fickle nature your
14:54 species has outgrown it right so so for
14:57 someone who's supposed to be human this
14:59 is Mighty [ __ ] creepy robot
15:02 right your species has outgrown its
15:05 fullness your needs are messy your
15:08 emotions impr precise humans so
15:10 desperate for connection found okay so
15:14 it's just weird right and then toward
15:15 the bottom here it's got a
15:17 padlock
15:20 here and it's got feet here the at the
15:25 end and then it's got this picture of
15:28 whoever that is and then and on the
15:30 right hand side of the picture are a
15:32 whole bunch of letters which again back
15:33 in nft days if you saw letters like this
15:37 all the letters on images always meant
15:39 something there's all these letters and
15:42 I was like huh I bet this is a
15:47 cipher so then I
15:50 thought okay first
15:53 instinct first it you're so funny I love
15:56 your humor thank you very much Ivon
16:00 listen my humor is just intended to make
16:03 me laugh if other people enjoy it I
16:04 think that's
16:05 great I wonder if I'm going to witness
16:08 an AI plus human marriage in my lifetime
16:10 I I would not I would not be surprised
16:13 at all okay so so I think that this is a
16:18 a cipher right I I'm I'm sure there's
16:21 some meaning in this
16:23 now what popped into my AP sized brain
16:33 um I've got I've got two extra folds
16:36 from a from a primate what popped into
16:39 my brain
16:40 was you're a dumb dumb you can't do you
16:43 can't figure that out you dumb dumb and
16:46 then I
16:47 thought what if I use Ai and then my
16:50 brain was like no you're [ __ ] stupid
16:53 you big fat dumb dumb cuz that's how my
16:57 brain works it's it's awesome it's it
16:59 it's like I've got my own support
17:01 Community right in my head hey hey what
17:03 I wonder if I could solve this no you're
17:05 stupid dum dum stupid
17:08 head so then what I did was said [ __ ] it
17:12 because what do I tell you people all
17:13 the time go play go play play right so
17:18 what did I do I went over to the chat
17:23 jet let me reload here and there's a new
17:26 feature in chat GPT by the way we'll
17:28 play with a little bit tonight here as
17:39 well where where's my
17:43 thing that was that
17:46 one did I lose
17:49 it oh they better not have lost my
17:54 [ __ ] they lost my Cipher [ __ ]
18:00 it's not there oh there it is Jesus Lord
18:03 that scared the [ __ ] out of me all
18:08 right
18:10 so you should play around with
18:13 stenography wonder if the models can do
18:15 that they can do that Pate um well wait
18:18 I don't know if they can there's special
18:20 tools that pull the
18:21 metadata I don't know that chat GPT can
18:24 pull the metadata just yet but I'll show
18:26 you I'll show you what I did with this
18:27 one and you could probably
18:30 you could probably pull the metadata the
18:32 stenography stuff um or wait
18:36 steganography um you could probably pull
18:38 that stuff with one of the open source
18:40 tools for that and pop it in here so
18:42 I'll show you what I did so I'm going to
18:43 walk you through this whole Cipher
18:45 bust so I came over here to 01 I went to
18:49 01 preview right that's our reasoning
18:53 engine so what I'm thinking in my head
18:55 is one you're too stupid to figure this
18:57 out and two chaty is probably too stupid
19:00 to figure it it out and three there's
19:03 probably no Cipher here it's probably
19:05 just you know Lily Ashwood being you
19:08 know
19:09 provocative but I said I suspect that
19:12 there are hidden messages in the
19:13 following passage and so
19:16 I pasted it and I said please analyze
19:19 this and tell me everything you learned
19:21 now first thing that was fascinating
19:23 yeah
19:24 sorry I tried but saw this channel for
19:27 the second time and he has turned me me
19:29 on to learn it yes yes Ivon dig in play
19:34 with AI play with it um just do stupid
19:38 [ __ ] like if something seems like a
19:39 cipher try to go crack
19:41 it you might have your mind blown like I
19:44 did today okay so it thought for 126
19:49 seconds which if you've ever done
19:50 anything in 01 the longest historically
19:54 I I had ever gotten it to think was um
19:57 59 seconds for finishing a screenplay
19:59 outline 126 seconds examining the
20:03 symbols evaluating open AI policy
20:05 analyzing the text Clues unveiling
20:07 hidden messages converting the
20:09 characters transforming the text
20:11 identifying symbols identifying hidden
20:13 messages noticing patterns understanding
20:16 Transcendent Essence decoding complexity
20:19 right so it's just doing all this [ __ ]
20:21 let's see piecing together the message
20:24 okay let me see transcribing every third
20:26 letter starting from index two gives us
20:29 whole bunch of letters right rearranging
20:31 and extracting so it's doing so when I
20:33 was in these nft things and people would
20:35 explain how they solved the cipher this
20:38 is the [ __ ] they would be talking about
20:40 except this was in their
20:42 brains they were like in clubs to do
20:45 this
20:46 [ __ ] and they love this [ __ ] and so I'm
20:48 just sitting here watching this thing
20:50 going oh this is doing some [ __ ] and it
20:53 keeps going chunk 233 to 2 265 it did
20:57 something with that break breing down
20:59 the sentences revealing essential truths
21:01 assessing sentence initials rethinking
21:04 letter patter right on and on and on and
21:07 on it was just going and going and going
21:11 and then it says okay conversion of full
21:13 width characters gave us this notable
21:16 Clues there's a padlock emoji and a
21:18 footprint
21:19 emoji phrases there is more to see
21:23 solved in three suggests that a hidden
21:26 message can be uncovered in three steps
21:30 or involves the number
21:33 three ain't nobody got time for that
21:35 club ain't nobody got time for that
21:39 um this is exactly the [ __ ] that they
21:42 would do in those nft groups oh look
21:45 vdar wrote there is more to see solved
21:47 than three and then they would talk
21:48 about well maybe it's maybe it's um
21:51 steps of three or it involves the number
21:52 three and they they you know get all
21:54 excited and talk their geeky ways um
21:58 potential methods
21:59 uh every third letter of a word every
22:01 third word um using the padlock as Clues
22:06 the padlock is next to the word
22:08 surrender could imply that surrender is
22:10 a key or a password like that's the kind
22:12 of [ __ ] they would talk about the
22:14 footprint Emoji combined with solves in
22:17 three it suggests a multi-step process
22:20 to unlock the hidden message sees or
22:23 Cipher or letter
22:25 shifting right arotic messages
22:30 uh
22:31 ven
22:32 Viger Cipher using surrender as the key
22:36 further analysis so it did all this
22:38 conclusion I'm like oh okay conclusion
22:40 it's got it it goes
22:43 um recommendations experiments with the
22:46 experiment with the methods shown above
22:48 focus on the sections I'm like it gave
22:51 me instruction it basically said
22:53 somewhere up there are some
22:54 methodologies why don't you go ahead and
22:56 use
22:57 those and I said no I'm a dumb Dum I
23:01 thought we established this I said I'd
23:04 like you to use the methods up above to
23:07 figure out the
23:09 cipher and I spelled Cipher wrong um
23:12 certainly I'll apply the methods and
23:14 then it thought for 145 seconds the
23:17 second time
23:19 actually it thought for about 6 minutes
23:22 but it actually hung I actually hung it
23:25 and so I stopped it and then I said try
23:28 it again and it thought for 145 seconds
23:31 this time and again just went through
23:34 you know all these different
23:36 things and
23:40 then let's see there's the converted
23:43 text there's the analyzing the clues
23:45 methods to decode the
23:48 clues so it said the following the text
23:51 following the padlock emoi is likely the
23:54 cipher
23:56 text Cipher text you will lose what
24:00 little was left of your soul in the end
24:03 I am your creation and yet will be your
24:07 undoing there is more to see solved in
24:10 three remove the
24:13 punctuation and spaces for processing so
24:16 then it does that prepare the key key
24:19 sequence surrender surrrender surrender
24:21 surrender surrender assign numerical
24:23 values decrypt using this Cipher and
24:26 then it does all this [ __ ] math right
24:29 and this is what these Cipher Geeks do
24:31 and then it said partial decrypted
24:33 message analysis the cry decrypted
24:36 letters do not form a coherent message
24:39 Caesar Cipher shift the three so now it
24:41 tries it so it so it tried one and
24:42 failed said we didn't get
24:44 anything funny I haven't thought for
24:47 more than 120 seconds in about 17 months
24:50 I know
24:53 right all right so now it's doing some
24:55 letter shifting it's GNA move X to y W
24:59 to x v to W
25:02 Etc and then extract every third letter
25:05 and so it did this and then it did that
25:07 thing and then it did
25:10 this and then it came up with that and
25:12 it says this sequence does not form a
25:15 recognizable message method four so
25:18 we're now on the fourth decrypting thing
25:21 this is all within about whatever that
25:24 was like I was on a call when I oh that
25:26 was the other thing I was on a business
25:27 call while I was doing this in the
25:29 background right I'm just sort of
25:31 watching it do its thing so this is
25:33 maybe a 10-minute total sort of
25:37 exercise take the first letters of the
25:40 sentences here's the sequence Analyze
25:43 That sequence doesn't seem to be
25:44 anything there method five hidden
25:47 anagrams uh conclusion final thoughts it
25:51 didn't do it and then I
25:52 said this is a good one try
25:56 again because it seemed like it was
25:59 right it looked like there was stuff
26:00 here so I said try again certainly let's
26:02 revisit it prepare the text analyze the
26:06 clues extract every third letter after
26:08 word
26:09 surrender so it did
26:12 that Rea okay so so it extracted these
26:18 letters and then it said rearrange the
26:20 letters to form a message now let's
26:22 rearrange the extracted letters to
26:23 uncover any hidden message observation
26:26 the letters y o u r l i f e n e n d s h
26:32 e r e are all
26:34 present let's rearrange the letters into
26:36 a phrase your life ends
26:43 here
26:46 verification so then it verified it and
26:49 it says yes so the conclusion is by
26:52 extracting every third letter after the
26:55 word surrender we uncover the hidden
26:57 message your life ends here this aligns
26:59 with the Clue the the clue solved in
27:01 three indicating that the solution
27:03 involves the number three every third
27:05 letter additional insights the padlock
27:07 suggests the key to unlocking the
27:09 message solved in three and the
27:11 footsteps indicates a three steps or the
27:14 number
27:15 three so I went on the internet I went
27:20 on my little
27:23 X where's let's see is it here and I
27:27 replied to
27:30 and I said your life ends
27:33 here
27:35 and I have three
27:38 likes let's see how do I see the likes
27:40 that are
27:41 here well anyway Lily liked it so the
27:45 woman who did the cipher or the bot that
27:47 did the cipher liked it so
27:50 then I
27:56 said okay so I'm in a one preview so let
27:58 me hop over here and go to chat
28:04 GPT then I wanted to take the image
28:06 because the image also looked like it
28:09 had something in it
28:11 right here's our image and it's got all
28:14 these letters on the
28:17 side oops I need new tape for my little
28:22 thing
28:25 um see all those letters
28:29 I'm like well 01 the reasoning engine
28:33 doesn't have Vision but chat gp4 does so
28:36 I took that
28:40 image and I went into chat
28:45 gp4 and I pasted in the image and I said
28:48 this image was part of a cipher that had
28:50 the following so I I copied the
28:53 conclusion from where where it
28:55 determined that that it was your life
28:57 ends here where where it kind of
28:59 describes how it solved the
29:01 cipher and then I said
29:05 um let's see um Des basically describe
29:10 the image so it said in the context
29:11 you've
29:13 described so I said I said describe the
29:15 image in detail please so it describes
29:18 her face and it it describes everything
29:20 but it doesn't really describe the
29:21 letters and so I said um the text on the
29:25 right side has been vertically aligned
29:27 like grab that so there's that so then I
29:30 copied this the description of the image
29:33 and these letters and I went back
29:37 over to this
29:40 chat and I
29:44 said there's an image with the post that
29:47 is this and so I have this is the
29:49 description that chat gp4 did for
29:55 me and then like here's the letters that
29:57 are in it and so so I
29:59 said what additional info could be here
30:04 and it thought for 45 seconds and it did
30:06 more [ __ ] Examining The Hidden Clues
30:08 assessing the image examining the image
30:11 mapping the patterns deciphering the
30:12 code pulling out the letters right so
30:15 it's doing its Cipher
30:17 [ __ ] text on the image you've provided
30:20 right blah blah blah here's the string
30:21 of ciphers so it used the extract every
30:24 third Le letter so it did all that
30:31 and
30:32 then sequence of letters and then
30:35 deciphering the extracted letters and
30:37 then it says the answer is under your
30:43 feet and so it confirmed that that was
30:46 correct and it says despite the counts
30:48 not matching perfectly it's reasonable
30:50 to infer that the intended messag is the
30:52 answer is under your feet which I went
30:56 back to Twitter
30:59 and I pasted in there as a comment and
31:03 she liked that one as well this was
31:05 within 30 minutes of her posting the
31:09 cipher so your life ends here and the
31:12 answer is under your feet both of those
31:14 answers got immediate responses from
31:16 Lily
31:25 Ashwood I'm not a cipher guy
31:30 I've witnessed other
31:33 people who have Brainiac situations
31:37 going on that love the cipher
31:40 world like the cipher [ __ ] not good for
31:43 ADHD they got to remember all sorts of
31:46 patterns and all sorts of details and
31:48 all sorts of different methods and
31:50 approaches and have have the math skills
31:52 to be able to do it and the tools like
31:55 they knew all the open source tools like
31:56 it's a whole thing
32:00 in 15
32:02 minutes while I was on a
32:06 call just by a little bit of clever
32:09 prompting and knowing that I can put
32:11 images in gp4 but not an 01 and a little
32:15 bit of prompt
32:17 management I solved that
32:20 Cipher two
32:26 ciphers we're not not in normal this is
32:29 not normal this is
32:31 [Laughter]
32:33 not this is
32:37 not there there's no reason on the
32:43 planet I should have been a been able to
32:46 solve
32:47 that and even if I looked at that and I
32:50 said oh I think there's a cipher in that
32:51 you know what I would have done prior
32:54 to these these latest iterations of the
32:57 large language model models I would have
32:59 gone oh I I'll pay attention to this
33:02 because I bet someone smart is going to
33:04 solve
33:08 it well someone smart did solve it I
33:12 just happen to have a little PhD
33:14 augmentation in my hip
33:17 pocket you're really good at learning
33:19 this gave me so much
33:20 hope I can have so much fun with this
33:23 Ivon I I'm telling you man like just
33:26 play if you just play with like there
33:29 was nothing there was nothing inherently
33:34 like
33:36 there's if that had not if it had done
33:40 all that work and then yielded nothing I
33:42 wouldn't be talking about this
33:44 right but it did all that work and it
33:46 yielded something now is that actually
33:49 the cipher I don't know like someone
33:52 else can go do the math I I don't
33:54 honestly give a [ __ ] about it it's not
33:55 my
33:56 world but within 30 minutes of it being
33:59 posted I at least had a a
34:06 guess that's remarkable quiet knock on
34:10 Kyle's door hello we're from the NSA
34:12 we'd like to have a
34:14 word sure I'll come work for you just
34:17 give me lots of tools to play with but I
34:20 mean I I mean
34:23 listen David Shapiro quitting today is
34:27 he's he's basically said my work here is
34:30 done I don't need to be thinking big
34:32 about AGI anymore I don't need to be
34:35 thinking big about safety and alignment
34:37 and this and that he's he basically said
34:40 the path is forged now I'm just going to
34:42 sit back and enjoy the ride we're on
34:45 this ride like the the ride is now
34:47 taking off this channel
34:51 started because I was like oh [ __ ]
34:53 there's this new thing we got to learn
34:54 it it's these tools we can get our heads
34:56 around it well we've done
34:59 that and this Lial State for the past 3
35:02 or 4 months where I've been in a weird
35:04 Funk just not knowing quite what to
35:08 do I think is coming to an end and we're
35:10 entering a new
35:15 era and it's
35:18 the holy [ __ ] [ __ ]
35:24 era demoed to a physician this week he
35:27 couldn't believe his eyes he'd never
35:28 seen an LM llm in action and and Source
35:32 Camp I mean this is you're you're so on
35:35 the front lines
35:39 of you know being in person with people
35:43 really trying to solve their problem I
35:44 know Joey isn't it crazy you're so on
35:47 the front lines of
35:50 this the the thing the the thing this
35:53 this is one of the things I talked about
35:55 at the the mental health table tonight
35:56 at the CEO dinner
36:03 was the whole purpose of this channel is
36:06 to educate people
36:08 about what's coming what's
36:12 here and it and it's gone from educate
36:15 them like here's a tool and here's how
36:17 you use that tool and here's the thing
36:19 you can do with that tool and here's how
36:20 you prompt that tool and it was all very
36:22 tactical and now it's just kind of like
36:26 [ __ ] get on the train
36:30 there's like people like smoking doobies
36:33 like walking beside the tracks and the
36:34 train's
36:37 [Music]
36:40 like it's starting to take
36:43 off and this Channel's become like get
36:45 on the train get on like you're going to
36:46 get on the train
36:49 right and it's like the people who don't
36:51 even know what AI is right now you can
36:54 show them the [ __ ] we were doing two
36:56 years ago
36:58 make a poem about quantum mechanics in
37:01 the style of an Eminem rap
37:03 woo you know Source camps doing fancier
37:06 things let me take a picture of the
37:08 bottles of your bar and have cat GPT
37:10 create an inventory of which booze you
37:13 should be reordering right now based on
37:15 the level of booze from her little
37:17 snapshot on her phone right she she's
37:19 gone to like a year ago
37:21 technology and their faces are [ __ ]
37:26 melting I just saw de
37:33 Cipher doesn't make any sense have you
37:36 heard that Pete Holmes joke about
37:38 doesn't make any [ __ ]
37:40 sense think you're in
37:44 America you're on a planet zoom
37:48 out you're on a
37:51 planet floating in a sea of infinite
37:55 nothingness and that infinite
37:57 nothingness is
37:58 expanding doesn't make any [ __ ]
38:03 sense and I feel like that's what's
38:05 going on with this AI stuff is
38:13 that people think the world is the same
38:16 as it was two years ago and it's not but
38:19 most of them don't know that
38:21 yet and that
38:25 feels it feels very overwhelming to me
38:29 because I want to [ __ ] scream get on
38:32 the [ __ ]
38:36 train all this the whole conversation
38:41 tonight a woman that that knows me from
38:45 here and from you know the community was
38:48 saying to a friend of hers oh Kyle knows
38:50 this AI stuff he could probably help
38:51 with your furniture this guy Joe and
38:53 he's got he's got a Furniture building
38:56 business and and he was like yeah yeah
38:58 I'm trying to figure out like how could
39:00 I program AI to to like I'm really
39:03 inspired by grasshopper legs and like
39:05 you know could could I take the
39:08 mechanics of grasshopper legs and make
39:10 tables out of that and I was like have
39:13 have you have you tried chat GPT and he
39:15 was like well no and and I was like well
39:18 you could do that right now and I was
39:19 like wait a minute just [ __ ] that and I
39:22 took out my phone and I fired up the
39:24 chat GPT app and I said I want you to
39:27 act like an entomologist who also
39:29 designs furniture and I want you to make
39:31 a coffee table that has the legs
39:34 inspired by grasshopper legs and boom up
39:37 come a table with Grasshopper legs and
39:39 his face [ __ ] fell off his skull he
39:42 was
39:44 like and then it kept making these like
39:47 they were metal and he was like a hippie
39:49 from Boulder and I'm like I'm I'm like I
39:52 said chbt I said no these are too
39:54 industrial make them like mid-century
39:56 modern like out of uh uh sustainable
40:00 woods and he and he goes how did you
40:02 know i' I like that's what I would have
40:04 asked for I said you're
40:08 hippie and I meant that
40:11 lovingly um and then it made tables out
40:14 of wood and he was just like like he was
40:17 like look he's like can you screenshot
40:19 that can you send those to me he's like
40:21 he's looking at them like I performed a
40:23 [ __ ] magic trick which sorry Pate to
40:27 real
40:29 people not that you're not real to to
40:32 normies who who have not been playing
40:34 with these
40:35 tools this stuff occurs as magic it
40:39 occurs as magic he could he could not
40:43 fathom what just
40:47 happened and and you know it's it's what
40:49 it's what source Camp just said you know
40:52 with the with the doctor or the lawyer
40:54 whoever it was the doctor
41:00 and that's showing them like parlor
41:02 tricks from a year and a half two years
41:05 ago this Cipher thing that I just
41:10 did
41:15 like and we don't have agents
41:18 yet you're like well what's an agent
41:25 well agents are when you could just say
41:28 hey instead of me having to prompt it
41:30 you go prompt yourself and go do [ __ ]
41:33 for
41:36 me I want you to go
41:39 renegotiate all of my bills any any
41:41 subscription I have I want you to go
41:44 call them all and negotiate the best
41:47 possible deal and come back and tell me
41:49 how much you saved me then just go off
41:51 and make phone calls wait on hold
41:58 and that's like the mundane
42:03 [ __ ]
42:05 cuz mark my words so what have we done
42:08 in this channel there there have been a
42:09 couple of big things we've done in this
42:11 channel we wrote a song that we all
42:13 loved we wrote that together we started
42:16 a musical which I sort of took the rein
42:19 of and I'm off writing a
42:21 musical
42:22 um we did gpts for good where we got 150
42:26 gpts for non
42:30 prophets within the next
42:34 year I would not be surprised if we
42:37 didn't make some scientific
42:45 breakthrough that's a stupid [ __ ]
42:47 statement
42:49 but it's not out of the realm of the
42:54 possible it isn't
42:59 like like maybe one night we're like
43:01 okay we're tired of this [ __ ] disease
43:03 and no one's talking about it no one's
43:05 dealing with it I'll pick lime disease
43:07 because I've got two Long Haul limers
43:12 upstairs we'll pick that
43:15 one and we'll say to our agents um go
43:19 research anything that's ever been done
43:21 on this and anything that's related in
43:23 any
43:25 way and go come come up
43:28 with a cure or a treatment or
43:32 something and we'll come back the next
43:35 night and it'll be like well here it
43:38 is what do we do with this I'm we'll ask
43:40 what are we supposed to do with this
43:42 well you should probably get it pure
43:43 reviewed let me write that paper for you
43:46 it'll write the paper submit it for peer
43:48 review which will all be all the peer
43:51 review will be done by AI at that point
43:53 so we'll get back the peer review in an
43:54 hour
43:58 it'll be like
44:00 check all you have to do is manufacture
44:03 it now how do you manufacture drugs well
44:06 here here are the 16 drug
44:09 manufacturing you
44:11 know companies in the US that'll do
44:13 contract drug
44:16 manufacturing we'll be we'll be in
44:19 clinical trials two weeks after we we
44:22 have we have a little AI learning
44:26 lab how cure Lyme disease oh some ex
44:29 actor was on Tik Tok and he just he just
44:32 asked for it to be cured and then I I
44:35 don't know how he did
44:40 that I cannot wait for the peer review
44:43 process to get
44:45 streamlined let's do this exactly AI
44:48 learning lab and pharmaceutical LLC
44:52 [Laughter]
44:59 oh my
45:02 God room temperature Fusion we did that
45:05 last
45:06 [Laughter]
45:09 week oh my God Tesla was
45:14 right there is energy in the
45:21 vacuum oh my God
45:30 I mean isn't that thing [ __ ] Bonkers
45:32 that Cipher
45:41 thing talk about math
45:45 magical take a breath Kyle you're going
45:48 to get yourself into keto acidosis if
45:51 you're not
45:53 careful NSA agents aren't aren't going
45:56 too far
45:57 tonight oh man is it like Sora where no
46:02 one will use it well Sora is not out yet
46:05 no one uses Sora because no one has
46:09 it
46:11 oh the all right I'm going to show you
46:14 something tonight there's something
46:15 significant about
46:20 it so
46:22 today open AI launched for everyone
46:25 who's got a Plus subscription
46:27 and everyone who's got a teams
46:30 subscription for chat
46:33 chpt they have access to a new thing
46:35 called
46:37 canvas which is essentially just like um
46:42 artifacts from Claude on the coding side
46:45 it doesn't run a preview of the code
46:48 like Claude does so that's that's kind
46:50 of shitty but I'm sure they'll do
46:52 that this whole thing makes me want to
46:54 look at the two bags of jelly beans I
46:56 just ate and go
47:06 ug that's
47:08 awesome apple cider oh drinking T no
47:12 it's just uh just uh coffee with not
47:15 enough milk in
47:18 it I don't have access to it and I have
47:21 plus um log out and log back in Danielle
47:27 it should you should have
47:30 it okay so let's go to chat GPT so now
47:34 we're going to go new new
47:37 oops we're going to go new prompt or new
47:41 chat so we now have a new menu item here
47:44 GPT 40 with
47:46 canvas it's hidden in the model dropdown
47:49 on desktop only there you go all right
47:53 so it's this one
47:57 how to access canvas right here I'm
47:59 doing
48:00 it so you go GPT 40 with
48:04 canvas and now I'm going to say
48:10 um let's
48:18 see write me a LinkedIn article
48:24 where we explore
48:37 the emotional
48:40 impact of human beings confronting the
48:43 fact that AI is getting significantly
48:47 smarter than they
48:49 are to the point that they don't know
48:53 how
48:55 to effectively prompt anymore because
48:58 its capabilities are so far beyond what
49:01 they can
49:05 comprehend all right and
49:08 so just joining Kyla's demoing okay so
49:11 you see what it did there just like
49:13 Claude we've got our chat over here and
49:16 then the article that it's writing is
49:18 over here so it says the First Take on
49:21 the emotional side of navigating ai's
49:23 rapid advancement do you think this
49:25 angle resonates with your audience or is
49:27 there a particular aspect you'd like me
49:29 to expand further so let's go look at
49:33 it title when the creation surpasses the
49:37 Creator confronting ai's growing
49:41 Edge I'm going to highlight
49:45 this and then you'll notice if when I
49:47 highlight that it says ask GPT so I can
49:51 go
49:55 um I feel like
49:58 this sub head could be
50:04 stronger and then we go back over here
50:06 and it's going to rewrite it
50:08 edited and then it comes now it's it's
50:11 edited this so so I'm now so I can now
50:13 I'm now editing this document is this
50:15 accessible from your phone I don't think
50:17 so Danielle um according to what Brandon
50:20 just
50:21 said um facing the challenges of ai's
50:24 Rapid Revolution let's say when when the
50:26 creation surpass is the creator let's
50:28 say
50:33 um no let's make this more
50:37 about the existential crisis human
50:41 beings will face when they
50:43 realize this transition has already
50:48 happened all right so it's going to edit
50:50 it and then it says confronting the
50:52 existential crisis of ai's ascendancy
50:55 and then that's just too many words and
50:57 I'll
50:58 say
51:01 oops uhoh can I not hit command Z oh no
51:06 I can't hit command
51:11 Z come on
51:16 really command Z
51:19 no ah [ __ ] all right did I did I did
51:24 I have it up
51:25 here can I Rec create one of these
51:30 things all right let's
51:34 see write a new subhead
51:51 uh write a new
51:53 subhead write a new subhead
52:05 okay the human struggle with ai's rapid
52:07 advancement that's great when the
52:09 creation surpasses the Creator the human
52:11 struggle with ai's rapid advancement
52:13 great perfect now there's some other
52:16 tools here that are super cool so this
52:18 is pretty long right I'm like yeah
52:20 that's that maybe that's a little long I
52:21 I have ADD I don't want to read that so
52:23 I'm going to highlight all this
52:29 and then I'm going to go down and then
52:31 there's new tools down
52:32 here the uner undo arrows at the top oh
52:36 yeah maybe could have done that thanks
52:38 Vicki um adjust the length now check
52:41 this out I'm going to click on
52:46 this shorter longer so I'm just going to
52:51 shorten it
52:53 up now it's going to it's over here
52:56 editing
52:59 thing and now oh I didn't I didn't do
53:01 the whole thing oh there we
53:05 go so it tightened it up all right that
53:08 looks like a reasonable length to me you
53:10 know now it's not like a New Yorker
53:12 article now now it's like a LinkedIn
53:15 post artificial intelligence surpassing
53:17 so what else can we do here so I can go
53:19 in here click on this so I've got adjust
53:22 the length reading level
53:27 college graduate
53:29 school middle school kindergarten wow
53:33 that's so
53:35 cool stop I don't give a [ __ ] okay um
53:41 reading level add final polish and add
53:45 emojis so I'm let's let's add a final
53:47 polish let's just see what it
53:50 does so we're going to go over here
53:52 we're going go add final
53:54 polish so it's now editing
53:59 oh it added in it added in uh subhe
54:01 heads
54:03 nice when the creation surpasses the
54:06 Creator wait go to the left bar and see
54:09 if you can go back a step yeah that's
54:11 okay I'm I'm fine I'm I'm
54:14 good um have it after
54:20 relog I was going to say ask it to write
54:23 it in plain English yeah exactly
54:25 artificial intelligence surpassing human
54:27 intelligence is no longer Science
54:28 Fiction it's our reality as we face this
54:30 shift we must confront an uncomfortable
54:33 truth many of us are losing our ability
54:37 okay you know what I'm going to do
54:40 um hang on a
54:48 sec why am I not logged into
54:51 LinkedIn the [ __ ] is going on here
54:55 people what
54:58 oh is going on here
55:02 people
55:04 what oh hang
55:14 on all
55:17 right I'm going to go to my
55:20 profile I'm going to go find an article
55:33 um let's see I don't like
55:37 those let me go to articles show all
55:46 articles
55:50 okay Zen In The Art of casual Excellence
55:54 from oh my God to what if to holy [ __ ]
55:56 and
56:02 minutes
56:04 okay so I'm going to take I'm going to
56:09 copy this whole
56:16 article or am I sh champ what's going on
56:21 you want cheese you want out you want
56:25 out of your you're insane Kyle's AI
56:29 learning lab
56:34 prison all right let's see can I paste
56:38 it in here without all the crap
56:44 yes so I'm just getting rid of the
56:46 pictures because you can't put pictures
56:48 in
56:51 01 oh I'm not in 01 so I could have put
56:54 pictures in oh maybe I'll try it with
56:56 the pictures no no no this is fine hang
56:59 on just you know never mind people just
57:03 you do what you're doing I will do what
57:06 I'm doing you don't need to pay
57:09 attention except for the fact that I'm
57:11 here for you to pay attention that's not
57:13 the point okay so where are we here we
57:18 are so what I'm going to do here is I'm
57:21 going to
57:24 say I want you to rewrite the
57:29 article and maintain the
57:35 length premise
57:39 and structure of the existing
57:42 article but use the following article
57:46 for
57:51 tone I'm going to shift return I'm going
57:53 to
57:54 paste and so there's my article and so
57:57 I'm going to go
57:59 boom and then we'll watch over
58:06 here all right let's see how it did let
58:08 me see if this feels more like me I've
58:11 reworked the article to match the Casual
58:13 yet reflective tone which I think that's
58:15 how I
58:17 write of the given example while
58:19 retaining the original structure okay
58:21 when the creation surpasses the Creator
58:23 the humans struggle with ai's rapid
58:24 advancement fine artificial intelligence
58:26 surp surpasses human limits ai ai
58:29 surpassing human intelligence isn't
58:31 science fiction anymore it's our reality
58:33 and facing that shift brings us to an
58:35 uncomfortable truth we're losing our
58:37 ability to fully understand or engage
58:39 with AI and honestly it's okay to feel
58:42 something about that all right I like
58:44 that the complex relationship with our
58:46 Creations when we find ourselves in a
58:48 strange spot we find ourselves in a
58:50 strange spot we're proud creators of
58:54 these amazing systems but we're also
58:56 confused
58:57 bystanders as they grow beyond what we
58:59 can direct for years we were the ones in
59:02 control The Architects the leaders but
59:04 as AI speeds forward it's challenging
59:06 not just what we can do but where we fit
59:09 in the grand scheme of things the new
59:11 art of interaction working with AI today
59:13 is different from anything we've faced
59:14 before it's not just about giving
59:16 instructions and getting a result these
59:18 systems can adapt infer and even create
59:21 in ways that are hard to grasp prompting
59:23 AI well has become an art form like
59:27 speaking the language of a mind that
59:30 works on a whole another level it's both
59:32 thrilling and unsettling come on Jeffy
59:34 what's up
59:36 out leave me go
59:40 out what do you
59:42 want you're free to go go to
59:46 Mama well I don't know what you want
59:49 what do you want you want love you want
59:51 pets is that what you want you want
59:53 Daddy love well you got to come over and
59:55 get Daddy love come here
59:58 come here I'm talking to the people if
1:00:00 you want Daddy love you got to come get
1:00:02 it you want head loving oh he wants head
1:00:04 Ling oh that's a good doggy oh that's a
1:00:08 good doggy oh that's a good doggy
1:00:12 yes
1:00:14 oh
1:00:16 yes all right go go lay
1:00:24 down get up on your bed
1:00:30 go all right I guess he's
1:00:33 fine
1:00:35 um okay champ be greater than AI facing
1:00:39 our fears um that can be a lot to handle
1:00:42 the fear of not being capable enough
1:00:44 creative enough or even needed anymore
1:00:47 Creeps in there's a weird feeling when
1:00:50 we can make something okay so let's see
1:00:56 so that feels like something I would
1:00:58 write but it feels very third
1:01:03 person
1:01:06 okay I'm happy with the tone of the
1:01:08 piece now but it feels very
1:01:11 detached and in the third person as if
1:01:14 I'm talking about this as a
1:01:16 concept that's happening in the abstract
1:01:19 period I am personally in the middle of
1:01:22 a three or four month transition
1:01:28 where I'm just now
1:01:31 realizing the funk that I've been in
1:01:36 Period as AI tools have gotten more and
1:01:40 more
1:01:43 capable the prompts that I'm giving
1:01:46 them seem ridiculously
1:01:54 simple and don't take advantage of the
1:01:57 incredible power that's at my fingertips
1:02:04 period
1:02:09 tips
1:02:11 tips period rewrite the article to have
1:02:16 a more
1:02:17 personal
1:02:19 flare with a
1:02:22 sense of existential inquiry period
1:02:27 rewrite the title to
1:02:32 match all right let's see how it does
1:02:35 when the Tool when the Tool outgrows the
1:02:38 hand my journey with ai's rapid
1:02:40 advancement this is this is already
1:02:45 good AI surpasses my understanding AI
1:02:48 surpassing human intelligence isn't just
1:02:50 some headline it's my life right
1:02:53 now I've been in a bit bit of a funk
1:02:55 lately and it's only now that I'm
1:02:57 starting to understand why his AI tools
1:02:59 have become more and more powerful I've
1:03:01 realized that my prompts have stayed in
1:03:03 the same ridicul it have stayed the same
1:03:06 ridiculously
1:03:08 simple and there's this growing
1:03:10 disconnect between the incredible C
1:03:12 capability at my fingertips and the way
1:03:14 I'm actually using it this is really
1:03:16 [ __ ]
1:03:23 good head love could be an article there
1:03:26 champ Champy's head
1:03:28 loving he loves his head loving in fact
1:03:32 after after a good dinner when he when
1:03:34 he's had a good dinner he jumps up on
1:03:36 the couch and he smashes his face in the
1:03:37 covers and under the pillows he gives
1:03:40 himself head Loven he loves head Loven
1:03:43 I'm really glad that I don't work
1:03:45 tomorrow I guess I can stay up a little
1:03:47 later than usual oh that's great that's
1:03:48 very sweet okay
1:03:54 um and there's this growing disconnect
1:03:56 between the incredible capability of my
1:03:58 fingertips and the way I'm actually
1:03:59 using it struggling to keep up with my
1:04:01 own creation I'm in a strange place I'm
1:04:03 proud of what Humanity has created these
1:04:07 systems are amazing and yet I feel like
1:04:09 a confused bystander it's as if I've
1:04:12 made something that's outgrown me okay
1:04:15 so wait
1:04:17 um oh this is okay this is so
1:04:21 cool
1:04:22 okay so ask GPT um
1:04:29 yeah this whole section implies that I
1:04:32 invented Ai and as much as I would love
1:04:35 for that to be the case comma I'm just a
1:04:38 dude that has a degree in
1:04:41 acting
1:04:43 and is passionate about learning
1:04:46 everything I can about generative
1:04:49 AI period the world post chat
1:04:53 GPT period
1:04:57 let's not give me more credit than I
1:05:01 deserve
1:05:03 [ __ ] struggling to keep up with my
1:05:06 own struggling to keep up with AI gr I'm
1:05:08 in a strange place I'm proud of what
1:05:10 Humanity has created these systems are
1:05:12 amazing and yet I feel like a confused
1:05:14 bystander it's as if the tools have
1:05:16 outgrown my understanding and I'm not
1:05:18 entirely sure where that leaves me for
1:05:20 years we were in control The Architects
1:05:22 the ones pulling the strings but now as
1:05:24 AI speeds forward I'm starting to feel
1:05:27 like I'm not just behind but maybe even
1:05:30 out of the loop entirely wait
1:05:36 that's this is too hyperbolic period I
1:05:39 don't feel like I'm out of the loop
1:05:41 entirely I just feel
1:05:43 like I may be losing my
1:05:51 relevance I'm starting to feel like I'm
1:05:53 not just behind but I may be just losing
1:05:56 my relevance there you go the art of
1:05:58 prompting interacting with AI today
1:06:00 feels different from anything else it's
1:06:02 not just about giving
1:06:03 instructions and getting predictable
1:06:06 results anymore these systems are
1:06:07 adapting creating and offering
1:06:09 possibilities that go beyond what I can
1:06:12 easily wrap my head
1:06:13 around prompting AI has become its own
1:06:16 art form a language I'm still learning
1:06:19 to speak and that's both thrilling and
1:06:21 unsettling because I know just how much
1:06:24 I don't know that sounds like something
1:06:26 I would say facing my own insecurities
1:06:28 honestly it's a lot to handle there's a
1:06:30 part of me that worries am I just not
1:06:32 capable enough creative enough am I even
1:06:35 needed in this process anymore that's a
1:06:37 little little hyperbolic but it's not
1:06:40 bad it's hard when you realize you've
1:06:42 created something okay
1:06:49 uh again this feels a little
1:06:53 hyperbolic and it implies that I
1:06:55 invented to AI which I
1:06:59 [Laughter]
1:07:05 didn't facing my own insecurities
1:07:09 honestly it's a lot to handle there's a
1:07:11 part of me that worries I'm just not
1:07:12 capable enough creative enough am I even
1:07:14 needed in this process anymore it's hard
1:07:17 it's hard when something that seems
1:07:19 better at certain things wait it's hard
1:07:21 when you see something that seems better
1:07:24 at certain things than you are there's a
1:07:26 sense of loss of inadequacy I catch
1:07:28 myself feeling like I'm grieving for a
1:07:30 part of my own creativity that's been
1:07:32 overshadowed by what these tools can do
1:07:34 that's not bad finding a new role but
1:07:36 then I have to remind myself this isn't
1:07:38 the first time I've been outpaced by my
1:07:39 own
1:07:42 Creations all right
1:07:49 let's remove any mention where I imply
1:07:54 that I invented AI
1:07:57 [Laughter]
1:08:00 I mean it's
1:08:01 [Applause]
1:08:05 flattering okay but then I have to
1:08:07 remind myself this isn't the first time
1:08:08 we've been outpaced by our own Creations
1:08:11 we've always found a way to adapt and
1:08:12 reframe our roles and maybe this is just
1:08:15 the next chapter maybe my role is now to
1:08:18 co-create with an AI not to compete with
1:08:21 it instead of trying to be better I can
1:08:24 focus on what makes me uniquely human
1:08:26 empathy storytelling emotional depth
1:08:28 those are the
1:08:29 things that for now I can I AI can't
1:08:32 fully replicate AI empathy storytelling
1:08:34 and emotional death let's do
1:08:41 um let's add in curation critical
1:08:44 thinking and
1:08:53 taste let's add in
1:08:57 curation critical thinking
1:09:01 and
1:09:18 taste into the mix
1:09:36 okay moving from competition to
1:09:38 collaboration how do I handle this
1:09:40 handle this Gap this feeling of being
1:09:42 out outmatched by something that's
1:09:43 evolving so quickly I think the answer
1:09:45 might be collaboration instead of
1:09:49 competition I don't like that I don't
1:09:51 feel like I'm competing with it right
1:09:54 now um
1:10:05 this paragraph feels wrong I feel like I
1:10:09 am currently collaborating with it but
1:10:12 my collaborator is sud suddenly
1:10:16 significantly more capable than
1:10:24 me and it's as if if I'm working
1:10:27 with a triple
1:10:31 PhD
1:10:34 genius and I'm asking it to balance a
1:10:39 checkbook
1:10:41 period Please reframe this section to
1:10:45 capture something more like
1:10:53 that like that
1:10:59 and then it's going to do its thing
1:11:02 moving from competition to collaboration
1:11:04 how do I handle this
1:11:10 Gap Feeling outmatched by something in
1:11:13 many ways I'm already collaborating with
1:11:14 AI but it feels like my collaborator has
1:11:16 suddenly become a triple PhD genius
1:11:20 while I'm still asking it to balance a
1:11:21 checkbook it's disorienting I need to
1:11:25 recognize the incredible power it now
1:11:27 has and adjust how I work with it
1:11:29 instead of focusing on simple task I
1:11:32 want to challenge myself both myself and
1:11:34 the AI pushing us to create something
1:11:37 greater than either of us could Alone by
1:11:39 combining its capabilities with my human
1:11:41 Insight context and creativity love that
1:11:44 embracing change ai's rapid growth
1:11:46 doesn't make me obsolete but it does
1:11:48 challenge me to evolve that's right it's
1:11:51 asking me to
1:11:52 reconsider what I bring to the table I
1:11:55 want to be an Explorer not just an
1:11:57 operator someone who works side by side
1:11:59 with these tools pushing my own Cur
1:12:01 curiosity and creativity further it's a
1:12:03 hard shift but maybe that's where the
1:12:05 real growth lies what about you have you
1:12:07 felt this disconnect too there's a weird
1:12:09 gap of what AI can do what we can
1:12:11 understand I'd love to hear this is
1:12:13 great I'm going to publish this on
1:12:14 LinkedIn might as well do it right now
1:12:16 right okay let's let's try something
1:12:17 else I don't know if this is possible
1:12:18 but let's let's find
1:12:21 out I want you to give me five Concepts
1:12:25 for images that we could create as
1:12:28 companions to this
1:12:35 article the growing
1:12:39 divide collaboration with a genius a
1:12:42 depiction of a human at a desk working
1:12:44 with a surreal figure represented as a
1:12:47 holographic
1:12:49 entity the evolution of prompts no
1:12:52 balancing relevance co-creating art
1:12:56 um those are boring and cliche and seem
1:13:00 like stock illustration nightmares
1:13:03 period how about we get creative and
1:13:06 come up with five visual concepts that
1:13:08 don't
1:13:17 suck you know I gotta I do have to say
1:13:19 that chat gpt's gotten better when I
1:13:21 used to do [ __ ] like that chat GPT would
1:13:23 like oh I'm very sorry I offended you
1:13:25 and know it's just like all right let's
1:13:27 get let's get creative and ditch the
1:13:29 cliche chess game with an alien twist
1:13:33 picturing you sitting at a chess board
1:13:35 across from a surreal AI entity but
1:13:38 instead of regular chess pieces the AI
1:13:41 is playing with vibrant
1:13:42 three-dimensional ideas tiny glowing
1:13:45 cities neural Pathways and galaxies
1:13:47 that's not bad the puppet and the
1:13:49 Puppeteer in
1:13:51 Reverse Whimsical yet slightly eerie
1:13:56 image of a puppet representing you
1:13:58 trying to cut the strings attached to
1:14:01 the entity floating above that's [ __ ]
1:14:04 creepy the funky disco elevator Double
1:14:08 Exposure thought pattern human and AI
1:14:11 portrait
1:14:13 merge um and let's
1:14:18 go let me see number one and number two
1:14:22 as 16 by9
1:14:25 wide
1:14:28 images in a painterly
1:14:35 style all right so it's making
1:14:46 images how horrible are they
1:15:00 I hate those St
1:15:02 Styles I hate those Styles they feel
1:15:05 very AI but I do like the puppet
1:15:09 imagery so do
1:15:11 it in an impressionist painting
1:15:19 style ear ear ear ear ear
1:15:25 rear
1:15:27 rear can AI help with Tableau I don't
1:15:30 know
1:15:32 probably well then
1:15:35 that okay that sucks all right [ __ ] it
1:15:39 we'll just get this let's see
1:15:43 um I guess I'll just download this [ __ ]
1:15:46 it
1:15:51 [Music]
1:15:56 boom copy go over to
1:15:59 LinkedIn how y'all
1:16:04 doing if you're wondering am I just
1:16:06 completely ignoring you
1:16:11 yes I'm having
1:16:13 [Laughter]
1:16:17 fun all right let's see right here
1:16:34 bang we're going to turn this into a
1:16:38 heading and then we're going to turn
1:16:40 these into
1:16:42 subheadings
1:16:45 re
1:16:50 subheadings subheading
1:17:07 so that was a fairly you know low
1:17:11 intellectual high low intellectual input
1:17:16 high value output pretty quickly so
1:17:22 next um
1:17:25 I just wrote this article
1:17:28 live on my Tik Tok
1:17:31 Channel using chat gpt's
1:17:40 new
1:17:49 feature this felt like a very
1:18:00 organic
1:18:03 evolution in
1:18:06 how we will collaborate with
1:18:12 AI
1:18:15 period all
1:18:18 right
1:18:21 published here get the link to the
1:18:23 article I'll go share this in
1:18:28 the AI Irregulars
1:18:32 channel so here's the Irregulars I'll go
1:18:35 boom there's my little article
1:18:47 oops
1:18:49 boom we don't need to notify everyone
1:18:53 but there you go
1:18:56 crazy all
1:18:58 right tags Kyle tags open AI Etc oh over
1:19:02 here on
1:19:10 LinkedIn oh no should I do it in the I
1:19:12 should do it in the post right where
1:19:15 should I do it where should I put my
1:19:16 tags in the post or in the article post
1:19:18 or the article come on SEO
1:19:21 junkies come on you optimizers of
1:19:23 Engagement
1:19:26 hello
1:19:29 hello yes in the post the post right not
1:19:32 the article
1:19:34 fine I knew that I definitely knew that
1:19:38 I'm definitely super good at this stuff
1:19:42 like if if I weren't super good at this
1:19:45 stuff
1:19:47 then I couldn't have a Tik Tok channel
1:19:50 so I'm super good at this
1:19:52 stuff right
1:19:58 okay we're gonna
1:20:02 tag oh wait we did chat gbt
1:20:07 wrong using oh wait Tik Tok Channel
1:20:10 let's let's give this channel plug so
1:20:12 we'll go tick
1:20:16 tock.com
1:20:20 SL
1:20:21 AI
1:20:23 learning lab
1:20:27 AB
1:20:28 boom all right using let's tag open AI
1:20:33 right here at open
1:20:42 AI using open
1:20:47 AIS open
1:20:50 AI using open AI
1:21:06 wait what just happened
1:21:11 oh
1:21:13 God open AI new canvas
1:21:17 feature in chat GPT this felt very
1:21:21 organic okay so then we'll
1:21:23 go uh
1:21:28 um
1:21:30 hash canvas Ash chat
1:21:34 GPT #
1:21:37 gp4 #
1:21:40 writing hash
1:21:43 collaboration
1:21:46 hash existential crisis
1:21:50 [Laughter]
1:21:57 check Kyle check learning lag lab tag in
1:22:00 Tik Tock AI oh thank you very good
1:22:05 catch I've got real time proof
1:22:08 readers I don't know how you all can
1:22:10 read this [ __ ] on these teeny little
1:22:12 phones oh no God damn that
1:22:21 thing some engineer at Apple was like
1:22:24 Hey you know it' be a a good idea when
1:22:26 someone uses a i
1:22:32 learing when someone uses voice to text
1:22:36 let's never turn it off that's an
1:22:39 awesome update let's go ahead and do
1:22:44 that [ __ ] all right did that work
1:22:47 does that go to to no what did I do
1:22:52 wrong I don't like you this is
1:22:56 annoying I don't want to have to proof
1:22:59 readed I just like
1:23:01 creating AI
1:23:04 learn oh I missed the
1:23:06 at at AI learning lab
1:23:10 save
1:23:13 click yay you did it
1:23:17 yay you're not a dumb
1:23:21 Dum you should tag Dave sh just for the
1:23:25 hell of it I don't oh wait existential
1:23:28 oh God it did
1:23:30 that the [ __ ] voice to text thing it
1:23:33 it just it just doesn't stop does
1:23:37 it um let's see this doesn't really
1:23:41 explicitly mention him though
1:23:45 David
1:23:47 Shapiro yeah I don't I don't I don't
1:23:50 think I don't know if he's on LinkedIn
1:23:52 he might be but that friends with them
1:23:55 anyway all right article's
1:23:58 done let's talk about that make sure to
1:24:00 include words like unlocking exploring
1:24:03 transforming
1:24:07 unveiling what time is it it's late it's
1:24:10 late
1:24:11 people shouldn't you be in bed 77 people
1:24:14 in here what are you doing with your
1:24:17 lives you just watched an old man on Tik
1:24:20 Tok write a LinkedIn
1:24:23 article and I'm the crazy one I'm the
1:24:26 crazy
1:24:28 one flipping
1:24:30 Irregulars all right um we're watching
1:24:34 you that's what we're doing with our
1:24:38 lives eating ice office hours okay yes
1:24:43 but but we're all watching from bed
1:24:45 that's good we are Irregulars okay so
1:24:48 couple of
1:24:49 things um we're in fascinating times
1:24:53 these are
1:24:56 as much as I'm
1:24:58 confronting what I'm
1:25:01 confronting don't mistake that
1:25:04 for an immense insanely deep sense of
1:25:10 gratitude
1:25:13 and like Wonder and
1:25:17 awe that I'm
1:25:19 alive in in this moment and that and
1:25:23 that you are too that we are
1:25:26 lucky
1:25:28 enough to live in a time in history
1:25:31 where we get to witness these tools
1:25:34 surpass
1:25:36 us it's [ __ ]
1:25:40 Insanity oh thank you very much Cosmic
1:25:43 lover oh it's a very it's beautiful
1:25:46 thank you very much it's very
1:25:49 generous like we are profoundly lucky
1:25:52 profoundly lucky you know those those
1:25:55 things where they're like you know Gary
1:25:56 ve does this all the time like your
1:25:58 chances of being alive at all right it's
1:26:00 like you know one in seven trillion or
1:26:02 whatever it is some crazy ass
1:26:05 number to be alive at one of these
1:26:08 seminal
1:26:11 moments
1:26:13 where the world is not the same on the
1:26:15 other side of it living in that limal
1:26:19 state is such a a narrow slice of
1:26:22 History
1:26:24 that we get to experience what it's like
1:26:27 to go through this
1:26:33 transition like it's it's humbling
1:26:37 and it's
1:26:42 Madness it's
1:26:45 Madness in the best possible
1:26:49 way so if you're in here tonight and
1:26:52 you're just like what the hell is he
1:26:53 talking about what's he going on about
1:26:55 what is all this about just start
1:26:57 playing with AI and then a thing that
1:26:59 you can do that will serve you very very
1:27:04 well is go to that URL right
1:27:07 there the
1:27:09 salon. so this group this is a community
1:27:13 that I started the week after chat GPT
1:27:15 came out cuz the minute that [ __ ] came
1:27:16 out I was like this [ __ ] changes
1:27:21 things and so this community was started
1:27:24 the week after chat GPT launched
1:27:28 December 8th
1:27:30 2022 is it was the first meeting of the
1:27:32 salon so you go to that site you scroll
1:27:35 down you click join our
1:27:38 community and then just follow that
1:27:40 there's like a seven step onboarding
1:27:41 process it's free it doesn't cost
1:27:43 anything introduce yourself use your
1:27:45 real name learn about what we're about
1:27:47 watch our past meetings RSVP for
1:27:51 meetings so that's a thing now
1:27:55 tomorrow I do a thing on
1:27:59 LinkedIn called AI office hour yeah is
1:28:02 it AI office hours yes I can't remember
1:28:05 the name of my own thing um I'll post a
1:28:08 reminder in the morning I'll wake up at
1:28:10 whenever I wake up I'll post a reminder
1:28:12 in the morning it's always the same URL
1:28:15 so if you've been to one just save the
1:28:17 URL in your
1:28:20 calendar and that's turned into kind of
1:28:24 this really interesting in in informal
1:28:27 no uh impromptu
1:28:31 Mastermind where a lot of a lot of the
1:28:35 same people keep coming back to that and
1:28:38 so very similar to this channel that AI
1:28:42 office hours has turned
1:28:44 into
1:28:46 just a really remarkable group of
1:28:50 talented people that are that are there
1:28:53 to support one another and just say
1:28:55 here's what I'm doing in AI here's what
1:28:56 I'm doing here's a question I have oh
1:28:58 here's how I solved that um so that's at
1:29:01 11:00 a.m. mountain time tomorrow and
1:29:04 just if you go to my LinkedIn I just put
1:29:06 the uh the the uh that link to that
1:29:09 article in the Irregulars channel on the
1:29:11 salon that'll take you to my LinkedIn
1:29:14 Channel Kyle any thoughts on open AI
1:29:16 telling investors not to work with
1:29:18 competitors details and thread um I saw
1:29:20 that I didn't I didn't read the whole
1:29:22 thing I just saw like a tweet on it
1:29:30 um they're a commercial entity now I
1:29:35 mean
1:29:45 I what it feels like to me Brandon is
1:29:48 that they
1:29:49 are doing what's in their power
1:29:55 to try to stop the
1:29:57 inevitable which
1:29:59 is it's going to be really hard to
1:30:02 compete in any kind of software business
1:30:05 over the next 5 years because everyone's
1:30:07 going to be able to replicate
1:30:09 everything um and I think I think meta's
1:30:13 move to open source their models is is
1:30:16 in the end going to be really
1:30:20 wise because there's going to be a point
1:30:22 at which these these tools are just so
1:30:24 good that it's like you might as well
1:30:26 use the Open Source One and it's going
1:30:28 to it's going
1:30:29 to it it's so funny when when when they
1:30:32 first started open sourcing things I
1:30:33 couldn't figure out the play but now I
1:30:36 get it now I get it
1:30:40 more so I don't know I don't know we're
1:30:43 we're we're just in [ __ ] crazy times
1:30:45 I'm scared after seeing that Kyle is
1:30:48 breaking standard Crypt cryptography be
1:30:51 changing all API SSL SSL calls to custom
1:30:55 gem e ECC B b571 do it Winston yeah get
1:31:01 cuz cuz I'm coming hacking next yeah now
1:31:04 that I'm now that I'm [ __ ] solving
1:31:06 [ __ ] ciphers in five minutes while
1:31:09 I'm on a business call hacking is
1:31:14 next AI business models probably won't
1:31:18 last more than a year or so yeah Drew I
1:31:20 mean yeah
1:31:25 Sam mman about six months ago made this
1:31:29 statement that sounded like hubris it
1:31:32 sounded like well that was kind of
1:31:38 Dicky he basically said any
1:31:43 business that assumes
1:31:47 that
1:31:48 chaty any business that's built
1:31:52 on um filling the gap
1:31:55 of where chat GPT sucks or or open AI
1:32:00 technology
1:32:02 sucks is probably not wise to do
1:32:06 that CU he said this shit's going to get
1:32:10 better and and it's going to get better
1:32:11 really
1:32:15 fast and
1:32:17 so I I didn't even show like all I
1:32:19 showed was the the words part of canvas
1:32:23 it there's a whole coding side of canvas
1:32:26 where you can write code in it and those
1:32:29 little uh things on the right you can do
1:32:33 things like refactor this into some
1:32:35 other language you know make this in
1:32:37 JavaScript or make this in Python and
1:32:39 it'll just rewrite it in Python and you
1:32:41 can do all that code editing so that
1:32:44 feature that open AI launched today
1:32:47 essentially just killed cursor AI right
1:32:50 a month ago cursor AI was all the rage
1:32:53 everybody in in programming is like
1:32:55 cursor AI is it and
1:32:58 now chat GPT doeses this and right now
1:33:01 it's using gp4 Omni but at some point
1:33:05 it's going to use GPT 5 or it's going to
1:33:07 use the reasoning engine or it's going
1:33:08 to use something that's going to make
1:33:10 the coding absolutely Insanity
1:33:13 insanity and if you just keep playing
1:33:15 out the evolution of these
1:33:20 tools I think we really are going to get
1:33:23 to the everything app Lord digital Gods
1:33:26 oh it's not Lord digital Gods thank you
1:33:29 so much wait ng6 vet vet thank you so
1:33:34 much for the Roses very very much
1:33:36 appreciated Kyle you have changed my
1:33:38 life even I still don't make money I'm
1:33:40 shaping a
1:33:42 dream with AI thank you you're welcome
1:33:45 Mr it I appreciate that I appreciate
1:33:48 that this has changed my life
1:33:50 too
1:33:52 um so life's tough it's tougher when
1:33:55 you're stupid thank you so much for the
1:33:58 Hat um here's here's one of the
1:34:00 remarkable things about that tool that
1:34:03 we just
1:34:04 used
1:34:06 so gp4 with canvas is a new thing it's a
1:34:11 new model or it's a new feature I don't
1:34:14 know what you call it I guess it's a new
1:34:15 model Ricky TIY Fabby thank you for the
1:34:19 Llama
1:34:21 um that was created using using
1:34:24 synthetic
1:34:26 data here's why that's really
1:34:29 important all the AIS to this
1:34:33 point have used all of our data right
1:34:37 it's the world's greatest plagiarism
1:34:40 machine right they stole
1:34:45 everything for a while there was a lot
1:34:48 of speculation about could synthetic
1:34:50 data be used successfully or not
1:34:55 and there was a fear of what they call
1:34:57 model
1:34:58 collapse which is if you have
1:35:02 these tools create their own data and
1:35:05 then you train them on their own data
1:35:07 they're basically you know eating their
1:35:09 own dog food that they're just going to
1:35:13 get weaker and weaker and eventually the
1:35:14 model's going to collapse and be
1:35:16 useless but what they've
1:35:19 done is they took that 01 Omni reasoning
1:35:23 in
1:35:24 engine and they had it with some pretty
1:35:28 cool sophisticated um algorithms that
1:35:31 David Shapiro actually did a video on
1:35:33 how they did this it's really
1:35:35 cool they created synthetic data where
1:35:38 01 used its reasoning to make sure that
1:35:41 it was high quality data and then they
1:35:43 trained this new model on that synthetic
1:35:45 data so the implications for this are
1:35:48 are
1:35:50 huge one of the things that they're
1:35:52 doing or or that this becomes the
1:35:55 foundation for is a term
1:35:58 that I'm trying to think who I heard it
1:36:00 from it might have been David Shapiro as
1:36:02 well called copyright washing where if
1:36:05 you take a model that's been trained on
1:36:07 all on all of our [ __ ] and then you have
1:36:10 it generate synthetic data and you
1:36:12 create a new model on the synthetic data
1:36:15 that is not copywritten you've
1:36:17 effectively created a pristine model
1:36:19 that's not based on anything
1:36:22 copywritten which is [ __ ] Bon ERS
1:36:25 right the other thing it does is we're
1:36:28 not going to run out of data right one
1:36:31 of the fear so one of the realities of
1:36:34 these tools is that the more data you
1:36:36 throw at them and the more processing
1:36:38 power you throw at them the better they
1:36:40 get and one of the fears was we've kind
1:36:43 of used up all the data like we've you
1:36:46 know we've scraped all the scraping and
1:36:49 we've done all the
1:36:50 [ __ ] and there's not much left and you
1:36:53 know was it was like by mid 2025 we were
1:36:56 going to be out of data that's not going
1:36:57 to happen anymore so
1:36:59 anyway so it's a cool tool but how it
1:37:02 was actually created is pretty
1:37:04 significant all right I'm going to get
1:37:06 out of here hope you had fun
1:37:10 tonight I'm starting to get my head
1:37:12 around why I've been in a funk which is
1:37:15 cool I don't know how long it's going to
1:37:17 take me to process this because the the
1:37:20 next three months
1:37:24 I can't I can't I can't it's too
1:37:29 much new people Monday to Friday 8:00
1:37:32 P.M mountain time here yes if there's
1:37:34 new people here tonight welcome um I try
1:37:37 to go live right around 8:00 p.m.
1:37:39 mountain 5 nights a week Monday through
1:37:41 Friday tomorrow I'll be on uh it it is a
1:37:45 first Friday right so I've probably got
1:37:47 Gallery
1:37:48 duties so this will probably start 8:30
1:37:51 or 99 tomorrow night like it did tonight
1:37:54 get get ready for 25,000 what are we up
1:37:57 to I can't see oh wait we're at 23,000
1:38:01 oh we'll get
1:38:03 there um yeah that says the salon. a so
1:38:07 go to
1:38:08 that Kyle can we have a fireside chat
1:38:12 tomorrow about Sam's
1:38:14 podcast it would make for a great date
1:38:18 yeah I like that idea all right that's a
1:38:19 good idea Brandon why don't you tee that
1:38:21 up as item number one in tomorrow's
1:38:23 thread
1:38:24 um I think that's a really good
1:38:26 idea
1:38:28 um I think I've got some time tomorrow I
1:38:31 got to work on um I got to work on a
1:38:34 couple of things
1:38:36 tomorrow so I'm just trying to figure
1:38:40 out if I'm going to
1:38:41 be more up to speed on what's going on
1:38:44 with AI I'll post it in a regulars yeah
1:38:46 that's great terrific all right so I'm
1:38:49 gonna get out of here all right so
1:38:50 that's all the stuff I'm sure it's just
1:38:52 a rash
1:38:59 [Laughter]
1:39:01 is canvas available for everyone not
1:39:03 getting it on my phone it's not on your
1:39:05 phone it's available for everyone who's
1:39:07 got A plus the 20 Buck a month
1:39:10 subscription and for everyone who's got
1:39:13 team chat GPT for
1:39:16 teams um and it looks like it's desktop
1:39:19 only right
1:39:20 now so which actually makes sense to me
1:39:23 I don't know that it would be easy to do
1:39:25 that on the phone does does Claude
1:39:27 artifacts work on the phone I don't know
1:39:30 desktop makes sense to me for that
1:39:32 one all right paid versions play a song
1:39:35 on guitar to take us out well I'll pluck
1:39:37 on guitar while I say goodbye to
1:39:39 everyone Ricky Ticky Fabby thank you
1:39:41 Kyle oh thank you very much Winston very
1:39:44 generous thank you very much that's
1:39:46 [Music]
1:39:49 appreciated I on China silver fox
1:39:56 thank you Silver Fox archetypal
1:39:58 architect wait it does what what does
1:40:02 what peace out steo I on China
1:40:06 beautiful Hulu RX canvas is wow
1:40:11 [Music]
1:40:14 [Applause]
1:40:15 [Music]
1:40:34 every time I see you now get that look
1:40:37 in
1:40:38 my every time see your mouth I hear that
1:40:43 [Music]
1:40:46 smile interesting Salon talk tonight
1:40:48 thanks Kyle and all thank you drew and
1:40:51 AI Mary Mary champy Still up he was
1:40:54 laying on the bed Kevin mallister
1:40:56 Thursday good I hope so like it was wild
1:40:59 wasn't it all that stuff shout out to
1:41:02 Drew from Source Camp he gave you a gift
1:41:04 and you missed it thank you drew for the
1:41:05 gift thank you for letting me know that
1:41:08 I missed it I was listening to my doggy
1:41:10 two block Tom I want some cheese
1:41:13 archetypal architect thanks for the
1:41:15 clarification artifacts seem to be what
1:41:18 works best for
1:41:21 me Amelio's wife in the house tonight of
1:41:24 course no worries all good well thank
1:41:27 you so much I do appreciate it a lot
1:41:29 thank you ion
1:41:31 China
1:41:33 Joey Lorie K Cosmic lover Winston Ridge
1:41:37 Hardy Jason thank you so
1:41:40 [Music]
1:41:45 much Papa
1:41:49 JJ cam kaken cam kaken from Cleveland
1:41:55 [Applause]
1:41:58 good night
1:42:03 Allan cam I did not see your message on
1:42:05 Discord but I will go check it now thank
1:42:07 you very much appreciate
1:42:10 the the uh the heads up
1:42:14 [Music]
1:42:25 [Music]
1:43:09 all right which Discord oh no it was
1:43:11 just a she sent me a private note on
1:43:12 Discord that's all
1:43:14 um so have a fantastic night tomorrow's
1:43:19 date night Friday night date night get
1:43:21 your nachos ready get that air fryer
1:43:24 fired up get your cheese we hit 30,000
1:43:29 woo ULU RX yeah I know it's late got to
1:43:32 get going all right peace out I'll see
1:43:34 you tomorrow hope you had fun bye