2/3/2025 - Musical Update & OpenAI Deep Research: A Quick Look

Description

Chapters

Transcript

0:00 w
0:09 [Music]
0:29 [Music]
0:43 yo yo yo everybody what's
0:47 happening what's shaking what's going
0:51 down good people on the internet I am in
0:55 New York
0:56 City in uh Time Square area
1:01 I got I got busyness I got business to
1:05 do I'm doing
1:08 business did that work yes no
1:15 why I'm dealing with technical
1:17 situations here people calm down all
1:19 right um I can't talk super loud because
1:23 I'm in a hotel so I don't want to get
1:26 told to shut
1:28 up so give me a little dude uh we're not
1:31 doing a deep dive
1:33 tonight um just because I'm too busy and
1:38 traveling and stuff like that so these
1:39 will be the travel ones will be like a
1:42 little little they'll be a little chill
1:44 they'll be chill they'll be chill so got
1:47 a handful of folks it looks like we've
1:49 got it's just a pile of Irregulars here
1:51 so um if you have anything you want me
1:53 to talk about pop them in there
1:56 um we will we'll make a jelly pod
1:59 tonight
2:01 because why wouldn't we oh we've got
2:03 some exciting news jelly pod is loving
2:07 the AI learning lab making making
2:10 podcasts for them um so we've got some
2:14 good news there
2:19 um
2:21 the new deep research from open AI is
2:25 out I've used it for two pieces of
2:28 research um
2:30 it was quite
2:32 good one of the one of the pieces of
2:34 research I used it for
2:38 was nothing wrong with chill exactly
2:41 James um one of the things I use deep
2:44 research for
2:47 was what was the first thing I did oh a
2:49 sales thing this was actually really
2:52 cool and you could
2:54 if I the problem the problem with both
2:57 operator and deep research which are
2:59 these new kind of of powerful things is
3:01 they're only available in Pro um so
3:04 talking about them is kind of shitty
3:07 because I have access to it but but most
3:09 people don't it's it's like well it's
3:12 nice nice for you Kyle isn't isn't that
3:15 nice you get to play new exciting things
3:18 um and you're correct you should talk to
3:20 me like that um let's
3:26 see I did
3:29 um so the way deep research works I
3:32 think I've got one left I think you're
3:34 allowed three a day like even for the
3:36 200 bucks a month or it's a hundred a
3:38 month or something like that I don't
3:40 [ __ ] know
3:42 um I'll find out um but it goes off and
3:47 it researches right it finds websites it
3:51 reads websites wait you're still in New
3:53 York yeah so I was in Upstate New York
3:56 for the reading tonight I'm in New York
3:59 City we've got city meetings tomorrow
4:02 then we go to
4:03 Jersey
4:06 for half of Tuesday that's
4:10 tomorrow then Wednesday we're supposed
4:12 to go to South Jersey but that meeting
4:14 got canceled so I think we're coming
4:16 back to the
4:17 city and then Thursday we're somewhere
4:20 in Jersey again and then late Thursday
4:22 we come home and then we've got virtual
4:25 meetings when it doesn't [ __ ] matter
4:27 but yeah I'm in New York uh um so
4:30 excited about the music how did the
4:32 music Go the music reading go tell us
4:35 okay we are fans and the team team
4:38 Irregulars um so the the um the reading
4:44 was Saturday um it went very very well
4:48 like better than I expected so when you
4:50 do a reading like that there's kind of
4:54 three three
4:56 possibilities um one of the
4:59 possibilities is
5:01 the thing you thought was great
5:03 isn't right you you have a piece of
5:08 crap the other possibility
5:11 is the thing you have has
5:16 possibility but it's
5:19 structurally it it needs
5:22 rearching and then the third possibility
5:25 is you have a
5:27 show and I was pretty confident we
5:30 weren't going to be in the first
5:31 category but you really don't
5:33 know um I actually was leaning toward I
5:37 thought we were gonna have to
5:38 restructure the show um because in
5:42 reading the show it reads like act one
5:45 doesn't end strong enough strong enough
5:48 um and when we heard it we've got a
5:49 [ __ ]
5:50 show um the woman that played Sydney was
5:53 awesome it was a husband and wife team
5:55 played Kellen and
5:56 Sydney um she was great he was really
5:59 good good um the woman that played
6:01 terara was also really good they brought
6:04 all sorts of cool
6:06 choices
6:08 um there's one there's one scene that we
6:11 wrote so so there's there's a scene
6:14 where um where Sydney says I have a I
6:19 have a confession to make thank you
6:20 Silver Fox Sydney says I have a
6:23 confession to make you know I have a I
6:25 have a secret I want to tell you and the
6:27 secret is she loves him right and so
6:29 he's like like freaked out and he he
6:32 says just I'm going to change the
6:35 subject like just talk about
6:40 gutters and so and so there's this funny
6:43 scene of like I don't know three or four
6:46 you know little monologue little speechy
6:48 things from Sydney where she's talking
6:50 about gutters and they're funny like
6:52 they're really funny um you know you
6:56 know gutters gutters are an often
6:58 overlooked but important part of your
7:00 house you
7:01 know you should clean them regularly
7:04 things like that um and so it's this
7:07 funny little scene and the way the
7:09 actress played it was
7:13 um she played it as if she wanted to
7:17 change the subject back to
7:19 love but couldn't because she's
7:22 programmed to give him what he
7:24 wants and it was [ __ ] tragic it was
7:27 really like oh so
7:30 sad um so that was it was a really good
7:33 choice and then there's there's another
7:34 scene where Kellen and his wife are
7:37 fighting they have a fight in the First
7:38 Act and they make up in the second
7:40 act and so we wrote it like he brings
7:43 her a sandwich and she instantly
7:45 forgives him and you know and they have
7:48 a nice scene together and the actress
7:50 played it like she didn't let him off
7:53 the hook like all the lines are like she
7:55 let him off the hook but she played it
7:57 like no I'm still [ __ ] with you I'm
7:59 still pissing off that was really
8:02 good look like you're in a holding cell
8:06 pretty much that's welcome welcome to
8:09 the the world of corporate hotels it's
8:14 pretty much a holding
8:18 cell I have little little
8:21 coffee hermetically
8:24 sealed paper cups because you know you
8:27 wouldn't want an unsanitary
8:30 raw paper
8:34 cup all
8:36 right um okay couple of things so on
8:41 YouTube now we've made it to the point
8:45 that that we can do business stuff on
8:47 there so there's things like super
8:49 thanks and stickers and things like that
8:51 there's also um two different
8:54 subscription levels if you want to
8:56 support the channel if you want to
8:57 subscribe on YouTube you can do that now
9:00 um we're not we we haven't put enough
9:04 time into like I just haven't had time
9:06 to design what it means to subscribe but
9:08 if you want to support the channel you
9:09 can just know that that's there now um
9:12 yeah Brandon do you want to pop up on
9:14 stage and just talk about it a
9:18 bit yeah hey Kyle uh and hey everyone uh
9:21 live from Cleveland uh not in a holding
9:24 cell uh at the switchboard here the cup
9:28 cont I just read that and I was trying
9:31 trying to pretend I didn't um so first
9:36 of all thank you to everybody
9:39 who uh got viewing hours up that liked
9:43 that followed the channel um it was a
9:46 Herculean effort over the past couple of
9:47 months to get to the point that we could
9:51 get to the level where we could support
9:53 it financially and um Now We're Here
9:55 YouTube um said that we've done enough
9:59 and they're going to let us uh open so
10:02 what we've got we've got two different
10:04 ways um that you four different ways
10:06 really that you can engage with us and
10:08 this is going to be an evolution of
10:10 process uh as well just like everything
10:12 else and subject to change without
10:14 notice but the um as Kyle said there's
10:17 two different levels of subscription uh
10:19 and memberships um which you can join at
10:22 The Irregular level of $3.99 a month and
10:25 these are all monthly memberships um and
10:28 then and you could be a learning lab
10:30 loyalty member at $9.99 a month and um
10:34 you get to even uh have a star next to
10:37 your name you get things that some of
10:39 the things that we've talked about um
10:40 member only live chats if we do any uh
10:44 demos or previews we're going to offer
10:46 those to Members First and then the fun
10:50 one for me is you get if you click if
10:53 you're on the YouTube live stream and
10:55 next to the chat and next to the hearts
10:58 and other um motive icons there's a new
11:01 dollar sign and if you click on that
11:03 you've got four options you can give
11:05 super stickers which I gave one earlier
11:07 in the chat just to see what it did uh
11:11 you can also do super chats which you
11:13 have my word as your producer if you do
11:16 a super chat I will break into whatever
11:18 Kyle is doing and make him read and pay
11:21 attention to what you're doing at that
11:23 moment um because you're paying to play
11:26 effectively uh and then you can also uh
11:28 join memberships in you can also gift
11:30 memberships so if you've got somebody in
11:32 your life that you want to or somebody
11:34 you don't like and you want to subscribe
11:36 them to AI learning lab out a spite um
11:39 that's also
11:40 acceptable it's a good strategy I like
11:42 it so yeah uh it's going to be I would
11:45 love your feedback um in the AI Salon as
11:48 far as anything else that you'd like to
11:49 see from that in their regulars channel
11:51 uh and yeah we'll really more to come
11:54 after um uh after Kyle gets back from
11:59 New York
12:00 and
12:02 uh Vicky just had a comment about
12:05 YouTube lives automatically posting to
12:07 my channel that it's weird they're not
12:09 saved into the videos page they're saved
12:12 into the lives page so they're there we
12:15 might want to look at Brandon is is
12:17 there a way to move them around I don't
12:18 know if there is or
12:19 not okay sounds good I'm gonna back into
12:23 hiding for sure okay cool um and uh
12:27 let's see Taki Taki said I joined both
12:30 YouTube and Tik Tok just in case I need
12:32 to switch sides awesome thank you very
12:34 much um I think Tik tok's going to stay
12:37 but we're still in this weird thing
12:39 where if someone doesn't buy it I think
12:41 that the current Arrangement is I think
12:43 it was 70 days so we're probably 10 days
12:46 into that so there's probably two months
12:49 left of Tik Tok so we'll see whatever I
12:52 you know the there was someone wrote
12:55 legislation
12:57 um that if you download deep seek R1
13:01 it's punishable by 20 years in
13:04 prison so it's like you know
13:08 I it's just going to be an exhausting
13:11 couple of years with legislation because
13:13 the
13:15 legislators this thing is moving so fast
13:18 right like we're on this bullet train
13:21 and the legislators are sort of looking
13:23 at us as it passes Cleveland and they're
13:25 like let us write the law about
13:27 Cleveland and it's like already in
13:29 Pittsburgh you
13:30 know so it's gonna be weird was that
13:36 Elon I don't think so it's going for
13:38 cult status a tax
13:42 deduction was what Elon I don't
13:46 know I don't I didn't I didn't know what
13:49 that
13:51 me law lawmakers have no way of keeping
13:53 up they don't I mean they're going to
13:57 make it illegal to download deep seek
14:00 and then what three days after deep seek
14:02 was
14:04 quen which is apparently better than
14:06 deep seek and then three days after that
14:09 is the new 03 mini
14:12 High chat GPT still still has remarkably
14:16 bad yeah oh oh Elon did not do that
14:19 legislation elon's not writing
14:21 legislation um he's just got control of
14:24 the purse strings of the government it's
14:27 nothing to worry about there I'm sure
14:29 I'm sure it's going to be fine I'm sure
14:32 there's it's going to be
14:35 [Laughter]
14:46 just
14:51 um Bam um the yeah the naming
14:55 conventions that open AI with their with
14:58 their stupid [ __ ] names is they they
15:02 like honest to God like I think they
15:04 think it's cute at this point they know
15:08 it's a [ __ ]
15:10 nightmare
15:12 gp40 great for most questions GPT 40
15:16 with scheduled tasks
15:18 thanks um Brandon let me I'm gonna go to
15:22 this
15:24 view because it's because because I'm
15:26 sharing a window instead of a a tab it's
15:30 too skinny so that puts me a little
15:33 closer to the window anyway um let's go
15:35 look at these
15:37 again GPT
15:39 40 right that's your main model and then
15:42 there's GPT 40 with tasks not to be
15:45 confused with with reasoning or research
15:49 or
15:52 operator then there's 01 for advanced
15:56 reasoning they got rid of 01 Mini but
15:59 they replaced it with 03 mini and then
16:02 because I'm on the Pro Plan I've got 03
16:05 mini
16:07 high and then I've got 01 Pro mode and
16:11 then I can still go to GPT 40 mini and
16:15 gp4 the
16:18 original mixture of experts model um at
16:22 this
16:25 point what what started as a
16:31 a
16:34 good
16:36 um simple interface for
16:39 everyone you now essentially have to be
16:42 a developer right like you have to have
16:44 a developer mind you have to give a [ __ ]
16:46 about learning what these different
16:48 models do you have to figure out what's
16:51 possible um they need to create a chat
16:54 GPT model that understands the GPT
16:58 models okay they just name them after
17:00 Disney
17:03 characters I mean Disney characters
17:05 would be better well frankly Source Camp
17:08 what they need to
17:11 do what they need to do is we shouldn't
17:13 have to [ __ ] choose right we should
17:17 be able to just type [ __ ] in and it'll
17:19 go oh would you like me to go do
17:22 research on that for you or do you just
17:24 want an answer oh go do research
17:27 great right Disney would suit well but
17:29 just the the the basic idea of just
17:32 doing a name um the other thing you can
17:35 use now is co-pilot So within co-pilot
17:38 you now have
17:39 reasoning um you don't have deep
17:41 research um but they do have reasoning
17:44 Source Camp I think that's the only you
17:45 can't do Disney doesn't pay yes or name
17:48 them something we can keep straight yeah
17:51 like like reasoning
17:54 or or research
17:59 like descriptive names but again like I
18:02 I just I go back to the place of and I
18:04 think I think Google's going to need to
18:05 deal with this Google's got this really
18:07 bad right now because their products is
18:08 so fragmented at least chat gpts is sort
18:11 of in the same place you've got
18:14 operators a separate application you've
18:16 got Sora as a separate application and
18:18 then within chat GPT you've got 12
18:21 applications Google's got applications
18:24 you know scattered all over the place I
18:26 think all of these companies need to
18:28 create
18:29 [Music]
18:32 um AI simplification front ends that
18:36 just figure this [ __ ] out for us but
18:37 we'll you know whatever whatever we're
18:40 not there yet you know that's
18:42 what what this tells me is we're still
18:45 early right this is not going to be a
18:47 mass adopted thing while it's this janky
18:50 and and fragmented so but that's all
18:53 right um we have PhD level problem
18:57 solvers in our back pocket now all right
19:00 um so let me show you something really
19:03 cool I'm going to go over to jelly
19:07 pod and if you go to Jelly pod you go to
19:10 the like the main page and then there's
19:13 this discover
19:15 podcast little section wait hang
19:21 on make this bigger oh no I
19:25 can't and you go to discover podcast and
19:28 look what the fact is is a featured
19:32 podcast right in the top right in the
19:34 top our cute little thing my tabs are
19:37 all over the place my tabs are a mess um
19:40 but but yeah what the uh what the fact
19:43 is is we made the big time on the jelly
19:45 pood
19:46 homepage which you know I mean they're
19:48 not gonna they're not you know I asked
19:52 the uh the founder the other day are
19:54 they going to become a media company and
19:55 he said no but if you look at this this
19:59 kind of looks like you know a podcast
20:02 site doesn't it and then they've got
20:04 categories a bunch of different podcasts
20:06 but we're featured Numero
20:08 Uno so let's go make ourselves a new
20:12 podcast so if you haven't seen this
20:14 before this is jelly
20:16 pod they're breadcrumbing it out to us
20:19 yeah exactly uh FY FYI notebook LM the
20:23 speakers are somehow even better can't
20:25 tell their AI wow okay interesting
20:30 all right so we're going to go into what
20:31 the fact and we're going to go new
20:34 episode and then we're going to go to
20:36 chat
20:38 je we're going to go here we're going to
20:41 go
20:42 podcast so if you don't know how we did
20:44 this um the way jelly pod works is you
20:48 design a co-host or two you design what
20:52 you want them to sound like so we had
20:55 chat GPT do that for us
21:00 and then you create a show concept which
21:04 we had chat GPT do we had it come up
21:07 with I don't know 20 something like that
21:10 and then what the fact was the one that
21:11 we we
21:14 liked and then you basically just copy
21:17 your little outline from from the chat
21:25 je boom Oh wait what is this one haunted
21:29 dolls and creepy toys where playtime
21:32 becomes nightmare
21:34 cool all right so we're going to pop all
21:37 that in there we're going to do episode
21:42 7 we going to do add Source then we're
21:45 going to go back to chat Jeep we're g to
21:51 grab what's that
21:55 oh an alternate Wikipedia Source that's
21:58 kind of cool
21:59 cool grab
22:02 that did you bring any construction
22:04 paper with you I didn't but we've got
22:06 we've got a low angle on the keyboard so
22:08 we're not getting bad uh bad reflection
22:11 so it's not
22:14 [Music]
22:20 horrible Okay so we've got three
22:22 sources so we got
22:25 two two Wikipedia pages and our little
22:29 outline from chat
22:31 GPT and then we're going to say create
22:33 so the first thing it's going to do is
22:35 create an
22:41 outline chat GPT just puts a reasoning
22:44 button or deep seek just puts a
22:45 reasoning button which is that's what
22:50 open AI had that for a day and then they
22:52 took it away the more crap they add
22:53 without a good UI the more they create
22:56 the illusion of barrier to entry
23:00 that's an interesting
23:03 thought I don't know if
23:06 I agree with
23:12 that I just think it's I just think it's
23:16 they're a research
23:19 agency I mean they're they're a research
23:22 company and they you know they're not a
23:24 they're not a product company okay um
23:26 welcome people to to
23:30 the
23:32 to episode
23:38 7
23:40 welcome all right so we're going to move
23:43 this up here we're going to
23:46 say
23:48 episode
23:52 7 okay welcome people to episode
23:55 7 remind them that
24:00 that
24:02 this
24:03 podcast is
24:06 sponsored by the AIS salon atth
24:16 salon. and then we've got to go work on
24:22 our our little
24:24 images what the fact image is we got to
24:27 get number seven haunted dolls Oh this
24:30 is
24:31 cool oh wait
24:33 except hang
24:35 on uh I got to change my sharing for the
24:38 YouTube
24:40 people we're g to go we're going to go
24:42 full screen people we'll do it
24:48 live look how cool that is the uh the
24:51 lovely graphics for the what the fact um
24:56 podcasts were made by Corey Sandler she
24:59 made them all in one night she just like
25:02 blasted out 20 of them I sent her I sent
25:04 her all the outlines and she just
25:06 blasted them all out but what we have to
25:08 do here is we have to resize this
25:10 because whatever you can't upload an
25:12 image this big so I go
25:15 768 this is the kind of stupid [ __ ] you
25:18 discover when you're trying to you know
25:20 you're trying to move quickly and then
25:22 you realize
25:24 oh that's stupid
25:28 uh uh this is one of my um desktop jpeg
25:34 one of my skills as a product guys I can
25:37 usually find a bug
25:39 within within five minutes of the
25:42 engineers telling me that they've got
25:45 the perfect
25:48 build and it's always stupid [ __ ] like
25:51 uh could we upload an image bigger than
25:54 you
25:55 know the thumbnail file too large still
25:58 too large oh did I do the what what did
26:01 I
26:07 [Music]
26:10 do is that not the right um I saved the
26:15 wrong one hang
26:17 on duplicate as Tiff there we go export
26:22 that
26:26 as JPEG
26:29 with that with that go back over here go
26:33 back in
26:34 here get
26:36 there all right is that
26:42 working is that
26:53 working it does not seem to be working
26:59 M thanks Hotel
27:02 internet oh my God I just checked out
27:04 Jelly pod and I'm floored floor River
27:08 Willow neon Beats I know right I hate
27:11 not using a mouse I know it sucks
27:13 doesn't it why is this not
27:15 uploading hang on let me let me reload
27:18 the page maybe I'm having some you have
27:20 too many tabs open
27:23 sir sir sir we're going to need you to
27:26 close some tabs yeah listen that's the
27:30 file's too large I got the wrong
27:35 file okay hang on hang on what what is
27:39 going on here I'm going to do export I'm
27:42 going to do higher
27:48 compression I'm going to grab that I'm
27:51 going to
27:53 do
27:55 um we'll call this SM for small
28:01 all and we're going to save
28:06 it on the desktop
28:09 save come over here we're going to grab
28:13 this where's the SM
28:19 one I don't get it
28:22 man where' it go
28:25 man
28:26 um SM contains
28:33 SM the
28:36 [ __ ] this is cool you taught me about
28:39 GPT last year I use it all day every day
28:42 awesome River Willow welcome back
28:44 welcome back and I'm I'm so glad that
28:46 that the
28:47 channel helps generate script audio ad
28:51 music yeah yeah okay I I
28:54 gotta what what's happening this is
28:57 making me very sad
29:06 um I'm going to export
29:11 to why did it not save
29:14 it I do not understand we'll do seven SM
29:19 dash we'll put the SM at the beginning
29:21 this time we'll make it smaller
29:31 350k
29:33 okay do we have
29:35 it it's there says it's
29:38 there we're over
29:41 here
29:43 sm all right it's going good
29:49 Lord okay okay so this is one of those
29:54 this so reminds me of the early days of
29:57 the worldwide web
29:59 where you were on dialup modems and and
30:02 you you went to show someone a website
30:04 and you'd be like I oh I got this cool
30:06 website I want to show you I got this
30:07 cool website I want to show you and then
30:10 um you'd be like okay it's it's gonna
30:13 load in a sec it's gonna be so good okay
30:15 just hang on just it's gonna be so good
30:16 and then like 15 minutes later
30:18 everyone's left you know you're sitting
30:20 there with your stupid [ __ ] computer
30:22 all right we're going to generate the
30:23 script now bang all right let me go
30:25 change my sharing again
30:30 can delete that copy let's go back to
30:34 this
30:37 [Music]
30:40 um present I think I should do I'm going
30:43 to do this as a tab so we can share the
30:50 audio remember Joe Pizza's open till
30:52 about 3:00 a.m. I may just head out I'm
30:55 I'm in Time Square so I may just head
30:58 out after this and go walk the
31:01 streets because why [ __ ] not
31:07 um transcript editor what
31:12 why oh it's still generating
31:17 okay I could do a live I could do a live
31:20 uh Q&A from Time Square but I don't want
31:23 to I don't want I don't want to get
31:25 mugged with my little
31:29 with my
31:30 little Gimbal and my little stupid
31:34 [ __ ] iPhone in my face and someone
31:36 comes up and just punches me in the head
31:38 and steals my phone that would make for
31:40 a bad
31:41 [Laughter]
31:45 trip 93 94 something around there yeah
31:49 94 was when I started in all this web
31:52 stuff I was just talking about the
31:54 worldwide web early days this I mean
31:56 listen we are very much it yeah you said
31:58 93 994 I think we're closer to maybe 95
32:01 96 like it's decently it's decently
32:05 evolved at this point but we are nowhere
32:08 near the hype cycle not anywhere near it
32:12 this [ __ ] is still so janky and slow and
32:16 damn weird all right so Lily Vogue
32:19 welcome to episode seven of what the
32:21 fact so let's make generate the audio
32:28 I can save you about three hours and
32:31 Chase down the mugger New York walk and
32:34 talk we could I could do a New York
32:36 walking talk I like I should do it at
32:39 night because that's when that's when
32:42 the [ __ ] crazies are out I'm just
32:44 gonna be H I'm gonna have to just be
32:46 hyper aware because you know it's it's
32:49 [ __ ] tourist Ville right like if we
32:52 were in the village or we were in some
32:54 cool neighborhood I wouldn't be as
32:56 worried but you know it's time Square so
32:58 they're out to they're out to get all
33:00 the midwesterners and I look pretty
33:02 Midwestern you know I have New York
33:05 attitude but but I've got
33:07 Midwestern
33:09 chops I've got Midwestern
33:13 DNA we should put an Apple Air tag on
33:17 [Laughter]
33:20 Kyle okay are we generating audio is
33:23 that what we're doing yes that's what
33:25 we're
33:26 doing Lily V oh are we there I think
33:30 we're there where's the play
33:33 button oh here all right ready welcome
33:37 to episode seven of what the fact where
33:40 we dive into the Eerie world of haunted
33:42 dolls including Robert and Annabel dolls
33:46 possessed ones never thought I'd be
33:48 talking about that on a Tuesday but here
33:49 we are this episode and the chills it
33:52 may bring is brought to you by the AIS
33:55 salonthe
33:56 salon. a place where technology and
33:58 creativity meet to spark inspiration
34:01 that's fancy speak for they're pretty
34:03 smart folks over there right exactly
34:06 Buster now back to the scary stuff let's
34:09 set the stage for Robert the Doll one of
34:11 the most famous haunted dolls of
34:14 all oh that's pretty good all right so
34:17 we need to what the fact
34:19 MP3 so we'll go
34:23 mp3 mp3 audio
34:28 too good to be true
34:33 opening that's from our
34:40 musical where's the what the
34:43 fact I I have a feeling it got eaten by
34:46 my iCloud storage hang
34:49 on what search contains what what the
34:55 fact
35:00 what's the fact intro push play there we
35:05 go so we do that we'll do
35:10 that we'll do
35:14 that 30 years being on the forgot the
35:17 lead in music not yet I
35:21 haven't morning what tool are we
35:23 exploring so what this tool is is called
35:25 jelly pod and this is now our seventh
35:28 episode of what the
35:31 [Laughter]
35:36 fact into facts that'll blow your
35:41 [Music]
35:54 minding welcome to episode seven of what
35:57 the fact where we dive into the Eerie
35:59 world of haunted dolls including Robert
36:01 and Annabel dolls possessed ones never
36:05 thought I'd be talking about that on a
36:09 Tuesday oh Bucky mclin talk oh it's
36:13 Bucky McGill Bucky McGill Cy um Kyle did
36:17 you check out open AI deep research yet
36:19 yeah I have we'll play with that tonight
36:21 I think I have one one research left
36:23 today I think you you get a 100 a month
36:26 or you get three a day and I don't know
36:28 if it's three a day or 100 a month like
36:30 could I do 99 in a day and then just one
36:33 more for the rest of the month I don't
36:35 know um okay publish
36:40 episode ready to go I'm good right I'm
36:42 pretty sure I'm
36:44 good
36:46 boom and then what we're going to
36:49 do man we really need to update Tik Tock
36:52 I don't know what that
36:54 means oh my God the voices aren't they
36:57 hilarious well so and River Willow we
37:00 designed those like we had chat
37:03 GPT I I told chat GPT that I wanted it
37:06 to design voices that were comically
37:08 different from one another and that's
37:10 what it came up
37:13 with they're pretty
37:17 good he's like buer
37:21 buer okay episode seven so we're going
37:25 to click on this we're going to say c
37:29 copy and then we're going to go to the x
37:33 x is what they call Twitter oh there's
37:37 a priest piing a baby how nice
37:43 um let's
37:46 see oh you know what I need to do wait
37:50 can you all see this oh no tick to or
37:52 YouTube can't hang on hang on
37:55 people you think this is easy
37:58 I'm on a shitty little desk in a shitty
38:01 little hotel room um with a single
38:05 monitor trying to do all
38:07 this window streamyard Studio no that
38:12 window
38:14 yes
38:16 okay there we go um don't we recompile
38:21 late on Thursdays okay so let me go over
38:25 here go to my little profile find one of
38:28 these other things so we can copy all
38:30 the text from
38:31 it what the fact boom
38:36 copy go here we'll do a new one we'll
38:40 get rid of this
38:43 one we'll add in our little
38:47 image with the fact we'll put all these
38:50 on the same line
38:58 all right what the fact episode
39:02 seven haunted
39:17 dolls and toys I
39:20 think all right what the fact episode
39:24 seven okay so here's what I need you to
39:26 do people
39:28 um 100 queries per month okay great
39:31 here's what I need you to do go to the
39:33 Tik Tock no go to the Twitter go to the
39:38 X did you link the episode no I didn't
39:42 did you want me to do that I could do
39:48 that shut up oh I got to go get it again
39:51 hang on hang on people just calm down
39:54 cool your flipping Jets everybody's like
39:57 I he did it wrong he's doing it wrong
40:00 this is really painful watching an old
40:02 man try to use the
40:03 computer this is very
40:07 painful all right what the fact episode
40:11 seven looks great so you're going to
40:14 want to go to the X the Twitter the X
40:18 the Twitter the Twitter that where you
40:19 tweet where you post and you Z you Z and
40:23 you post the Z the tweets the twitters
40:30 and I want you to tag this and I want
40:32 you to like like it thumbs up and I want
40:35 you to share
40:38 it all right it's live we're live we're
40:43 live we did
40:45 it all
40:48 right all right I feel like K Kyle reads
40:51 Tik Tock check more than us not true
40:54 maybe I don't know
40:57 not saying a lot but still James low
41:00 shut up shut up shut
41:03 up I do what I
41:06 can
41:08 um you might need to unretire Frank
41:10 while you're in New York that's not a
41:12 bad idea actually let me tell you
41:14 something let me tell you something
41:16 about this whole naming convention open
41:18 AI got going with their [ __ ] 01
41:22 03 we got the 40 the 01 the 03 the 01
41:25 mini the 01 mini High what are they
41:28 [ __ ] high yeah we could do that I
41:31 can't be that loud in my little hotel
41:33 room they're going to yell at
41:35 me
41:37 um so we're live so that's good there we
41:41 go what the fact is out there did any of
41:43 you like it let's see Kelly Camp she
41:46 gets the gold start tonight she was
41:48 first she was first she was first people
41:54 [Music]
41:57 G okay so here's what we're going to do
42:00 we're going to make that go
42:02 away we have a deep research button that
42:05 we're going to turn
42:07 on and I think we're going to do a
42:09 little deep
42:11 researching I miss Frank laughing my ass
42:15 off hey hey hey let me tell you
42:18 something this [ __ ] this
42:20 [ __ ] Sam Alman with his little
42:21 [ __ ] cute little names the cute
42:23 little letters he's got like some sort
42:25 of [ __ ] code it's like if we stick
42:27 all the letters of the names together
42:29 it's going to it's going to make some
42:30 [ __ ] password that's going to unlock
42:33 the [ __ ] Universe [ __ ] yeah
42:36 exactly all right
42:38 anyway I don't know if deep research
42:41 Auto turns it to 03 mini I don't I think
42:46 you have to turn it I saw some people
42:48 bitching about that
42:52 today so so yeah so I'll go if if you
42:55 just leave it on 01 or o0 3 mini the
42:59 Deep research doesn't work yeah I was
43:01 using it today it was it was pretty bad
43:03 it was pretty janky it's so confusing it
43:05 is Dr Bud frost bitten but you can use
43:07 03 mini high for it that's what I'm
43:09 going to do some AI guy put Sam a in a
43:14 deep seek t-shirt that's pretty good
43:16 that's
43:17 solid Kyle can you explain the
43:19 difference between the models in open AI
43:21 okay I'll try
43:30 let me start at the bottom let me start
43:32 at the
43:33 bottom so
43:35 gp4 Legacy model the one that they put
43:38 all the way at the bottom that's
43:40 actually a really good
43:44 model that is
43:47 the that was sort of the Pinnacle of of
43:51 chat
43:53 gpt's um you know pure large language
43:57 model it was it was
43:59 actually 10
44:02 different 170 billion
44:08 parameter specialized models in what's
44:11 called an a mixture of expert models so
44:14 it was 1.7 trillion
44:17 parameters why they've got it at the
44:19 bottom there is I think it's really
44:21 expensive to run but it's really good so
44:23 if you're not getting good results with
44:24 the other things you can try it I mean
44:27 the thing about all of these models at
44:29 this point is they're all so good like I
44:33 I think it's nuanced differences between
44:35 them the I mean the reasoning engines
44:36 are very very different but like gp4 gp4
44:42 mini and GPT
44:45 40 like I don't know anyway um GPT 40
44:50 mini so 40 the O standed for
44:55 Omni so it's it's a
44:58 multimodal model meaning it it can see
45:02 images hear audio
45:06 speak um make
45:09 images but they still haven't released
45:12 it
45:13 fully we're still making images not with
45:16 40 not with the image generation part of
45:19 it so it's weird anyway um so 40 is just
45:23 your basic run-of-the-mill chat GPT 40
45:26 was scheduled
45:28 TKS is um if you want to say every
45:32 morning you want an inspiring
45:35 little they have more
45:39 models than you have tabs they do they
45:42 really do it's exhausting it's
45:45 exhausting um tasks is really cool like
45:48 if you wanna if you want to do an every
45:50 morning inspirational quote and have
45:53 chat GPT write it for you you can do
45:55 that it's pretty slick it's a expensive
45:57 to run compared to 40 yep exactly sunno
46:00 AI cussed in my song a few times today
46:03 yeah I got um refusion I want to go play
46:05 with refusion as well tonight um we
46:08 we'll go play with that we'll make a
46:09 song um I'm gonna let you listen to a
46:12 new Sydney song we we
46:14 um there were two songs in the script
46:17 that didn't exist um all the other songs
46:20 we had um and I did a really cool quote
46:24 a really cool prompt that I didn't think
46:26 would work and it worked beautifully and
46:28 we actually included the song in the in
46:30 the table
46:32 read um did you get deep research yeah I
46:36 have deep research we're going to play
46:37 with that in a second someone's just
46:38 asked me to explain all these models
46:40 okay
46:41 01 um is the advanced reasoning it's
46:46 really really good 0 one's really really
46:49 good um it's available for free users
46:53 and then down here at the bottom is 01
46:56 Pro so because I've got a pro account
46:59 that's the 200 Buck a month account and
47:01 you're like Kyle is it worth 200 bucks a
47:03 month not really but particularly for
47:07 for open AI I want to I want to have
47:09 access to the new thing because they're
47:11 kind of leading the charge with this
47:12 stuff so I want to have access to it so
47:15 thank you AI salon for sponsoring that
47:21 um and then um 03 mini and 03 mini High
47:27 these are the new super fast super good
47:30 better than deep seek models apparently
47:33 I don't [ __ ] know like everything is
47:36 they're all just sort of slightly better
47:38 in most of the benchmarks than each
47:39 other so at this point it's just for all
47:42 of us it's
47:43 it's way more than we
47:46 deserve is is how I would put
47:50 it okay so that's all that stuff
47:54 um Okay so
47:57 with deep research you you flip it into
48:01 gpt3
48:03 gp03 mini high and then one of the
48:07 things that I read on X today was that
48:12 when you use deep research it's actually
48:14 using the full 03 not not 03 mini I
48:19 don't know if that's true or not I
48:20 assume it's not or they would say that
48:25 Kyle have you played with or seen Quinn
48:26 to .5 I have I have seen it if you go to
48:30 chat I think it's chat. quen
48:34 lm.com
48:38 chat.
48:40 quen quen lm.com
48:46 no
48:49 chat
48:51 do quen ll.com
48:55 no is it chat quinn.com can't be
48:59 couldn't it couldn't be actually
49:04 good I bet it's quen ai.com
49:07 NOP let me try quen ai.com I bet that's
49:13 it
49:15 nope qu. a I don't [ __ ] know
49:29 anybody does anybody know what the Quin
49:31 thing is I I was playing with it last
49:34 week chat
49:36 dot
49:41 quen quen
49:44 lm. move your move your mouse Brandon I
49:46 can't see the quen LM so I had it I just
49:50 didn't have I just didn't have it I
49:53 didn't
49:54 have God damn I didn't have the the 17th
49:59 iteration of the stupid [ __ ] URL um
50:03 there it is if you go to it's a really
50:05 easy URL it's super easy to remember
50:07 chat. quen lm.
50:11 axyz
50:13 doart I am so sick of n.com things isn't
50:17 it on hugging face yeah but
50:21 so hugging next week I'm going to do a
50:24 deep dive on hugging face hugging face
50:27 for for normies and this channel is
50:30 really for normies for the most part
50:32 hugging face is confusing as [ __ ] I had
50:36 a doctor with easy Early Access to heds
50:41 on X and its results are
50:45 amazing okay great I I saw your I saw
50:48 your post thank you for tagging me in
50:50 that
50:53 Danielle and and what is heds that's the
50:56 Deep Oh
50:58 that's oh Early Access to chat GPT deep
51:01 research yeah it's I did um I did a
51:06 sales thing today and then I did oh you
51:08 know what I did which is really
51:10 cool I had it for Sydney for the for the
51:14 musical I had to go find any Play movie
51:18 or
51:20 book that had a relationship between
51:24 humans and
51:25 Ai and do research on that and then come
51:28 back and and give us a whole report on
51:30 it it it did it really really
51:33 well oh HDs is that that connective
51:36 tissue disorder yeah yeah
51:38 yeah which yeah that's probably
51:41 something that I need to need to look
51:43 into okay um let's go back to chat JP so
51:48 we got to come up with can you please
51:50 explain the Deep research
51:55 absolutely I wish could
51:59 explain I I wish this
52:02 [ __ ] weren't so nested but it's
52:10 unfortunately I need pizza boxes I
52:16 do
52:19 okay let me let me build up the let me
52:22 build up the Russian dolls here for
52:25 you so at the core at the core of what
52:29 these things
52:31 do you have a you have a large language
52:35 model like GPT
52:38 40 and that's that's the thing that
52:41 we've been using where you know you've
52:43 you've got your large language model
52:45 which is the cup and a condom and
52:48 they've put all of human knowledge that
52:50 they could get their hands on they've
52:51 packed it all in there so this is chalk
52:53 full of human knowledge and when you use
52:56 chat GPT you say hey do you know
52:58 anything and it does its probability
53:01 engine thing and it spits out an answer
53:03 and you're like
53:05 great and it hallucinates and it doesn't
53:08 know what it said and it doesn't know
53:09 how many RS are in Strawberry right all
53:12 that
53:14 [ __ ] trying to think how to do
53:21 this well
53:24 H metaphors are hard when you don't have
53:27 props when you don't have garbage all
53:29 around you that you can you can pull
53:32 um so that's the base
53:37 level the reasoning engines use what's
53:40 called Chain of Thought
53:43 reasoning so it's still making a call to
53:46 an
53:48 llm but it's doing what they call
53:51 thinking before it gives you the answer
53:54 and it do it it it writes its own
53:56 prompts so so you give it a prompt and
53:59 then it takes your prompt and it sort of
54:02 thinks about it and it says well Step
54:04 One is I should it basically does
54:05 planning step one I should do this and
54:08 then step two I should do that and then
54:09 I should think about this and then I
54:11 should think about that and so as it's
54:13 writing prompts it's getting back
54:15 answers and then it looks at that answer
54:17 and it creates the next prompt and so it
54:20 basically goes through this
54:22 reasoning series of prompts that you
54:24 don't have to write you basic basically
54:27 give it a prompt and then it
54:31 thinks
54:33 and it comes back with better answers
54:36 because it has looked at did I
54:40 hallucinate
54:44 whatever deep research takes it one step
54:47 further so you've got your basic prompt
54:49 you've got your reasoning right Your
54:52 Chain of Thought reasoning and then what
54:56 deep re search is doing is it's going
54:58 out to websites reading a single we so
55:01 it goes and finds a website it reads the
55:04 entire website comes back with a summary
55:07 and then based on what it found it adds
55:10 more steps
55:12 into the um the Chain of Thought
55:15 reasoning and then it goes out to
55:16 another website and it comes back and it
55:18 adds more and it's and so it's it's
55:21 literally doing what a researcher would
55:24 do right if you worked at Morgan Stanley
55:26 or you worked at some consulting company
55:28 and you would say hey super Highly
55:31 Educated researcher go find all of the
55:35 mentions go find all of the books that
55:38 mention humans and AI working together
55:40 and go find all of the movies and go
55:42 find and then come back and analyze them
55:45 and tell me which are the most important
55:46 ones and tell me what are the themes in
55:49 them and you know look at go read the
55:51 books and watch the movies and do all
55:53 that [ __ ] or read all the websites or
55:54 all that stuff's done and then so it
55:57 goes off and does that and it can it can
55:59 think for as much as 30
56:01 minutes where it's out looking at
56:04 websites analyzing it and and then what
56:06 it does is it writes up a research
56:09 report so we'll we'll do it right now so
56:11 what I'd like to do is give me some
56:13 ideas on what I could research and then
56:16 we'll go we'll go do that we'll just go
56:18 do
56:19 one um I'm I'm feeling guilty about um
56:23 the YouTubers because they're all
56:25 they're all Tik Tock jealous
56:27 there's there's you know they're like
56:29 our screen's bigger but Tik tok's cooler
56:32 so there's some insecurities going on in
56:34 YouTube so let me just go make sure
56:35 they're
56:37 okay um Rick olssen greetings Kyle
56:40 hugging face is confusing thank you
56:43 thank you for validating my like I want
56:45 to love hugging face but it is confusing
56:48 as [ __ ] you think you think chaty PT's
56:50 confusing go look at the go look at the
56:52 model names on hugging face try to
56:55 figure out what a [ __ ] model is if
56:57 you're not a
56:58 programmer um IKR always
57:02 been where C in a
57:08 cond how to make money with chat gbt I
57:11 didn't bring my button you can make
57:12 money
57:14 with
57:16 um all right let's see research Twi
57:20 training within industry I don't know
57:22 what Twi training
57:24 is I'm down with that research AI agents
57:27 good point Frost research AI agents oh
57:30 that's kind of
57:32 interesting kifs on both YouTube and Tik
57:35 Tock yeah the really cool people are on
57:37 both if if you're not dual screening Tik
57:41 Tok and YouTube you're not hip because
57:44 what you do you put the YouTube channel
57:46 up on your 80 inch
57:49 plasma qled screen that you got it
57:53 curved qled screen so I'm up there
57:57 because you want to see me as big as
57:58 possible because I mean look at
58:00 this and then you you come to Tik Tok
58:03 for your chat that's that's what the
58:05 cool kids are doing Danielle's on both
58:08 she's you know she's a showoff Vicky's
58:12 on
58:16 both K was uncool long before YouTube
58:21 Vicky Vicky concurs
58:28 triple screen valy Valerie Cox say
58:30 triple screening she doesn't want to
58:32 miss any comments on
58:33 X it's a good thing I turned off the uh
58:36 the LinkedIn
58:38 stream
58:40 um all right idea from last week find a
58:43 product on
58:45 ClickBank tell it to create a sales
58:47 funnel well I just want to do it I just
58:49 want to do the research part of
58:52 it um generates a response in a
58:55 particular format with without
58:58 hallucinating I'm on both hang on I have
59:00 to do a physical dexterity
59:03 challenge hey I did
59:07 it my record so far on deep seek is 17
59:11 and a half minutes that's pretty cool AI
59:14 agent um research the condition that
59:18 your boys have yeah maybe I should do
59:20 that let's let's do that it's gonna be
59:22 depressing
59:24 but um okay so let's see um bear with me
59:28 while I while I type I guess I could use
59:30 the talk the talkie talkie typy thing so
59:34 one thing that I noticed is when you do
59:37 when you do a deep research
59:41 query the next time you query the system
59:44 it has turned it
59:46 off so if you want it to keep
59:48 researching you have to keep hitting the
59:50 button the the UI for this is a
59:53 mess research the logistics of import
59:57 importing a 100 Irregulars from around
1:00:00 the world to an exclusive
1:00:02 event in Colorado that would be cool but
1:00:05 it when we're going to do an inperson
1:00:07 event it's going to be in
1:00:09 Vegas by the way because I'm a
1:00:14 degenerate oh is Pluto a planet that's a
1:00:17 really interesting
1:00:22 one let's do okay let's do
1:00:28 look for potential cures that AI can
1:00:32 create this reminds me the old old days
1:00:35 when Kyle took 45 minutes to get back to
1:00:38 us sh up sh
1:00:43 up Vegas for the win yeah baby I'm going
1:00:47 to Vegas [ __ ] [ __ ]
1:00:50 Colorado come to Denver Watch the watch
1:00:54 the Cowboys
1:00:57 um
1:01:03 okay let's
1:01:09 see here's my my reticence to
1:01:13 doing research on stuff that I not so
1:01:16 I'm trying to find something that I
1:01:17 could get my head around hang
1:01:20 on we could do a meeting at Casa Bonita
1:01:23 in Denver that's pretty funny
1:01:27 doing something at the sphere would be
1:01:31 cool all right my record so far 17
1:01:36 minutes we saw that just finished an
1:01:39 online AI certification course use a AI
1:01:41 to answer answer most of the assignments
1:01:44 that of
1:01:46 course
1:01:48 yeah did did you actually learn it
1:01:52 Rob or did you just want the sticker to
1:01:55 put on your LinkedIn profile
1:01:59 because it's all fun in games until
1:02:01 someone actually hires you to do the
1:02:03 thing that you used AI to learn I mean
1:02:06 for students that's one thing I don't
1:02:07 know how old you are Rob I assume if
1:02:09 you're on this channel you're probably a
1:02:11 little older than a student but I don't
1:02:13 give a [ __ ] if you cheat I first of all
1:02:16 AI detectors don't work and second of
1:02:18 all I don't think it's anyone's [ __ ]
1:02:19 business how you answered the questions
1:02:22 that's on you you're willing to put your
1:02:24 name and say you have that certification
1:02:25 and you didn't learn the crap cool go
1:02:28 for
1:02:29 it empowered fundraiser an Murphy was
1:02:32 shaking was shaking
1:02:34 lady um welcome back I'm I'm now on my
1:02:37 trip you were back from your trip trip
1:02:39 now I'm on my trip an and I are behind
1:02:41 on our homework assignment for the um
1:02:44 are you ready for AI training session
1:02:47 the only thing we have to do the only
1:02:50 flipping thing we have to do are we
1:02:52 talking AI Readiness yeah what I'm
1:02:54 talking about I'm shaming us I'm
1:02:56 publicly shaming us all we have to do is
1:02:58 make videos where we're like hi I'm Kyle
1:03:01 I'm an and you know we do the thing but
1:03:04 you know an was like I want to take a
1:03:05 vacation and I'm like I want to go get
1:03:07 business for my company and so we
1:03:10 haven't done
1:03:14 it no we're playing in with um deep
1:03:17 research from uh open AI I have access
1:03:21 to it then we're going to research
1:03:23 something I got to figure out what the
1:03:24 hell we want to research
1:03:28 um
1:03:35 [Music]
1:04:04 H I'm a doc research a medical condition
1:04:09 all right Dr
1:04:11 bod I'll do that I'll do the I'll do the
1:04:15 um I'll do the Lyme disease thing so
1:04:18 I've got if you don't know I've got a
1:04:20 family full of people with with long
1:04:23 haul lime and now all sorts of of other
1:04:26 special diseases like neuropathy
1:04:29 inspired by the
1:04:32 lime
1:04:35 um
1:04:38 okay I want you to write me a
1:04:41 report
1:04:43 on as much research as you can find
1:04:46 on L haul Lyme
1:04:49 disease
1:04:51 comma the related
1:04:54 co-infections like babesia bonella
1:05:09 Etc I want you to
1:05:11 find research on related
1:05:17 conditions like
1:05:18 inflammation neuropathy joint pain
1:05:23 autoimmune
1:05:25 disease cused by these little
1:05:30 [ __ ]
1:05:43 period once we learn
1:05:48 about the
1:05:50 research into the
1:05:54 organisms I want you to split your
1:05:56 research into two categories one would
1:05:59 be
1:06:00 treatments for the related
1:06:05 conditions and one would be research
1:06:11 on treatments that look
1:06:15 promising especially focusing
1:06:18 on recent advancements accelerated by AI
1:06:23 period
1:06:33 finally I want you
1:06:36 to
1:06:42 identify clinical
1:06:51 trials whether existing or upcoming
1:06:56 that we can look
1:07:00 into okay I think that's good so far so
1:07:06 good Dr Kyle in the house yeah I feel
1:07:10 like I feel like
1:07:12 house let's do a differential
1:07:15 diagnosis I love that show that guy was
1:07:18 so [ __ ] bitter he made me happy all
1:07:20 right here we go so thanks for your
1:07:24 detailed request
1:07:26 oh and then it does this which I think
1:07:28 is kind of interesting thanks for your
1:07:29 detailed requests to gather the most
1:07:32 relevant and comprehensive research on
1:07:33 Long hul Lyme disease and its co-
1:07:35 infections I'd like to clarify do you
1:07:39 want damn there's a lot of questions do
1:07:42 you want both mainstream and alternative
1:07:44 so let's see one
1:07:49 yes both main stream and alternative
1:07:53 that's actually a really good question
1:07:56 because if you don't know anything about
1:07:58 Lyme
1:07:59 disease our lovely uh insurance industry
1:08:03 um basically wrote the legislation
1:08:07 on the treatment for Lyme disease and
1:08:10 any doctors that go outside of the six
1:08:12 weeks of doxy cycling for Lyme disease
1:08:16 uh get sued to have their medical
1:08:18 licenses removed so no doctor will touch
1:08:21 it so there's this whole underground
1:08:24 medical community nothing is covered by
1:08:26 insurance but if you want your kids to
1:08:29 say you know not [ __ ] suffer um you
1:08:32 have to go to this underground so the
1:08:34 fact that it asked for that good sign
1:08:36 are you interested in specific types of
1:08:38 studies uh let's see uh yes both
1:08:41 mainstream and alternative all
1:08:44 [Music]
1:08:46 studies um
1:08:48 studies are fine I want a comprehensive
1:08:57 um
1:08:59 report we'll go two time frame should I
1:09:03 focus on recent advancements last 5 10
1:09:05 years uh
1:09:09 last five to 10 years is fine
1:09:16 since they will
1:09:20 likely site other
1:09:24 research okay
1:09:28 three do you want to focus on
1:09:30 conventional
1:09:32 medicine um three
1:09:35 both and then
1:09:37 four clinical trials should I prioritize
1:09:43 us um let's
1:09:47 see
1:09:49 um
1:09:51 prioritize
1:09:53 us and
1:10:00 um no
1:10:03 inclusion
1:10:05 criteria but I want you to
1:10:10 clearly
1:10:12 summarize the
1:10:14 inclusion
1:10:17 criteria C wait cry tier e and even
1:10:23 categorize
1:10:26 the
1:10:26 trials into logical
1:10:32 categories so like this
1:10:37 is this is pretty remarkable right that
1:10:41 that you can like this is what you would
1:10:43 ask if you hired someone to go do
1:10:45 research for you on
1:10:48 stuff let's see wait does YouTube not
1:10:51 have hearty Hearts I'm still a Tik Tok
1:10:54 refugee it's got thumbs up an you can do
1:10:58 thumbs up on tick on YouTube and we now
1:11:00 have you can do super stickers or I
1:11:03 don't Brandon knows I don't know what
1:11:06 they're
1:11:08 called this is what the doctor did with
1:11:11 with my report oh that's awesome okay
1:11:13 cool okay so here we
1:11:17 go boom
1:11:28 great I'll conduct a deep dive research
1:11:29 in the past 5 10 years the report will
1:11:32 be divided into these sections research
1:11:34 on the organisms covering their biology
1:11:37 mechanisms and impact on human health
1:11:39 treatments for related conditions
1:11:41 mainstream and alternative approaches
1:11:43 for inflammation neuropathy joint pain
1:11:45 and autoimmune
1:11:46 diseases promising AI treatments and
1:11:50 advancements clinical trials as I gather
1:11:52 information
1:11:55 okay so starting
1:11:58 research and then let me do so once it
1:12:03 starts
1:12:04 here it's it's going to pop up a little
1:12:07 research thing on the right that's where
1:12:08 we'll see its Chain of Thought reasoning
1:12:30 oh Danielle suggested a nod to effective
1:12:32 pain management in the most recent two
1:12:34 years of research
1:12:38 okay so there's the thing why am I not
1:12:41 seeing the little thing over on the left
1:12:43 or on the
1:12:46 right maybe I have to do
1:12:49 this
1:12:53 huh understanding babiosis
1:12:57 got
1:12:59 it can I click on
1:13:02 this yeah there we go
1:13:05 okay so here's the activity so this is
1:13:09 pretty slick let
1:13:11 me let me how how do I want to do this
1:13:14 let me just move the
1:13:16 window okay so for you ticktockers so
1:13:20 here's the
1:13:21 activity so I'm piecing together
1:13:23 Insights search for bellia Borg Deery
1:13:27 immune evasion review yeah these little
1:13:31 [ __ ] they they go into these things
1:13:33 called bofilm they migrate out of your
1:13:35 bloodstream into your muscles and they
1:13:37 hide in this [ __ ] called bofilm that you
1:13:40 can't get to them so you test negative
1:13:42 for them but they're sitting in there
1:13:44 like little
1:13:45 [ __ ] and then um when your immune
1:13:49 system drops they [ __ ] go hey party
1:13:56 [ __ ]
1:13:58 um so it's doing all that stuff look at
1:14:00 all that stuff look at all the work it's
1:14:02 doing it's found 15 sources so far so
1:14:05 here's the
1:14:06 sources the NIH the
1:14:10 mdpi frontier science hooks Hopkins
1:14:14 medicine neurology Hopkins lime Hopkins
1:14:18 is one of the big research centers
1:14:23 online we'll just let the activity keep
1:14:26 keep
1:14:28 rolling so
1:14:31 anyway so this is what it does pretty
1:14:34 pretty [ __ ] slick man it really
1:14:44 is $2 super sticker cool I guess I'm
1:14:51 imagining that commercial with
1:14:55 the Derma fights green slimy boogers and
1:14:58 party hats it's it's very much like
1:15:00 that and they and they have oh thank you
1:15:04 very much digital Gods I didn't realize
1:15:05 you sent me a super sticker I thought
1:15:07 you were just telling people how much
1:15:08 they were thank you sir very much
1:15:10 appreciate the
1:15:12 support that's very
1:15:14 cool you just shared the article in a
1:15:16 regulars okay great yeah so for for
1:15:18 Danielle's condition that that her
1:15:21 doctor researched she said he basically
1:15:23 did what I did the one thing they added
1:15:25 was pain management which is that's
1:15:27 probably a smart thing to
1:15:32 do oh that's pretty cool look at look at
1:15:35 this one comment I I love being able to
1:15:38 see the thinking of this Chain of
1:15:39 Thought reasoning is pretty
1:15:41 cool chat gbt look it says it read withp
1:15:46 power.com and then the The Prompt that
1:15:49 it gave itself I'm curious about the
1:15:51 with power s's mention of a trial in
1:15:55 Maryland testing an antibiotic for
1:15:57 inflammation reduction for long-term
1:15:59 Lyme patients I'm looking into specific
1:16:02 details for that trial name then it went
1:16:05 back to the
1:16:13 site so it found one of the lime trials
1:16:16 the
1:16:17 nct4 38 346 Lim trial at trial x.com
1:16:22 like this is stuff I mean you know this
1:16:25 is stuff I've been dealing with for a
1:16:26 while but like if you want to research
1:16:28 this like these are the rabbit holes you
1:16:29 run
1:16:30 down and it's it takes weeks
1:16:35 months and this is just doing it we're
1:16:38 at 28
1:16:40 sources oh it wasn't your doctor I was
1:16:43 with a doctor with over 300,000
1:16:46 followers on X who has open AI Early
1:16:49 Access that's super cool
1:16:52 Danielle that's very very cool
1:16:58 that's incredible it's incredible I mean
1:17:02 and again
1:17:04 like I you know I've done this is now my
1:17:08 third search using deep research I did
1:17:13 um deep research
1:17:15 with was it R1 had it I don't know
1:17:18 something else had it and that was kind
1:17:20 of cool like what it did was
1:17:23 cool um
1:17:26 I still haven't really played with
1:17:28 operator to the point that I understand
1:17:30 what it's really good
1:17:33 at that's why I was so excited and
1:17:35 tagged you that's awesome yeah thanks
1:17:38 sorry I've been a little distracted good
1:17:40 Tik Tock question can deep research
1:17:42 access all scientific papers because not
1:17:44 all are available for free right I
1:17:47 honestly don't know I am um I
1:17:51 assume that most of the research papers
1:17:54 that are behind pay walls stay behind
1:17:56 pay walls
1:17:59 um but I don't know I don't know how it
1:18:02 works I don't I I mean I assume it's
1:18:05 honoring those pay
1:18:09 walls
1:18:11 I'm
1:18:13 97% confident that the original llms
1:18:18 from all of the companies were trained
1:18:21 on data that was scraped from behind pay
1:18:24 walls
1:18:25 there's two ways to do pay
1:18:27 walls the one way is you actually
1:18:30 encrypt the [ __ ] and you can't you
1:18:33 literally can't access the stuff unless
1:18:36 you have the key triggered by your pay
1:18:38 wall the other way to do a pay wall is
1:18:40 you basically just the shit's all
1:18:43 sitting there in
1:18:44 folders and you have a pay wall in front
1:18:46 of it that just doesn't give you the
1:18:48 hyperlinks for all of those pay walls
1:18:51 that weren't doing the encrypted the
1:18:54 fancy version pay walls uh I'm pretty
1:18:57 sure open AI scraped all that [ __ ] and
1:18:59 it probably scraped the other [ __ ] too
1:19:03 um I I why would you not if if you're if
1:19:06 you're out trying to get as much
1:19:08 knowledge as you can I'm sure they did
1:19:10 all sorts of interarc things as did
1:19:12 everyone else I don't think they're just
1:19:13 particularly evil Kyle can you give me
1:19:15 the
1:19:17 tldr on on deep research I used it today
1:19:20 by accident yeah so so what it is an is
1:19:24 it it's um it's going to be really good
1:19:28 for your for your fundraising stuff it's
1:19:30 going to be really good for it so I was
1:19:33 describing it earlier it's a combination
1:19:35 of well that's doing that let me just
1:19:36 I'll just pop my my big fat face up
1:19:39 there again let me flip this over here
1:19:46 um you've got your
1:19:48 traditional like chat GPT query you ask
1:19:51 chat GPT it gives you an answer then
1:19:54 you've got reasoning model models where
1:19:55 you give it a prompt and it prompts
1:19:57 itself it does Chain of Thought
1:19:59 reasoning where it prompts the large
1:20:02 language model and it responds to it and
1:20:04 it it basically makes a plan right
1:20:07 before it gives you an
1:20:09 answer deep research takes it one step
1:20:12 further and goes out and searches the
1:20:14 web summarizes the websites it reads and
1:20:18 and is dynamically writing based on what
1:20:21 it learns in article one or website one
1:20:25 it changes its um Chain of Thought
1:20:29 reasoning so it's basically as it's
1:20:31 learning it's repr prompting itself and
1:20:34 and sort of refiguring out what to do
1:20:36 it's out doing what researchers
1:20:39 do right let me go look at this website
1:20:41 oh like the thing it said where it was
1:20:43 like oh it mentioned something about
1:20:46 some clinical trial but I didn't get the
1:20:48 name of that so let me go back and look
1:20:50 at that like it's comprehending what
1:20:53 it's reading oh we've got a report let's
1:20:55 go
1:21:00 back
1:21:05 [Music]
1:21:10 oh good lord look at our little
1:21:13 report
1:21:15 damn dag boy
1:21:24 [Laughter]
1:21:29 Long Haul lime disease a comprehensive
1:21:31 literature
1:21:33 review research on the
1:21:36 organisms bellia Borg
1:21:39 Deery lime disease
1:21:42 Spyro you ever seen little Lyme disease
1:21:45 Spyro they're like little cork screws
1:21:47 little [ __ ] swim through your
1:21:48 bloodstream and then they can cork screw
1:21:50 their way out of your bloodstream
1:21:56 a spiral-shaped bacterium transmitted by
1:22:00 ticks
1:22:03 um tick born
1:22:05 coinfections babesia Barella Etc same
1:22:09 tick that carries lime can also transmit
1:22:11 other pathogens called which often
1:22:14 co-infect patients and complicate the
1:22:16 illness notable infections babesia M
1:22:22 Micra malaria like protozoan bonella
1:22:26 Hensley
1:22:29 babesia it causes
1:22:34 biosis so I know all this stuff this
1:22:37 stuff I all know related conditions
1:22:39 persistent
1:22:44 inflammation neurologic issues and
1:22:47 neuropathy that's what my guys are
1:22:49 dealing
1:22:50 with joint pain and Lyme arthritis
1:22:57 autoimmune
1:22:59 sequi beyond the joints there's a
1:23:01 concern that lme and Co infections May
1:23:03 precipitate other autoimmune immune
1:23:17 phenomenon the main issue is once you
1:23:19 believe they've gotten rid of all these
1:23:21 pathogens then your body still interacts
1:23:23 with the remnant causing autoimmune
1:23:26 diseases yeah I know it's it's a [ __ ]
1:23:28 nightmare that needs to go in Notebook
1:23:30 LM oh that's a great idea let's let's
1:23:32 definitely do that oh this is so cool
1:23:36 dudes
1:23:37 okay
1:23:39 copy go copy this big
1:23:45 [ __ ]
1:23:48 boom go to notebook
1:23:51 LM oh no wait I got to go to docs first
1:23:54 let me turn turn this into a Google
1:24:02 [Music]
1:24:14 doc have you noticed by interacting with
1:24:17 AI that you're processing information
1:24:20 similar I've noticed that working with
1:24:23 AI
1:24:25 I
1:24:26 am it looks like I crashed my Google doc
1:24:30 here did
1:24:31 I why can I not
1:24:35 paste I've noticed that working with AI
1:24:38 I've
1:24:39 got way
1:24:44 more I'm more creative I have more ideas
1:24:49 I
1:24:51 have a more active intellect it's it's
1:24:55 it's the opposite of what you think it
1:24:56 will do it's
1:24:58 bizarre because you
1:25:00 think I think I I
1:25:03 thought
1:25:06 um here we're not going to do this
1:25:15 here
1:25:17 um I thought if if that stuff's doing
1:25:20 everything for
1:25:23 you that you you'll be less creative
1:25:25 you'll be less intellectual you'll
1:25:29 be all that stuff
1:25:35 wait
1:25:40 why it did not copy everything what the
1:25:49 [ __ ] oh I see what it it okay
1:25:56 wait can I
1:26:01 is they've got some weird ass formatting
1:26:04 in their
1:26:06 um in their documents is there a way
1:26:09 that I can hey Kyle yeah there's a copy
1:26:13 button at the bottom that you might want
1:26:15 to tap at the end of the at the end of
1:26:18 the research paper is
1:26:22 there oh yeah look at that
1:26:25 that's producer Brandon coming in for
1:26:27 the win it it doesn't mean it's going to
1:26:29 work it's it's still going to probably
1:26:31 copy the weird ass uh formatting but
1:26:34 let's go
1:26:36 see I've also got a Tik Tock pin for you
1:26:39 um okay did it
1:26:47 work oh yeah it worked all right good
1:26:51 job thank
1:26:52 you so what Brandon said said each of
1:26:55 these trials
1:26:57 assistance alternative therapy
1:27:00 immunotherapy okay all right so there's
1:27:03 a lot of [ __ ] in here good
1:27:06 Lord with all these all of these look at
1:27:09 all the citations all that yellow [ __ ]
1:27:11 are citations [ __ ] amazing all right
1:27:15 so now I'm gonna export this as a
1:27:19 PDF so this is going to be
1:27:23 um oh wait hang on let me get rid
1:27:27 of all that
1:27:31 [ __ ]
1:27:32 okay export as
1:27:36 PDF L haul Lyme disease we're going to
1:27:39 drop that on the desktop
1:27:43 boom did the AI just find all the work
1:27:47 that someone else did that answers your
1:27:52 prompt well it I asked it to go out and
1:27:55 find all of the research like all of all
1:27:59 of the research papers
1:28:02 on
1:28:05 no is the AI cheating off a human no no
1:28:08 no no empty bubbles no this this report
1:28:12 that it wrote it wrote it went out and
1:28:15 it read all of the papers like whatever
1:28:18 it went and found it went and read all
1:28:20 those websites it read the papers it
1:28:22 read the research papers and then it
1:28:25 wrote this
1:28:26 report this is what you would have a
1:28:28 researcher do if you wanted a researcher
1:28:31 to give you an update on what's going on
1:28:33 with Lyme disease you would pay a
1:28:35 researcher to probably spend I don't
1:28:38 know two three months six months going
1:28:41 and reading all this stuff categorizing
1:28:43 stuff putting it together writing it
1:28:45 into a thing it wrote this from all the
1:28:49 research it did so so just think of this
1:28:52 as this is consolidating you know two
1:28:55 months three months six
1:28:57 months oh I'm being a little bit
1:28:59 humorous okay good yeah yeah yeah I
1:29:01 figured you I figured you were yeah
1:29:04 it's yeah I listen
1:29:08 I we're in a weird we're in a weird ass
1:29:12 place where people are going to start
1:29:15 getting freaked out like if you're say a
1:29:17 researcher that spends six months making
1:29:19 these things and you watch what I just
1:29:22 did you be like
1:29:25 um okay we're going to go to notebook
1:29:27 LM do uh. goole.com
1:29:31 if you haven't seen this by the
1:29:36 way oh I didn't say notebook I said
1:29:40 note good Lord book LM all
1:29:45 right we're going to create a new
1:29:48 one we're going to upload our source
1:29:50 document
1:29:57 which we can't find of
1:29:59 course um
1:30:03 lime Long Haul lime disease let's make
1:30:06 sure it's our thing yep there it is how
1:30:08 many pages is
1:30:12 it I don't know
1:30:16 lot
1:30:18 okay so there's
1:30:21 that boom
1:30:25 Untitled
1:30:30 notebook Long Haul lime disease the
1:30:33 literature
1:30:36 examines Long Haul line focusing on
1:30:38 persistent mechanisms of lime
1:30:44 disease all right so we're going to go
1:30:46 to the
1:30:48 studio and we're just going to say
1:30:52 generate so what I'm going to do is it
1:30:55 generating yeah so I'm going to generate
1:30:58 an audio
1:31:00 overview it's basically doing a
1:31:02 literature review those are very time
1:31:04 consum very time intensive
1:31:07 reports yeah as someone with
1:31:10 ADD I couldn't read the first four
1:31:13 paragraphs of one study much less go
1:31:16 read all of
1:31:18 them it's like it's just unbelievable
1:31:22 the world we live in right now it really
1:31:24 is
1:31:25 is it's pretty pretty remarkable
1:31:29 stuff okay
1:31:32 where
1:31:34 um oh there it is
1:31:39 okay Becky Ru my youngest has lyme
1:31:41 disease he's 21 well that [ __ ] sucks
1:31:44 I'm sorry Becky yeah my guys are
1:31:47 25 yeah my guys have had it for I don't
1:31:51 know 12 years now sucks it sucks sucks
1:31:59 bad what do you think about attacks on
1:32:02 llms creating nukes destroying Humanity
1:32:05 what about llm
1:32:07 security that's a lot of attacks on llms
1:32:11 do you mean jailbreaking them I mean
1:32:14 you've
1:32:15 got I don't you don't really need to
1:32:17 jailbreak them you've got open-source
1:32:19 models like llama and deep seek where
1:32:23 you can now go do whatever you want to
1:32:25 do with them you don't you can you can
1:32:28 train around their safety guard rails so
1:32:35 um what I think about it is you can do
1:32:39 the same [ __ ] with existing technology
1:32:42 like any technology can use be used for
1:32:45 good or for
1:32:48 [ __ ] um and you know this channel is
1:32:53 about why don't we try using it for for
1:32:57 good you know you know like you
1:33:00 do we're not we're not making dirty
1:33:03 bombs here on this
1:33:07 channel um let's see
1:33:11 um what are
1:33:15 the most
1:33:17 exciting
1:33:20 new studies that point to an actual
1:33:31 [Music]
1:33:35 cure we have issues and problems that we
1:33:38 keep pointing the finger at each other
1:33:40 in any type of technology that we don't
1:33:43 understand I
1:33:48 agree
1:33:50 okay what's it
1:33:52 say several new
1:33:55 studies persist cell treatments a three
1:33:59 dug three drug
1:34:02 cocktail of dapto myosin doxy cycline
1:34:06 and whatever I think that's from
1:34:09 um horor wits I think that's horowitz's
1:34:16 thing D suaram traditionally used for
1:34:22 alcoholism huh
1:34:29 AI accelerated drug Discovery ah here's
1:34:32 our
1:34:34 thing hang on let me change the thing
1:34:37 let's listen to a little of
1:34:42 [Music]
1:34:44 this is this it yeah that's it
1:34:48 yeah lme disease and co- infection oh
1:34:51 yeah topic that's uh more incate than a
1:34:54 lot of people realize it is you sent
1:34:56 over some really fascinating research
1:34:58 for this deep dive a comprehensive
1:35:01 literature review on Lime the co co-
1:35:04 infections and their long-term effects
1:35:06 right so let's dive in yeah let's break
1:35:08 it all down okay so the research really
1:35:10 digs into the bacteria behind Lyme
1:35:12 disease yeah brilia bergd door
1:35:15 fery what makes this
1:35:18 bacteria and modulates the nervous
1:35:20 system which can have a positive exactly
1:35:22 if hyy a proves effective in human
1:35:25 trials it could be a real breakthrough
1:35:27 in Lime treatment that's great to hear
1:35:30 and there are also advancements being
1:35:31 made in treating lime Co infections
1:35:33 that's right what are some of the
1:35:35 promising developments for babesia and
1:35:37 bonella for babia scientists are
1:35:40 exploring new antiparasitic drugs okay
1:35:43 and even using computational methods to
1:35:45 identify potential compounds from
1:35:47 natural sources computational methods
1:35:49 that sounds Hightech it is artificial
1:35:51 intelligence is playing a growing role
1:35:54 in drug Discovery wow AI can analyze
1:35:57 massive amounts of data to identify
1:35:59 potential drug candidates much faster
1:36:02 than humans can so AI is speeding up the
1:36:04 process of finding new treatments
1:36:05 exactly and this is crucial for Lyme and
1:36:07 other complex diseases where we
1:36:09 desperately need new and effective
1:36:11 therapies right and for Bartonella
1:36:13 researchers are finding that some
1:36:14 existing drugs not what if you have
1:36:17 certain research papers as PDFs in one
1:36:19 folder how do you ask chat GPT to review
1:36:22 them well you can if you've got a bunch
1:36:24 I am if you've got a bunch of PDFs the
1:36:27 the tool that I'm in right now is
1:36:30 notebook LM you can upload I think it's
1:36:34 50 documents and they can be up to
1:36:38 500,000 words each so it's like a
1:36:41 ridiculous amount of data you can put in
1:36:43 here it's like 25 million
1:36:46 words and then you put the the sources
1:36:49 in over here you can chat with them in
1:36:52 the middle here and then you can turn
1:36:54 them into documents over on the right so
1:36:57 you you can create study guides a
1:36:58 briefing document a timeline um and this
1:37:02 audio
1:37:03 overview and then check this
1:37:07 out Welcome to our Deep dive today we're
1:37:09 going to be tackling Lyme disease and
1:37:11 co-infection I'm going into into
1:37:13 interactive
1:37:20 mode Welcome to our Deep dive today
1:37:22 we're going to be tackling l mind
1:37:24 disease and co-infections oh yeah a
1:37:26 topic that's uh more intricate than a
1:37:30 lot of people realize it is you sent
1:37:32 over some really fascinating research
1:37:33 for this deep dive a comprehensive
1:37:36 literature
1:37:38 review oh hey there how can we help yeah
1:37:41 hey I um I'm the one that initiated this
1:37:45 research study and I'm doing it because
1:37:47 I've got twin 25y old boys who've been
1:37:50 dealing with Lyme disease for about 12
1:37:51 years now
1:37:56 wow that's um that's really something
1:37:59 yeah thank you for sharing that with us
1:38:00 he gives us a lot of perspective knowing
1:38:02 the personal Drive behind this research
1:38:04 absolutely and it's important to keep in
1:38:06 mind the real life impacts of this
1:38:08 disease we really appreciate your
1:38:10 insights and well we'll definitely keep
1:38:11 your sons in mind as we go through this
1:38:13 yes oh back to our listener what's up
1:38:16 yeah so they've got one of the one of
1:38:17 the primary things they're struggling
1:38:19 with right now is myin sheath neuropathy
1:38:22 so is there anything in the
1:38:24 study that points to uh ways to deal
1:38:27 with
1:38:32 that okay so milin sheath neuropathy
1:38:35 that's definitely a tough one and it's
1:38:36 something that the research touches on
1:38:38 though not always by that specific name
1:38:40 right and what's fascinating here is how
1:38:42 lime can affect the nervous system in
1:38:44 various ways yeah it seems that the
1:38:45 research talks about how lime can cause
1:38:47 damage to the nerves specifically small
1:38:50 fiber neuropathy yeah small fiber
1:38:52 neuropathy or sfn is characterized by
1:38:54 damage to the thin nerve fibers that
1:38:56 transmit pain and autonomic signals and
1:38:58 from what I was reading this can cause a
1:39:00 range of symptoms exactly it can lead to
1:39:02 things like widespread pain Burning
1:39:04 Sensations and disautonomia like
1:39:07 lightheadedness or heart rate issues so
1:39:09 while the research doesn't use the term
1:39:11 milin sheath neuropathy all right so
1:39:13 anyway so that's that if you haven't
1:39:15 seen that
1:39:16 like as cute as the podcast is it
1:39:21 actually does really good analysis of
1:39:23 the do doents you put in there and the
1:39:25 fact that you can jump into it and ask
1:39:27 it questions and have them explain it to
1:39:29 you like this is you know we're in a
1:39:33 whole other thing all right listen I
1:39:34 gotta get going it's it's late here it's
1:39:36 late where you are um I'm Crispy and
1:39:40 dehydrated um so um yeah so Becky Becky
1:39:45 didn't know my kid said Lyme disease
1:39:47 yeah I know it sucks and I'm sorry that
1:39:48 yours does too like this is this is a
1:39:51 club none of us want to be in it's like
1:39:54 you get in this club you're like
1:39:56 we I've got
1:40:00 community um
1:40:02 okay couple of things um You probably
1:40:06 have all heard of the Webby Awards the
1:40:09 Webby awards are
1:40:12 um you
1:40:14 know awards for websites and [ __ ] like
1:40:17 that right they they came out during the
1:40:20 hype cycle of the dot bubble
1:40:25 everyone is like you know the we the
1:40:28 web's exciting Let's do an award show um
1:40:31 they just added an AI category so I
1:40:34 think they extended the deadline of the
1:40:36 Webb's to Friday that's what Brandon
1:40:39 told me if it's wrong blame him
1:40:44 um but like Joy pie Corey I mean you
1:40:50 could maybe I'll maybe I'll submit Sin
1:40:53 City sipper no one will care but it's
1:40:57 funny um I'm going to do Quantum
1:41:03 Cipher I'm GNA do the one I think I'll
1:41:05 do the one with uh with
1:41:09 uh Kelly Bosch unit maybe I'll do the
1:41:12 Civil War project I did I don't know but
1:41:15 but anyway go do the go do the
1:41:17 wees um
1:41:20 okay if you don't know go to
1:41:27 [Music]
1:41:30 wait go
1:41:33 to oh
1:41:35 please not banners media assets go to
1:41:39 the salon.
1:41:44 a see that
1:41:46 ticktockers um if you haven't joined the
1:41:48 AI Salon go to the salon. click on the
1:41:51 join our community button uh um and and
1:41:55 and go join it's a community one of the
1:41:57 things that's going to be important for
1:41:59 everyone in the next two years three
1:42:02 years as this stuff gets more and more
1:42:05 and more intense 2025 mark my words 2025
1:42:09 is going to be one of those years where
1:42:15 um A lot's going to change and this
1:42:20 technolog is going to get really good
1:42:22 really fast like it it already is like
1:42:24 what we did tonight is kind of [ __ ]
1:42:28 Bonkers what it just
1:42:30 did and that was like that was without
1:42:33 even really trying right that was just
1:42:34 like me sort of pulling a prompt out of
1:42:36 my butt and just kind of asking it for
1:42:38 what I thought would be good imagine if
1:42:41 you actually put you know like a week
1:42:42 into using that
1:42:45 tool
1:42:48 um a lot of shit's going to change um
1:42:51 there's going to be a lot of disruption
1:42:52 so get your ass in community the AI
1:42:56 Salon is a community of AI optimists and
1:42:58 creators who are generous and um helpful
1:43:04 and smart and
1:43:07 accomplished and
1:43:09 humble something you don't get
1:43:13 [Laughter]
1:43:17 here in spite of me there're a really
1:43:21 humble bunch there's not a bunch there's
1:43:22 not a lot of ass holes in there they're
1:43:25 like sweet empathetic supportive human
1:43:29 beings that are trying to figure this AI
1:43:31 [ __ ] out too and they happen to be some
1:43:34 of the people doing the most interesting
1:43:36 work not building the AI we got a couple
1:43:39 of those we got like Pate who's building
1:43:40 this
1:43:42 [ __ ] but but most of the com the the
1:43:45 community of the AI Salon it's like
1:43:47 3,000 people are people that are there
1:43:50 to just try to figure out how do we use
1:43:51 these tools how do we use them in our
1:43:53 Liv how do we use them to help do the
1:43:55 [ __ ] we do so go do that tomorrow
1:43:59 night is the first Tuesday of every
1:44:01 month we have a meeting called AI Salon
1:44:05 presents and tomorrow night we're going
1:44:07 to have a return guest we had an artist
1:44:10 named black
1:44:13 Blac um who's an nft artist he was a he
1:44:16 was a consultant for many years business
1:44:18 consultant he was really good at Art and
1:44:21 he got way into digital art got way into
1:44:23 nfts all that sort of stuff and now he's
1:44:26 coming back around he's he's done all
1:44:28 that and he's going to talk about coming
1:44:30 back around and now he's starting an AI
1:44:32 consultancy which I think is absolutely
1:44:35 fascinating so um so that's tomorrow
1:44:37 night from 5:00 pm to 7 PM mountain time
1:44:42 um Brandon and Eric Sanner uh are going
1:44:46 to be co-hosting the event Leah can't
1:44:49 make it and I might be able to make it
1:44:51 but I don't think I can because I'm
1:44:53 doing business [ __ ] um so so we're going
1:44:56 to have guest co-hosts cool and we're
1:44:58 going to have black speaking and there's
1:45:00 just so much news has happened you
1:45:02 should go to that thing and you should
1:45:04 meet the people there and join the salon
1:45:08 and be cool all right that's it I'm
1:45:11 gonna get out of here um I've got a cool
1:45:15 I think I'm gonna take that um that
1:45:18 podcast and publish it um I'll I'll put
1:45:21 that up on LinkedIn and I'll put it up
1:45:23 on
1:45:25 uh Twitter X you know the thing all
1:45:30 right everybody
1:45:33 Coolio peace out I will be back here
1:45:36 tomorrow night um yeah it's I'll be back
1:45:41 here I'll probably be a guest on the
1:45:44 salon presents and and then I'll do this
1:45:47 probably normal time 8:00 p.m. mountain
1:45:50 but I'll be it'll be 10 p.m. East Coast
1:45:55 why do I do this to people why do I
1:45:57 start these things at 10 o'clock at
1:45:58 night for East Coasters I'm an [ __ ]
1:46:01 that's why okay
1:46:04 everybody has deep seek put every other
1:46:06 competitor out of business no Snowman no
1:46:10 no no the Deep seek thing was completely
1:46:14 blown out of proportion is it a good
1:46:15 model yes did they train it for $6
1:46:18 million no it looks like it was closer
1:46:20 to 1.5 billion it looks like
1:46:24 um deep seek has about 50,000 h100s
1:46:27 which they're not supposed to
1:46:30 have so that was that was some uh how
1:46:34 you call it in America a little bit of
1:46:36 misinformation to drive the stock market
1:46:39 crazy no um quen came out with a better
1:46:42 model three days later um open AI just
1:46:45 came out with this deep research thing
1:46:47 and and 03 mini and 03 mini high and no
1:46:51 it's that that is a small blip in a much
1:46:54 larger War um it's like that I don't I
1:46:56 wouldn't even consider that a battle did
1:46:58 it did it drain whatever it was 1.2
1:47:03 trillion out of the markets in a day
1:47:05 yeah it did do that is that going to be
1:47:08 a long-term impact I don't think so but
1:47:09 we'll see who knows I don't [ __ ] know
1:47:11 anyway all right grock 3 will be very
1:47:14 late
1:47:16 yep um all right peace out everyone have
1:47:19 a good one I'm going to say goodbye to
1:47:21 the YouTubers and xers peace out
1:47:24 everyone hope you had fun tonight