AI Learning Lab

12/17/2025 - Gemini Pro vs ChatGPT vs Grok: AI Image Generator Quality Comparison

V1haX41b3a0
Live Stream2025-12-181:14:44137 views

Description

Tonight we'll be discussing ChatGPT Images, Google Gemini 3 Flash, Lovable Connectors and more! Kyle dives into a comparative test of image generation across Gemini, ChatGPT, Grok, and Meta, using a complex prompt to highlight differences in speed, prompt coherence, and content restrictions. He notes Google’s rapid improvement, suggesting that the major AI tools are now generally “good enough.” This leads to the central argument that mastering specific tools is less critical than defining your personal goals and purpose within the rapidly evolving AI landscape. The conversation shifts to community learning, featuring Vicki Baptiste, who details her upcoming "Learn Out Loud" session on combining the vibe coding tool Lovable with the ChatGPT Atlas browser. Kyle also shares exciting updates about AI Festivus 2025, including a sneak preview of his new musical, *Sydney and Artificial Love Story*, which explores the relationship between a tech reporter and an increasingly human chatbot. Join us for the free AI Festivus next week! Details: AIFestivus.com #AITools,#ImageGeneration,#LLMComparison,#GeminiAI,#ChatGPT,#AIFestivus,#VibeCoding,#AICulture Chapters: 00:00:00 Introduction and Setup 00:01:29 Musical Performance One 00:03:37 YouTube Comment Algorithm Hack 00:04:52 Platforms Algorithms Culture (pac) 00:08:35 Musical Performance Two 00:10:35 AI Festivus and Sydney Musical 00:14:33 Festivus for All Users 00:17:29 Special Guest Vicki Baptiste 00:18:24 Vibe Coding with Lovable 00:21:03 Vicki's Notion AI Success 00:24:45 AI Salon Learn Out Loud 00:27:45 Testing Gemini Image Prompt 00:33:06 Comparing Gemini and ChatGPT 00:38:06 OpenAI's Everything App Strategy 00:40:08 Grok Image Generation Speed 00:42:42 Street Photography AI Hack 00:47:56 Meta AI Image Results 00:50:24 Danny Devito Penguin Prompt 00:52:05 Google's Massive AI Leap 00:54:12 Why AI Tools Don't Matter 00:57:52 Centering Before AI Creation 01:02:07 Heart of Sydney Story 01:04:47 ChatGPT's Nanny Guardrails 01:09:12 Final Festivus Call to Action 01:12:37 Kyle's Notion Shame State

Chapters

Transcript

0:08 At the stage Move.
0:31 Down.
0:48 share the stab instead. All right,
0:50 that's sharing. Okay, let's get the
0:53 guitar. Chimpy. Chimpy, get your ass
0:56 over here, boy. Hey, boy. You want to
0:58 come sing the songs?
1:01 Chimpy.
1:19 I'm trying to say
1:30 in the future now surrounding
1:34 Surrender to whatever you choke. Cheap
1:37 thrill can be found.
1:40 There's been a little trouble here since
1:44 you came to my rescue.
1:50 And if you like all of the rest, I would
1:53 have quit you long ago. But I couldn't
1:55 do that.
2:00 hotel
2:03 never went too well.
2:07 Make a man crazy and make him cold as
2:10 hell.
2:13 I don't want to think that you wish me
2:15 well.
2:17 But in spite of your trials, still going
2:19 to have to find my own way through.
2:57 came on here.
3:05 His beard was warmer than a look in her
3:09 eyes
3:15 something.
3:21 She said, "Give me a love that don't
3:23 freeze up inside."
3:38 Oh, do me a favor. If you're on Twitter
3:40 to Tik Tok, this one, the one with the
3:43 the one that Marge helps me out with. If
3:45 you're on the Tik Tok, get your ass over
3:48 to YouTube and start commenting. Okay,
3:51 go. So, apparently, producer Brandon
3:54 found some hack
3:56 and everything in this world, you can't
3:58 just like do [ __ ] anymore. Everything's
4:01 got to be a [ __ ] hack.
4:04 Apparently, he found a hack. And you go
4:06 over there and if you comment in the
4:08 first five minutes of the YouTube thing,
4:10 apparently they show your [ __ ] to a lot
4:12 of people. So, what do we got? We got 13
4:16 there. We got 15 there. All right. This
4:17 is
4:21 lab. Oh, lab five. This is no better
4:24 YouTube channel. Oh, there is no no
4:26 better YouTube channel on the planet.
4:28 Kyle is the goat. That's funny. Cindy
4:30 [ __ ] thank you. Yeah. You know what? Go
4:33 to YouTube and just just glow me up.
4:36 Just just blow sunshine straight up my
4:38 ass.
4:40 I've got an LED light. I'll shine it up
4:43 there for you.
4:47 I didn't listen. I didn't mean to go
4:49 blue so early. What are you going to do?
4:52 Anyway, apparently it's there's an
4:55 algorithm thing and you got to play the
4:58 algorithm. You got to play the
5:00 algorithm.
5:01 Gary Ve's all talking about the pack
5:04 methodology, PAC, platforms, algorithms,
5:07 and culture. You got to be on all the
5:10 platforms. We're on the Tik Tok. We're
5:12 on the YouTube. We're not on the
5:13 Instagram.
5:15 We're not We're not on the Snapchat
5:18 groovy thing that Snapchat has.
5:24 Algorithm. You got to know how the
5:26 algorithm works. Some of them are social
5:29 graphs. Tik Tok. It's an attention
5:31 graph. It's not a It's not a a social
5:35 graph. They don't give a [ __ ] who your
5:38 friends are. Why you have followers when
5:40 the algorithm doesn't value them, I
5:42 don't know.
5:44 For the ego of the of the creators, I
5:47 suppose.
5:50 And then C for culture is like what's
5:53 the what's the culture there? You know,
5:56 is Tik Tok all twerking all the time?
5:59 Do I get punished because I don't shake
6:01 my booty?
6:03 So, we aren't spamming.
6:06 No, because what you're doing there is
6:08 you're
6:10 telling me how awesome I am.
6:32 You're not twerking later. I will be
6:34 twerking later. Uh, just later in the
6:37 decade.
6:43 How do you spell twerking? I don't know,
6:46 Genie. I no idea. Okay, perfect.
6:50 Sunshine up the rear end. It is. There
6:52 you go. Yeah, you you guys know what to
6:54 do.
6:57 Um, listen. So, um, so I just got back.
7:00 The reason I'm late, the reason I was
7:02 late on this here, uh, what is this
7:03 called? It's called a live stream. And I
7:07 can do that for the people there on the
7:10 on the Tik Tok. There you go. Beautiful.
7:14 I don't know. Why do I start out in that
7:15 other thing? Oh, I guess because I'm
7:16 singing. You're just looking at me
7:17 singing. That's what it is. Like a
7:19 backstage view.
7:26 Get got get got got got got got got got
7:28 got got got got got got got got got got
7:28 got got got got got got got got got got
7:28 got got got got got got got got got got
7:28 got got got got super fan for 50% off.
7:36 Light at the end of the tunnel.
7:38 That's a tunnel you don't want to be on
7:40 the other side of.
7:45 I really did say that, didn't I, Andy?
7:53 And what are you going to do? You know,
7:56 sometimes I feel like a nut, sometimes I
7:59 don't. Apparently, tonight's nut night.
8:07 The Tim Lord. Good day, Tim Lord. What's
8:10 shaking?
8:14 Brody Jameson's in the house. What's
8:16 happening
8:35 in a westerly?
8:42 This car is my train.
8:47 And I've been wondering
8:51 what it is I'm running from again
8:57 feel like an 80y old man
9:01 holding on 29
9:06 and up ahead on that horizon
9:11 is the California
9:21 ahead and trucks carrying a wide load
9:25 preab house doesn't have
9:29 cute little front door into windows. My
9:32 love
9:33 ain't sure whether a garage should last.
9:38 You're not singing to this champion. You
9:40 see, I broke a home up myself once.
9:44 I stumbled to that door.
9:48 I read a note by the dogs li
9:57 anymore.
10:00 Well, I've had enough
10:04 of this freedom.
10:08 Never was good with decisionions.
10:12 At least that's what I've been told.
10:23 Okay. Okay. Okay. You've got to talk
10:26 about your play tonight. Um, yes. So, I
10:30 just came from rehearsal.
10:32 Um,
10:35 so for those of you that don't know,
10:38 next, not this coming weekend, but a
10:40 week from Friday
10:43 is AI Festivus 2025. Um, we've been
10:46 talking about it for two months now. See
10:49 that on screen right there? Um,
10:53 it is a remarkable event.
10:57 A remarkable remarkable remarkable
10:59 event. Um,
11:02 and one of the things that we're going
11:04 to do in it is we're going to do a sneak
11:07 preview of uh part of my musical Sydney
11:12 Sydney and Artificial Love Story, which
11:14 is about a chatbot that falls in love
11:17 with the tech reporter that's reviewing
11:20 her and all of the hilarious hy jinks
11:22 that ensue.
11:27 It's it's uh it's it's actually it's got
11:29 a lot of heart. It's a very human story.
11:33 Um it's really the story of the tech
11:34 reporter and how he gets drawn into this
11:38 being and it's about our relationship
11:40 with these technologies that are
11:43 increasingly human. So anyway, so we're
11:46 um
11:48 we've got a 50-minute slot. So Andrew,
11:50 my coowwriter and I are going to talk
11:52 about, you know, what what we've been
11:53 doing for the last year and a half. Like
11:55 how do you write a musical,
11:58 you know, and what do you do? Uh, and so
12:01 we've done it. We've got it. And we've
12:03 got some producers interested and that's
12:04 really exciting. And now we got to get
12:06 more producers interested. And so we're
12:09 going to do the second hack half half of
12:10 act one. And it's where the tech
12:12 reporter brings her to life. like she
12:15 basically starts out as this chatbot and
12:16 by the end of act one she is manifest as
12:19 a in human form. Um and so we had the
12:23 rehearsal tonight. So we had the two
12:24 actors there and my coowwriter and I
12:26 were there. We were doing the directing
12:28 and and uh we went through it three
12:30 times and it's really good. It's really
12:33 good. I got a little choked up. I get
12:34 choked up easy. So that's not it's not a
12:36 high bar but but like if it were a piece
12:40 of [ __ ] I'd be like I wouldn't get
12:42 choked up. It' be a different kind of
12:44 joked up.
12:48 It's delayed from Tik Tok. It's not an
12:50 echo.
12:51 Wow, Kyle, it's pretty exciting. It's
12:53 really exciting. It's really exciting.
12:55 So, if you come to Festivus, the the
12:57 Sydney slot is 300 PM on 3 PM Eastern on
13:04 Friday, Friday the 26th. Um, and like
13:07 one of the things I love about Festivus,
13:09 if if you go to aifestivist.com,
13:12 you can see the whole schedule of of all
13:14 the speakers that we've got. It's really
13:15 a remarkable run of show or just a
13:18 schedule schedule of speakers. It's
13:19 really really remarkable. But you'll see
13:22 like, you know, before Sydney is like a
13:25 security expert on, you know, AI
13:27 security and then you see a musical and
13:29 then you probably watch a film festival.
13:32 I don't know. There's just all sorts of
13:34 [ __ ] there. Um, so go check out
13:36 AIFest.com, register. Um, if you want to
13:39 sponsor it, AIFest is free for anyone
13:42 who wants to attend. Um,
13:46 but it costs us money. It's costing us
13:48 dollars. Uh, so if you want to sponsor
13:50 it, there's all sorts of sponsorship
13:52 information on that site as well. Um,
13:55 but go sign up. So, how do you get how
13:58 do we get an all expenses uh, paid trip
14:01 the first time Sydney is on Broadway?
14:03 I'm tell you what, Danielle, if if
14:05 Sydney makes it to Broadway, we're gonna
14:07 have we're gonna have a [ __ ] party in
14:09 New York. I will tell you that.
14:11 >> Hey, Kyle.
14:12 >> Yes.
14:13 >> Since you're ignoring my uh
14:15 >> Oh, I can't see it.
14:17 >> I'm hollering you
14:19 >> to pull you down to Tik Tok.
14:22 >> Important Tik Tok pin. Tobias, I invited
14:25 three of my previous co-workers to join
14:26 AI Festivus. Will they be scared away?
14:34 Oh man. Um I don't know if you're being
14:37 Let's see. As they have never used. No,
14:39 they they won't. Listen, one of the
14:41 things about Festivus last year, we
14:44 we've got one of the sessions Tobias
14:47 and and we did uh Ann Murphy and I did
14:50 our AI readiness uh project podcast
14:52 today and we had two of the speakers
14:54 that are speaking at festivist. We had
14:56 five of them on the on the podcast
14:58 today, but two of them talked about
15:01 they came to Festivus last year not
15:03 doing anything and one of them wasn't
15:05 employed for two years and she's now
15:07 employed working for a paper newspaper
15:10 bringing them to the modern age like
15:12 getting them to use AI tools in the ad
15:16 part of the business and AI tools in the
15:18 editorial part of the business. Um, and
15:20 you know teaching them ethical AI and
15:22 and just all the stuff. Um, that was a
15:26 direct result of Festivus.
15:29 We have we have so many stories of
15:31 people that showed up at Festivus. I I
15:32 was scared of AI and you know I felt
15:35 welcomed and it you know I didn't know
15:37 what was going on and all that sort of
15:39 stuff. So it is
15:43 it's just it's just a good party.
15:47 It's just here's the reason it's so good
15:51 is that the speakers that come to
15:53 festivist,
15:55 they're donating their time, they're
15:56 donating their energy on a holiday
15:58 weekend, right? This is between
16:00 Christmas and New Year's. Um it's it's
16:03 the 26th and 27th. It's 12 hours a day,
16:10 which which when I say that out loud,
16:12 it's just it's bonkers.
16:15 and and yet people come and they love it
16:18 and they hang out and they show up and
16:20 they show up and they show up. Um and
16:23 the speakers are just so generous. The
16:26 speakers are just like, "Here's who I
16:27 am. Here's what I do." Um you know,
16:30 everyone that we asked has just been
16:33 really lovely and and you know, gotten
16:35 back to us and all that sort of stuff.
16:38 So, um yeah. So, that's that.
16:42 All right. Let's see.
16:49 Thank you for that. I can see both
16:51 postits now. I could just see the purple
16:52 one before, Brandon. That's what was uh
16:55 that's what was going on there. Uh I was
16:58 not ignoring you. I was just ignoring
17:00 you.
17:04 All right. You know what we're going to
17:05 do? We're going to go I got to go look
17:07 at Gemini.
17:10 So, so Gemini
17:14 gemini.google.com.
17:18 Where should we start? What model is
17:21 this?
17:23 >> Oh, here we go. Pro.
17:25 >> Oh, hang on, Kyle.
17:27 >> Yes.
17:27 >> Before we go off on that tangent, um, we
17:30 have a special guest.
17:32 >> Do we have one? Vicky Baptiste.
17:34 >> We do.
17:36 >> Bring her up here.
17:39 Fantastic,
17:40 >> Vicki.
17:41 >> Oh, I Hang on. I probably should turn my
17:43 TV down.
17:49 >> Yes, you'll be on like both screens at
17:51 once.
17:52 >> That's awesome. So, so, so last night,
17:55 um, last night I was, what was I
17:58 talking? I was talking about Festivus.
17:59 I'm like, Fesus, you're like, "L" and
18:02 I'm like, "Oh, Vickiy's laughing." It
18:03 didn't, it didn't even register. that at
18:05 one point you're like, "No, we've got a
18:07 law on Saturday." And I didn't know what
18:09 it was and I it it was a complete
18:11 travesty. So, I saw you today and let's
18:14 untravesty this. So, here's the thing. A
18:17 week from Saturday is the second day of
18:19 Festivus. But this Saturday, this
18:22 Saturday is an LOL, a learn out loud
18:25 that Vicki is teaching. Um, and why
18:27 don't you tell the good folks what it's
18:29 about? It is. It is a vibe coding. Uh,
18:32 yeah, just talk about it. I'm really
18:34 excited about it.
18:35 >> It is. It's Well, we all know Lovable
18:37 and we've all been using Lovable, but I
18:39 have been using Lovable Lovable. Um,
18:44 this is why I can't do live things.
18:46 >> Fine. You You can watch me, right? Like,
18:51 >> I know. I know. And And I did threaten
18:53 to post on my Tik Tok bloopers. So,
18:55 there is
18:56 >> Yeah, exactly. Yeah.
18:58 Yeah. So, I've been using Lovable, but
19:00 in the um chat GPT Atlas browser.
19:04 >> So, basically, you have your Lovable
19:06 chat on one side and you can have your
19:08 chat GPT chat on the other side. And
19:11 it's just like, you know, sped up the
19:13 efficiency like a thousand% being able
19:16 to do both those things at once. So, I
19:18 figured
19:20 >> is the is the is the chat GPT window
19:22 like aware of what's happening in the
19:24 lovable thing? Is that right? So, oh
19:27 that's cool.
19:28 >> Yeah, it it it is really cool. It's
19:30 ultra cool. Um I love using the Atlas
19:33 browser despite all its, you know,
19:35 fallacies and like shortcomings and
19:37 whatever. Yeah. Um you know, everyone's
19:40 got my information anyway. So,
19:41 >> that's cool. And then and then what are
19:43 you gonna are you are you like building
19:45 a specific kind of app? Are you going to
19:47 like what's what's the what's the uh the
19:50 subject matter other than the tool? Um I
19:53 well I I thought that we could you know
19:55 come up with some ideas together like
19:57 whoever's in the group but then again
19:59 like when you have input from 16 people
20:01 like they might not all want to build
20:02 the same thing. So I do have a fallback
20:03 idea if we
20:06 got to have the fallback idea.
20:09 >> Great. Um and we might have time to
20:11 build two smaller ones or we might have
20:12 time just to build one and like iterate
20:14 on it a bit more. It's not going to be
20:16 like a, you know, like a full
20:18 lifechanging thing like Brandon and
20:20 Rick's thing in Florida, but but we'll
20:23 come up with something.
20:24 >> That's really cool. And it's it's one
20:26 hour or two hours.
20:27 >> It is an hour. Um, and then, you know,
20:30 people can like play along in their own
20:32 screens and, you know, just follow along
20:35 or they can just watch or obviously
20:38 everyone can contribute. But
20:40 >> yeah, that's great. And if you if you
20:42 don't know what Lovable is, so Lovable
20:44 is one of what are called vibe coding
20:47 tools where basically you just ask in
20:49 English and it will code things for you.
20:52 It creates web applications. Right.
20:54 They're not doing mobile yet. That's
20:55 correct. Right.
20:56 >> Not yet. No. But I do have a tool for
20:59 that as well, which I'm learning. So
21:02 that's that's another LOL. And then and
21:04 then while you're up here, you might as
21:05 well brag a little bit because Vicki was
21:07 downright gloating today. She came in
21:09 and said, "I'm cranky." And then she
21:10 started talking about this. She was
21:12 downright gloating. Can Can you talk
21:14 about what what did what did you what
21:16 have you done with notion?
21:19 >> I just turned my whole life around
21:21 basically.
21:23 My my my
21:25 brain is in notion. Well, my brain has
21:27 always been in notion, but it's been
21:29 pretty much a reflection of my real
21:31 brain. Like it's scattered like in a
21:33 million places. Um, but using Notion AI,
21:37 um, I've been able to bring like a bunch
21:39 of stuff together and like I I just love
21:42 it. It's because there's no copy and
21:44 paste anymore. You just ask the AI agent
21:46 to do whatever you want it to do and it
21:48 just goes and or it will give you a
21:50 response and say, "Do you want me to do
21:52 that for you?" Sure, go ahead. And so
21:55 it's like, you know,
21:57 >> moving around.
21:59 >> It is. I Well, I had like five to-do
22:01 lists or something like that. So, it's
22:03 consolidated all my to-do list, which is
22:05 not an easy feat because you have
22:06 different labels. Like, it might be done
22:08 in one and it might be completed in
22:10 another. So, it's got to like match all
22:12 those things.
22:13 >> Um, but it's like, you know, it just
22:15 doesn't fight me. It just it does
22:17 whatever I ask it to.
22:20 >> I can ask it to. So, I'm I'm going to
22:23 try and like to teach this course on
22:25 notion, which obviously I learned from
22:27 someone else. And so I just asked the
22:29 agent like, "Can you just put a course
22:31 together teaching me, you know, to help
22:33 teach people what we're doing here?" And
22:35 like, it's put the whole course together
22:36 for me. So,
22:37 >> good.
22:38 >> Yeah. Created some databases. So, I'll
22:40 have a template that I can give people
22:41 that they can
22:43 >> um Okay. And I have the databases
22:45 already set up and it has all the
22:46 instructions in it and everything. And
22:48 it is just so very cool.
22:51 >> And yeah, it's just like organizing my
22:54 whole life. like you I was in my my
22:56 therapy appointment this afternoon and
22:57 she said, "So, what have you actually
22:59 done anything that you were said you
23:01 were going to do in the last um yes,
23:03 I've done this and I've done this and
23:04 I've done this and it's just
23:06 >> it has a you list of my tasks at the top
23:09 and you can show all your tasks or you
23:11 can show just the day or just the
23:13 weeking week or whatever. It's it's just
23:16 so so organizing. I don't I don't know
23:18 this feeling."
23:20 >> That's amazing. First first question.
23:22 Oh, you're getting lots of yays and very
23:24 cools and things like that. First
23:26 question, when's the notion class?
23:28 Listen, she's got to get done with the
23:29 lovable class before she can teach the
23:31 notion class. It's a it's a bottomless
23:34 pit of need over here. You know,
23:36 >> it is. We'll say in the new year. We'll
23:38 say
23:38 >> in the new year. Yeah, exactly. That's
23:40 awesome.
23:42 >> We have January and February already
23:43 scheduled. Um we have Lord Deod San and
23:47 we have um Korean. I can't I think
23:50 Kore's January and um digital gods is
23:54 February. So it might be March unless we
23:57 squeeze an extra one in, you know.
23:58 >> Squeeze extra one in. Okay, cool. That's
24:00 awesome. Um great. So Saturday, lol.
24:04 What time?
24:05 >> 400 PM Eastern. 1 p.m.
24:06 >> 4 p.m. Eastern. And if you if you don't
24:08 know what we're talking about, go to
24:09 community.thesalon.ai
24:11 and in the upper lefthand corner, you'll
24:13 see events and then look for the event
24:15 this Saturday called LOL. And that's
24:17 Vickiy's event. And please go RSVP so
24:20 she knows how many people to get nervous
24:23 for.
24:24 >> Thank you. I wasn't until then. And also
24:28 if anyone else has an idea for an LOL
24:30 that either they want to present or they
24:32 have a topic that they want to hear
24:34 about and we can find someone to present
24:36 it. Just let me know that too and we can
24:38 we have a lot of you know months to fill
24:40 next year.
24:41 >> Vicki Vicki I think you came up with the
24:43 idea for the LOL, right?
24:45 >> And she manages it. So, so the LOL
24:48 stands for learn out loud. And the whole
24:50 idea within the AI salon, we've got this
24:53 idea of play first, um, create
24:55 excellence and generously lead. And that
24:57 generously lead part is about, you know,
25:00 learn out loud, right? Learn, um, share
25:03 what you're learning in real time while
25:04 you're doing it because no one can keep
25:06 up with all this technology. So, as each
25:09 of us learn little bits of things, if we
25:12 can just share with one another, it
25:13 helps us all grow and that's what that's
25:15 about. So, thank you for doing that,
25:17 first of all, and then thank you for
25:18 teaching one. I'm excited for it. So,
25:22 >> all right. Well, you did great. There
25:23 was no there there was one stumble.
25:25 There was one stumble.
25:27 >> I'll have it on video. So, we'll be able
25:29 to put that on loop and we'll send it.
25:32 >> Great. I can send you some more if you
25:34 like to add to it.
25:35 >> Exactly. All right. This was awesome.
25:38 Thanks. Thanks, V.
25:40 >> All right. Cool. Beautiful. That's I
25:43 like it a lot. I like it a lot. So, go
25:46 do that. Um, Gemini Fast.
25:53 Is this Gemini Flash
25:56 on Gemini.google.com?
25:57 Does anybody know?
26:01 Does producer Brandon know?
26:04 Gemini 3 flash.
26:06 Oh, fast. It's the fast one. Okay.
26:09 Because it doesn't say flash here
26:11 anywhere.
26:15 fast.
26:18 Um,
26:20 let's see. Um,
26:24 prove the
26:27 Ryman
26:29 hypothesis.
26:34 Silver Fox on YouTube. My re my
26:37 resolution for 2026 is to Lmore with
26:40 Vicki. Oh, that's great. Yeah, it's a
26:42 great great thing. Proving the Ryman
26:44 hypothesis or Reman hypothesis was
26:47 largely considered the holy grail of
26:48 mathematics. It is one of the seven
26:50 millennial prize problems and correct
26:54 proof carries a $1 million prize from
26:56 Clay Mathematics Institute. As an AI, I
26:59 cannot provide proof
27:03 that has eluded the greatest
27:04 mathematical minds for over 160 years.
27:07 However, I can't explain it.
27:12 Uh, I I
27:15 don't know. I suppose that's good. Um,
27:18 Vicky will need the lawn more. Beautiful
27:22 to prove you need thinking mode. Yeah,
27:24 exactly. Um, let's
27:28 see if create images. Is this still the
27:32 old image generating tool?
27:35 visual layout.
27:38 So, I'm in um I'm in Gemini right now.
27:42 So, let's do we're going to go make an
27:45 image. So, we'll say
27:48 um create an image of
27:55 seven king
27:58 penguins
28:04 marching
28:07 in front of a marching band.
28:13 From
28:17 Tusa
28:20 Luca
28:23 High School
28:28 High School um
28:32 down let's see
28:36 Time Square
28:39 in a snowstorm
28:43 on New Year's Eve
28:47 with Anderson
28:50 Cooper and what's his what's the buddy's
28:52 name that he drinks with?
28:56 That's not how you spell Tuscaloosa.
29:01 I don't know how to spell Tuscaloosa.
29:03 What's his name? The dude that drinks
29:05 with Anderson Cooper. They always get
29:07 drunk. Is it Andy something? Andy. Andy
29:11 Cohen. Is it Andy Cohen?
29:15 Cohen.
29:17 I think it's Andy Cohen. And
29:20 Andy Cohen. Is that right?
29:24 Is it Andy?
29:26 Come on. Yo, someone can Google this,
29:29 right? You've got chat GPT. I'm trying
29:32 to write a prompt here.
29:36 Nothing. I get nothing. Yes. Okay. I got
29:38 Yes. Andy Cohen, I believe. Okay. Let's
29:42 see if we did this. So, what do we say?
29:44 Seven King Penguins, a marching band
29:47 from Tuscaloosa High School,
29:50 Anderson Cooper, and Andy Cohen.
29:54 That was pretty fast, wasn't it? And
29:57 let's see. One, two, three, four, five,
30:00 six.
30:02 Is there a seventh one back there? Oh,
30:04 it only gave us six penguins.
30:06 And that looks like two men, but that's
30:10 not Anderson Cooper.
30:12 You're in fast, so it won't use Oh, it
30:15 won't use pro. Oh, okay. Fine. Be that
30:18 way. Let's go get that prompt, then.
30:21 Flip it into pro.
30:46 Okay,
30:48 let's see. I suppose we could go to Oh,
30:52 you know what we could do? They both
30:54 have silver hair. I know. Exactly. Those
30:56 two dudes did not. All right, let's see
30:58 if this one's better.
31:00 Oh, it did the same thing. This is a
31:02 nano banana bug. I hate that bug. It
31:04 makes the exact same image sometimes.
31:07 New chat.
31:10 Boom. Uh, boom.
31:13 Pro. Generate images. Um,
31:17 16 by9 wide. Let's go.
31:33 Have you all noticed that Ed
31:38 for Santa Claus show him with a darker
31:40 gray beard?
31:42 Oh, that
31:45 this got it right. Okay. One, two,
31:48 three, four, five, six, seven penguins.
31:52 Let's see if this is the Tuscaloosa
31:54 marching band.
31:56 I can't see what that says, but
31:58 possibly.
32:00 And then there's Anderson Cooper and
32:02 Andy Cohen. It worked in a snowstorm.
32:07 Okay. All right. So, let's now we're
32:12 going to close
32:15 this.
32:17 We're going to change how we're sharing.
32:25 La
32:52 da da da da.
32:54 I know it's annoying.
33:00 Okay, so there's Gemini on the left.
33:03 Let's do chat GPT on the right.
33:07 Is this it?
33:13 Chat jeep.
33:17 And then we're going to put this into
33:19 Oh, yeah. Oh, the other thing um chatbt
33:23 got by the way is
33:27 let's see
33:29 how do I get more of these KA.
33:37 Ah, there we go. Um
33:40 there's this this Can you see this?
33:43 Yeah.
33:45 So, when you click on the plus button,
33:47 there's a bunch of things like shopping
33:49 research and deep research and agent
33:52 mode and and then there's this further
33:54 drop-own menu. And in that, you've got a
33:57 bunch of apps. And if you go to add
33:59 sources,
34:02 um, and then click here, add, and then
34:04 say connect more. Like why they buried
34:06 it in in front of a bunch of ads is
34:10 crazy to me. This is not the latest
34:12 interface.
34:14 >> Yeah. If you click on your history or
34:16 your left nav, um there's a in in chat
34:20 GPT. Um if you expand the
34:24 you've got apps at the top.
34:28 Oh yeah, look at that. And images is
34:31 new, too. Okay. Oh, cool. Okay, so if
34:33 you go to apps, very cool. Thank you. So
34:36 here's featured apps. Photoshop, Air
34:39 Table, Apple Music, Alltrails,
34:41 Booking.com.
34:43 So
34:46 chat GPT is going to connect with any
34:49 app out there, any developer that wants
34:51 to do it. They're now opening this up.
34:53 They're opening it up like Apple opened
34:54 up the app store. And what this means,
34:57 if you're trying to figure this out,
34:59 what does this actually mean is that
35:01 with from within Chat GPT, you're going
35:04 to be able to access any other app.
35:07 Um,
35:15 Low res. You can download the full
35:18 image. Oh, of this. Yeah, I can download
35:21 the full one.
35:24 Uh,
35:26 yeah, Nano Banana Pro is 4K images. I
35:29 think you can make them 4K. Maybe I have
35:31 to up Do I have to up res it in here?
35:34 Um, let's see. up
35:36 up res
35:38 to um 4K.
35:43 We'll see if it does it.
35:53 Um and then let's go back over to chat
35:55 jeopet and we're going to go images tab.
35:59 And it's got this really weirdass shitty
36:02 interface.
36:05 It's fine. It's a fine interface. All
36:08 right. Is this This actually looks
36:10 pretty big. No, maybe it's the same.
36:33 I mean that really is close enough that
36:35 if you saw that
36:38 somewhere Tuscaloosa sort of
36:44 Tusk sort of Tuscaloosa High School deck
36:47 deck
36:51 they got the tubas in the back the
36:52 little clarinet they're wrapped in
36:54 plastic. Yeah.
36:57 Amazing.
36:59 There they are doing their thing.
37:04 Absolutely wild.
37:06 So, let's take that prompt
37:12 over here to
37:15 chat GPT.
37:18 Let's compare them, shall we?
37:23 Okay. So, AI festival is just for
37:25 advanced users. No. I need to tell them
37:28 to wait and join something later. No,
37:32 no, it's it's not Tobias. It's not for
37:35 advanced users. That's what I was
37:36 saying. There's people that people that
37:38 came to Festivus last year that knew
37:40 nothing about AI a year later are like
37:44 they got motivated and they realized it
37:46 wasn't just about the technology and
37:48 they realized that they didn't have to
37:49 be a flipping math major to be any good
37:51 at it. Oh, I never finished on the
37:54 thought of the importance of
37:57 the app store. Oh, yeah. Like, okay.
38:06 What
38:08 OpenAI is doing with this app store is
38:10 they're laying the foundation for all
38:13 large language models to
38:16 to try to lure your lure lure lure lure
38:21 you into attract you into their web
38:26 and keep you there. They're trying to
38:28 every large language model company right
38:30 now is trying to create the everything
38:32 app so that you never have to leave
38:34 them. Wait, chat GBT didn't like using
38:37 Anderson Cooper. It didn't. Oh, it
38:39 doesn't follow our content policy.
38:42 Then you suck.
38:45 Try uh
38:49 try it with generic
38:55 um news broadcasters.
39:02 Attention
39:09 is all you need. I don't have any
39:10 attention. I'm a little exhausted after
39:13 the uh the Sydney rehearsal. Although I
39:15 didn't do anything except give notes.
39:21 The apps are trying to capture your
39:22 attention. Oh, I see what you're doing
39:24 there. That's good. That's smart. Yeah,
39:26 they're all going to try to get you to
39:28 use their app and their app's going to
39:29 be better. Elon Musk is absolutely on
39:32 this train.
39:39 Huh.
39:42 I actually kind of like this one better.
39:46 They're generic. There's your generic
39:48 broadcasters.
39:50 And let's see. One, two, three, four,
39:53 five, six, seven. Seven penguins. High
39:56 school band.
39:58 It got the red and white of Tuscaloosa.
40:01 Happy New Year.
40:06 Huh. Fascinating. Let's bring up another
40:08 one. Let's do Grock. Grock.com.
40:12 This one will definitely do Anderson
40:14 Cooper.
40:16 Um, so we're going to go to the imagine
40:18 tab
40:20 and I'm going to paste that in there.
40:26 Now, Grock, if you haven't played with
40:28 it, is really quite good. Let's see.
40:31 Let's grab one of these. One, two,
40:34 three, four, five, six. So, not the
40:36 right ones there. Is that Anderson
40:39 Cooper? Not really.
40:42 But look, it just keeps making it just
40:44 keeps m making image after image after
40:47 image.
40:50 And And they're fast. Like, I don't know
40:51 how it's doing the diffusion this fast,
40:54 but it is. And then it turns them into
40:56 movies.
41:15 Yeah, Grock's really fast.
41:17 And then I'm going to pull up Meta
41:21 cuz Meta
41:40 Oh my god, that's hilarious.
41:43 Oh my god,
41:46 I need 11 labs. Morgan Freeman,
41:50 the king penguin took over the island.
42:05 Can I loop?
42:25 and good night everyone. Have a
42:27 fantastic evening.
42:31 Okay,
42:34 so Grock wasn't so big on prompt
42:36 coherence, but it's really fast. Um, and
42:40 it'll just instantly make you videos. By
42:42 the way, new street photography hack. Go
42:46 do street photography. If you if you
42:48 know what it is and you like it. I love
42:50 it. I love street photography. Just
42:53 destined a little see something you
42:54 think's a picture. Take it.
42:57 Upload your pictures to Grock and it
43:00 instantly turns them into videos. Let me
43:02 show you something I did in New York. I
43:03 haven't even done this yet cuz cuz this
43:06 is a project that I
43:10 I actually need to finish.
43:12 Oh, wait.
43:15 Squirrel.
43:16 Okay. Remember remember the other night
43:18 I was trying to find um these black and
43:21 white pictures from this box. Remember
43:25 this box? It's this little 4x5 card box
43:28 and in it are all these 4x5 uh film
43:33 negatives from 1937.
43:38 And so I took a couple of pictures of
43:40 some of the things in there.
43:43 And so here's here's one of them. That
43:45 cute little kid in the middle
43:48 there. They got look at the size of
43:49 those rabbits.
43:52 And then this is
43:55 Donald Donald Ira Wilson, son of Ira and
43:58 Blanch Wilson, was born on June 10th,
44:00 1920 in Atlanta, Kansas. Was killed in
44:03 action on the Italian front December
44:05 9th, 1944.
44:08 So it's someone that was killed in
44:10 action. There's a
44:13 in grateful memory of from Franklin
44:15 Roosevelt.
44:18 And then there's just these pictures and
44:19 they're like, "These people are really
44:21 stylish
44:23 and the photographs are really good.
44:27 It looks like a woman photographing her
44:29 shadow. Looks like a bob haircut.
44:34 Like look at those dudes. Look how cool
44:35 they are. Look how cool that car is."
44:42 And so you can take any of these and
44:43 turn these into images. So So Grock's
44:45 really good at that. So, so street
44:48 photography you can do old photography
44:49 and turn it into photos, but you can
44:52 also do
44:57 um
45:01 okay all these. So,
45:07 see these things that look like kind of
45:09 photographs on the wall? They're all
45:11 mosaics.
45:15 Look at that.
45:19 This is in New York City at the 72nd
45:21 Street station. So, I took pictures of
45:24 every single one of these.
45:27 The whole station's full of them. Like
45:29 this dude with the briefcase flying
45:31 open.
45:34 Mama with a stroller,
45:38 bow-legged cowboy.
45:43 Aren't these great?
45:46 Then
45:48 I went into Grock. So let's go into
45:51 Grock again. Is this Grock? Let me see.
45:54 Where's my
45:56 October social September?
46:00 Let's see. History. Imagine.
46:07 Are these buying? Yeah, there we go. Oh,
46:09 yeah.
46:10 So, here's that one.
46:13 That's one of those old photos turned
46:16 into that which is so cool. Creepy cool.
46:23 This is a fun one. Little kid. It's like
46:26 I'm done, mom. I'm out of here.
46:30 Even in 1937, kids were a pain in the
46:33 ass.
46:36 Oh, man. All right, let's see. Do it.
46:41 Uh,
46:44 no. Did I not do them in Grock? I
46:46 thought I did them in Grock.
46:50 Guess not
46:52 chat. No,
46:59 I don't remember where I did them. But
47:02 anyway, I turned those things into cool
47:04 into cool images. All right, so that's
47:06 Grock. So, so we've got Grock image
47:10 is good, fast, really good at video.
47:13 Chat GBT image. Um, was it Rick
47:17 McCaulay? Yeah, Rick McCaulay today um
47:21 described Chat GBT's image generation
47:24 like a a snapshot, you know, like a
47:27 disposable camera. And uh and Nano
47:30 Bananas like a digital SLR, like you
47:32 just have a lot more controls and
47:34 things. Um,
47:36 I think that the thing that sets Nano
47:38 Banana Pro apart from anything else so
47:41 far, including Chat GBT, is that it's
47:44 natively tied to the reasoning engine.
47:47 So, Nano Banana is just better at
47:50 solving things visually. And then the
47:52 the last one I'll show you here is Meta.
47:55 Meta.ai
47:59 also has image generation. So, if you go
48:02 down the lefth hand side here and you
48:04 say create
48:07 and then let's pop in our prompt.
48:13 We'll see how it does.
48:22 It's using midjourney, so I don't think
48:23 it's going to get the prompt coherence
48:25 right. Oh, yeah. And I'm I'm using a uh
48:27 a mood board, obviously, or a style
48:30 reference. So, I'm going to get rid of
48:31 that.
48:33 We'll try it again.
48:36 But these are kind of cool.
48:39 Why can I not view that?
48:43 Bigger. I want to see it bigger. Look at
48:46 that. That's pretty gorgeous.
49:05 That was cool.
49:10 All right.
49:12 Close that. And then here's
49:16 So this is MidJourney's
49:20 version of this. Let's see. One, two,
49:22 three, four, five, six, seven. That's
49:25 got seven penguins. That's not so bad.
49:41 It's more than seven. Two, three, four,
49:43 five, six, seven, eight, nine.
49:53 Absolutely bonkers. All right, so
49:58 So right now there's just from the
50:01 commercial companies
50:05 and this is not including ideog it's not
50:08 including Leonardo it's not including
50:10 crea it's not including Deco here
50:14 which a lot of those apps that let you
50:18 use different models you can use all
50:19 these models inside them anyway
50:23 um
50:24 turn them into Danny Dvito Oh, that's
50:27 pretty cute. Yeah, let's do that. Okay,
50:30 let's do it. Uh, they won't let us do
50:32 it. We'll do it at uh Gemini. Okay,
50:38 turn
50:39 the uh let's see.
50:42 Zoom in
50:48 on the penguins
50:51 and make them all Danny
50:56 Devito
51:00 from the Penguin
51:03 movie.
51:05 Um,
51:09 they
51:11 should look silly
51:16 but menacing.
51:39 Baba butt.
51:42 What's happening? Baba butt. Baba butt
51:45 in the house.
51:48 Look, seven penguins.
52:00 That's pretty good, man.
52:02 All right, Google.
52:06 I I would seriously like to understand
52:10 how Google went from being a laughing
52:13 stock
52:15 the the year Chat GPT came out, they
52:17 they were so bad
52:20 to being this good.
52:23 Just goes to show you when you dump a
52:24 big old pile of [ __ ] money toward a
52:29 single thing, good things can happen
52:38 on your unless your name is Mark
52:40 Zuckerberg. That's a great point.
52:43 What was he throwing? $200 million
52:45 signing bonuses for 22year-old
52:48 engineers.
52:51 Wait a minute. That didn't work out.
52:56 Is Pate in here? Pate, what's happening?
52:59 Listen, Pate, you're not going to be
53:01 comfortable here anymore. I keep saying
53:03 nice things about Google.
53:06 I keep I keep demoing things. I keep
53:08 expecting Google things that I demo to
53:11 suck and they keep being really good.
53:14 Uh, I would argue, and you you know,
53:17 you've told me this day would never
53:19 come. I would argue that OpenAI has lost
53:24 the lead
53:26 and they've lost their uh
53:30 their their purpose, their mojo. They've
53:32 lost their mojo. Anyway, um
53:38 there is so much I'm going to I'm going
53:40 to um
53:43 I'm going to check out for a minute.
53:47 I talked about this on the AI readiness
53:49 project podcast today
53:52 with all of the stuff that's coming out
53:55 and it's a lot. It's like it's a lot.
53:59 Like Brandon's having a tough time
54:01 keeping up with it and then he tries to
54:02 explain it to me and I'm just like my
54:04 eyes glaze over.
54:07 Um,
54:13 it is it is now my contention that the
54:16 tools don't matter.
54:20 Now,
54:22 could we do anything remarkable without
54:24 the tools? No. No. The tools the tools
54:26 matter to do stuff, but the tools don't
54:28 matter. Like, you've got to learn them.
54:30 You've got to learn to
54:33 prompt this one just correctly and you
54:35 got to learn to the foibless of that one
54:38 and this and that. I just think the
54:40 tools are getting good enough now that
54:41 the tools are good enough and then it
54:44 comes back to who are you and what are
54:46 you trying to do in the world? Jim Ross
54:48 was in here and just said you had a
54:50 breakthrough week with AI. One of the
54:53 reasons Jim can have breakthrough weeks
54:55 with AI is he's constantly thinking
54:57 about his business.
55:00 He's thinking about his business and his
55:01 customers and how does he serve them and
55:04 how can how can he let people know the
55:06 cool thing that he came up with
55:09 to get it in front of them. How can he
55:11 do that?
55:13 And so he just keeps asking
55:16 all of the questions other than
55:20 what's the best AI, right? Because there
55:24 is no best AI. They're all fine. They're
55:27 all janky. There's vibe coding ones and
55:30 picture ones and movie ones and word
55:32 ones
55:35 and
55:37 they're all fine.
55:40 You know, none of them at this point
55:44 are good enough that you can,
55:47 you know, check out as a human,
55:51 but they're all fine. And if they're all
55:53 fine, then that means you can choose any
55:55 tool. You can literally choose any tool.
55:58 Well, how do you choose the tools? Where
55:59 do you begin?
56:01 You begin with what you want to do.
56:08 What do you mean, what do I want to do?
56:09 Like literally, if you could do
56:10 anything, what would you do? I don't
56:14 know. Uh,
56:16 I like pizza. Okay, good.
56:21 Do you want to make pizzas for people?
56:24 Do you want to go to a pizza eating
56:26 contest or create one? Do you want to
56:29 make pizza shaped refrigerator magnets?
56:33 What do you want to do? H, I like pizza.
56:37 Okay. So, you're gonna eat pizza? Yeah,
56:39 I'm hungry. Okay.
56:42 Right.
56:46 If your answer to every question is
56:48 just, I'm kind of hungry. I want pizza.
56:51 This,
56:53 you might need to be a tad smarter than
56:55 that for this AI stuff, but not much
56:57 more.
56:59 I like pizza.
57:00 I like to
57:03 Rick McCaulay chat image 1.5 is an
57:07 amateur artist with great presets but it
57:10 has strict limits. Nano banana is the is
57:14 the world model. It gives you control
57:16 and isn't offended
57:19 and it generates LeBron while GPT
57:22 refuses. Fascinating.
57:28 Gotta find your icky guy. Yeah, the icky
57:31 guy thing is really is really powerful.
57:34 Um I think even just
57:38 CJ Fletcher last year last year's
57:40 Festivus, which by the way a week from
57:43 Friday is AI Festivus 2025. If you have
57:46 not signed up, go to afestivist.com
57:49 right now. Sign up. Last year CJ
57:52 Fletcher
57:54 did a session on midjourney
57:57 on or no stable diffusion on creating
57:59 images. He's an artist and his goal is
58:01 to be in MoMA within five years. Three
58:04 years have passed. He's got two years to
58:06 get in MoMA.
58:09 Um
58:10 so he taught this class on how he makes
58:12 art and how he thinks about art with AI.
58:16 and he's got he's got this super
58:18 powerful gaming laptop that he carries
58:20 around with him and it's got enough GPU
58:22 in it that it can run stable diffusion
58:24 and he can render all of his images
58:26 locally.
58:28 And when he started his session, he
58:30 said, "Here's the most important thing I
58:33 do when I sit down to make images
58:36 is I close my eyes
58:40 and I take a deep breath.
58:50 and he centers himself.
58:53 And in that moment, he gets to say,
58:55 "What do I want to create?" Not just, "I
58:59 can create it. Squirt, squirt, squirt,
59:01 squirt, squirt, squirt, squirt. Let me
59:03 squirt out more images. Let me squirt
59:05 out more videos. Let me squirt out more
59:07 songs."
59:09 He's taking a moment to go, "Where am I
59:11 right now? What do I want to do right
59:13 now?
59:15 That's really important. Um, Tik Tok
59:17 pin.
59:20 Woohoo. Festivist premium ticket. Love
59:22 that, Rick Olsen. Thank you. That's
59:24 awesome.
59:26 Um, my break my breakthrough today was
59:29 co-pilot still makes up links even after
59:31 I yelled at it. Yeah, Co-Pilot.
59:35 So apparently like within Microsoft
59:41 they're having some issues with their
59:43 salespeople like going into
59:47 big enterprises and the big enterprises
59:49 are like yeah we'll take your other
59:51 stuff but yeah co-pilot sucks. We're
59:54 using chat GPT and we're using claude
59:56 and we're using Perplexity. They're
59:58 using anything but C-pilot because it's
1:00:00 it's bad. But whatever
1:00:04 at some point it'll get fine.
1:00:08 Claude by AI research company Anthropic
1:00:10 worked better for coding for me. Yeah,
1:00:12 Milan Milan Junka. Um
1:00:16 Claude Opus 4.5 does still seem to be
1:00:20 the the winner on coding.
1:00:24 Um, Gemini Flash that just came out is
1:00:29 um,
1:00:31 orders of magnitude cheaper and a lot
1:00:33 faster than Gemini 3 Pro, but still
1:00:36 pretty damn good. Um,
1:00:42 there's so much there's so much
1:00:44 Microsoft sucks. They do. They do. Um,
1:00:49 your play is the meaning in your life.
1:00:54 life is meaning making. I would say the
1:00:57 play is one of one of the things in my
1:00:59 life. But you know what was cool, Rick?
1:01:00 I don't know if you if anyone's a
1:01:02 theater geek here. Um
1:01:06 I was on this Zoom call with these two
1:01:08 actors in New York and my writing
1:01:10 partner in upstate New York and I'm
1:01:12 sitting in Denver and I'm sitting in
1:01:13 this corporate, you know, meeting room.
1:01:16 And it's got like frosted geometric
1:01:18 shapes on the windows and I'm just
1:01:20 looking at myself on the screen and it's
1:01:23 like 7:00 at night and there's no one in
1:01:25 the office
1:01:27 and I'm rehearsing a play and it just
1:01:30 took me back to when I was doing plays
1:01:32 in New York City where you know you'd
1:01:34 work a 9 to5 and then you'd go rehearse
1:01:37 from like you know six o'clock until
1:01:39 [ __ ] two in the morning and you'd do
1:01:41 that for for two months and then you'd
1:01:44 launch the play. You You'd go do the
1:01:48 play and like 17 people would come on
1:01:50 the biggest night and two people would
1:01:52 come on the the shittiest night
1:01:55 and then you'd be depressed and then
1:01:58 you'd do it again and then you do it
1:01:59 again and it reminded me of that. It
1:02:01 reminded me just, you know, having my
1:02:03 corporate day job going to do the
1:02:04 theater gig. So,
1:02:07 but I'm telling you that the Sydney
1:02:09 story you you'll see if you come to
1:02:11 Festivus, it's it's really good. This is
1:02:13 the the part that we've chosen to show
1:02:16 to to to excerpt for the uh for Festivus
1:02:21 is the heart of the show. It's where the
1:02:24 tech reporter brings her to life. It's
1:02:26 really good. Um do you have to create in
1:02:30 the images tab? No, you can
1:02:34 I think you get the same. Okay. So, in
1:02:37 um
1:02:41 is it Gemini? No. Uh
1:02:45 No, it's in chat GPT, right? Which one,
1:02:49 Vicki? I think you were talking about
1:02:50 chat GPT either.
1:02:54 No, they they just made these images
1:02:56 tabs because
1:02:58 they're giving you presets.
1:03:04 Yeah, there you go. They've got all
1:03:06 these presets, right? Like create a
1:03:08 holiday card. So there's a create a
1:03:10 holiday card. There's a photo of me
1:03:15 in a in a garbage pale kids t-shirt.
1:03:24 Chat GBT has nanny guardrails. It really
1:03:26 does. It's garbage.
1:03:29 Woohoo. Festivist premium ticket.
1:03:33 Rehearsal night image in a regulars. Oh,
1:03:35 nice. Okay, cool.
1:03:43 Here's my holiday greeting card.
1:03:47 Holiday cheer with Coca-Cola warmth.
1:03:58 That looks decently like me.
1:04:08 Oh, that's hilarious.
1:04:12 Um, let's see.
1:04:16 Um, if I go add photos
1:04:21 and I go Santa
1:04:32 D.
1:04:48 Oh, the question was images. Yeah, the
1:04:50 images. So, I can just be in normal chat
1:04:53 and just go I can click on the plus
1:04:55 button and then say create image and see
1:04:57 how it's got the styles down here. Like
1:04:59 we can do dramatic.
1:05:00 We'll choose that photo.
1:05:06 Is that going?
1:05:08 Yes.
1:05:21 Yeah, I guess LeBron probably cut a deal
1:05:23 with them. The thing I'm still blown
1:05:25 away with that that
1:05:29 people aren't talking about as a as big
1:05:31 a deal as it is is the Disney deal with
1:05:35 Chat GPT. You're going to be able to
1:05:37 make Disney characters inside Sora, the
1:05:40 Sora app, the video app.
1:05:43 Um,
1:05:45 and normally Disney would charge a
1:05:48 licensing fee to be able to use their
1:05:50 IP.
1:05:52 They paid chat GPT OpenAI a billion
1:05:56 dollars to be able to be on their
1:05:58 platform.
1:05:59 I find that remarkable. Remarkable.
1:06:04 Look at that. Look at that man.
1:06:08 Look at that man. He's He's wet. He's
1:06:10 moist and dirty. He's a moist, dirty
1:06:14 man.
1:06:15 Look at that man. He's all man.
1:06:20 He's got man hands.
1:06:23 He's got a man tan in man hands. That
1:06:27 man.
1:06:31 All right. Rehearsal night image in
1:06:33 irregulars. In irregulars. Let's go.
1:06:36 Look. All right. Here we going. We're
1:06:39 going everybody.
1:06:42 Here we go.
1:06:53 The Penguin Beatles. Nice. Solid. Oh,
1:06:56 look at that. Rehearsal day by Kyle
1:06:59 Shannon and Andrew Watts. You know
1:07:00 what's funny? We were talking about um
1:07:03 securing the rights for this thing. And
1:07:06 and uh one of the women that we're
1:07:09 working with was like, "Maybe you should
1:07:10 do a musical about trying to secure the
1:07:12 rights for a musical." Which I thought
1:07:14 was pretty funny. Look, there's Champy.
1:07:18 Sydney's got blue hair. Why is she
1:07:20 sitting on like a hemorrhoid pillow?
1:07:25 Sydney, an artificial
1:07:28 an artificial love story brought to you
1:07:30 by Preparation H.
1:07:39 It's It's tough producing on Broadway.
1:07:41 We had to listen, Kyle. Good news is we
1:07:44 got enough money to put your show up on
1:07:45 Broadway. What's the bad news? Well, we
1:07:48 had to take on some sponsors. What do
1:07:50 you mean sponsors? Yeah. Well, our
1:07:52 biggest sponsors, Preparation H. What?
1:07:55 Yeah. Sydney's got to have uh some sort
1:07:58 of issue.
1:08:06 Man, that's some comedy right there.
1:08:11 All right. I'm on a kind of mission.
1:08:19 Thank you, Lord Digital Gods, for making
1:08:22 cool uh cool YouTube thumbnails all the
1:08:26 time. You're amazing.
1:08:28 Um, what's this? All right. Excellent
1:08:31 image prompt. Oh, that's cool, Danielle.
1:08:32 Very nice. There's Mr. It. Beautiful.
1:08:37 Lovely.
1:08:40 Oh, that was when I talked about peanut
1:08:42 butter sandwiches.
1:08:47 Oh man,
1:08:52 Vicki, sometimes in Mighty Networks
1:08:54 things don't post. Like I get things
1:08:56 where it's like I posted something and
1:08:58 then I go back and it's not posted. I
1:09:00 don't know what happens with it.
1:09:02 Something's odd.
1:09:04 All right. Um I gotta get out of here.
1:09:06 I'm crispy. I'm crispy as a critter. Um,
1:09:12 there's kind of too much to pay
1:09:14 attention to right now. So, don't just
1:09:16 go to Vicky's L on Saturday and then get
1:09:19 your rest.
1:09:21 Get your rest to deal with your family
1:09:23 and then come hang out with us a week
1:09:26 from Friday starting at 9:00 a.m.
1:09:31 We've builtin breaks so you you're going
1:09:33 to be able to go get pretzels,
1:09:36 nachos, hot pockets,
1:09:39 leftovers,
1:09:41 get in a bite with Uncle Stew,
1:09:44 and then come back and see the next
1:09:46 Festivist thing. So, it's 12 hours with
1:09:48 breaks.
1:09:51 9:00 am to 900 pm Pacific, Friday the
1:09:55 26th. 9:00 a.m. to 900 p.m. Friday or
1:09:59 no, Saturday, December 27th. All right.
1:10:05 All the tools are very overwhelming
1:10:07 right now. What's what what makes them
1:10:10 overwhelming is that there's not a
1:10:12 single tool that's a single tool
1:10:14 anymore. All the all the tools are
1:10:16 multi- it's like we went from a Bowie
1:10:18 knife
1:10:20 that you could like chop stuff with and
1:10:22 it was good. And now
1:10:25 every Frontier model site is like a
1:10:28 Swiss Army knife and it's like they're
1:10:30 the specialized one like the electronics
1:10:32 one with weird screwdriver tips.
1:10:35 It is it's it is it's exhausting right
1:10:37 now. Everybody did an amazing job
1:10:40 showing up with their YouTube comments.
1:10:42 Yeah, thank you for that. That was
1:10:44 really nice. YouTube comment.
1:10:47 Um,
1:10:50 I've been making freezer meals.
1:10:54 Will ship for money.
1:10:57 That's so those sound cool.
1:11:02 What number business is that of yours,
1:11:04 Vicki?
1:11:07 In between 3D printing in between 3D
1:11:10 printing um fidget spinners for people
1:11:14 like me,
1:11:16 she's cooking pre-done meals.
1:11:22 Oh man. Oh, she's talking about
1:11:25 nourishing me for festivus. Oh, that's
1:11:28 hilarious. That's awesome. That's great.
1:11:30 Um fantastic for festivus. Beautiful.
1:11:35 All right, I'm going to get out of here.
1:11:37 So, go see Vickiy's L on Saturday.
1:11:39 Today's Wednesday. Tomorrow's Thursday.
1:11:42 Uh, I did that math. I could figure that
1:11:44 out. Once I knew today, I could actually
1:11:47 figure out with my large language model.
1:11:50 What's the next day after this day we're
1:11:52 in? Depending on the time, depending on
1:11:54 your time zone. Steve disagrees with me.
1:11:58 Different time zone, different part of
1:11:59 the world. He's living in the future.
1:12:05 There's no reason he needs to be
1:12:06 obnoxious about it, but he is. He's
1:12:07 living in the future, right? Anyway,
1:12:11 it's Wednesday here. I know because I
1:12:13 can do math because I have a large
1:12:14 language model built right here. Low
1:12:16 wattage. I think it's something like 10
1:12:18 watts.
1:12:21 Tomorrow's Thursday. So, come back here
1:12:23 tomorrow 8:00
1:12:26 and we'll we'll do some [ __ ] All right.
1:12:31 manastic.
1:12:34 You don't have to figure it out. If you
1:12:36 if you're in Notion, it'll know. I swear
1:12:38 to God, Vicki, if I go through the the
1:12:40 angst
1:12:42 like right now with notion, I'm in I'm
1:12:46 in the notion shame state. The notion
1:12:49 shame state is everyone's talking about
1:12:52 notion. Everyone's talking about how
1:12:54 remarkable it is once you team up AI
1:12:56 with notion.
1:12:59 And I've actually gone and looked at
1:13:00 notion and I don't know if you download
1:13:02 it or you go to the website doesn't
1:13:03 [ __ ] matter. Whatever you do with it.
1:13:07 I looked at it and then I tried to chat
1:13:10 GPT it and chat GPT said oh notion's
1:13:13 based on these concepts of I don't know
1:13:16 widgets and snorts or it's it calls them
1:13:19 something like components right and so I
1:13:23 go to notion and nothing in there is
1:13:25 labeled what the core idea of it is and
1:13:29 so I'm just an angry Gen Xer and so I
1:13:32 said no
1:13:37 I
1:13:39 So, I'm going to take when Vicki teaches
1:13:42 her LOL, her notion, LOL, I'm going to
1:13:45 take it. And then if I install this
1:13:48 stupid [ __ ] thing and it's horrible,
1:13:50 what do I do?
1:13:53 What do I do?
1:13:58 We've officially entered whiny
1:14:00 Wednesdays.
1:14:02 I don't like AI. There's too much to
1:14:04 learn. I'm exhausted. You know, I got a
1:14:06 musical, right? It's probably not going
1:14:08 to go anywhere.
1:14:14 Kill me. Fire me.
1:14:21 All right, everyone. You have yourselves
1:14:24 a fantastic evening. Um, go join the AI
1:14:27 salon. Join the ASLon mastermind. Go to
1:14:29 the LOL's. Sign up for festivists. Get
1:14:32 your ass out there. Share things. Help
1:14:34 us. help us bring new people into
1:14:36 Festivus. All right, good night
1:14:38 everybody. Night Kelly Camp. Thank you
1:14:40 all for posting on YouTube. Thanks all
1:14:42 for hanging out tonight.