AI Learning Lab Preshow

Transcript

1. 1:18Wait a minute. [laughter] I didn't put myself on stage for the YouTube stream. [laughter]
2. 1:36>> [music] [music] >> Good evening, good people. Welcome to the AI Learning Lab pre-show.
3. 1:50Getting ready for the big show to start at. It's 7:30 Mountain time. Little over little under 20 minutes.
4. 1:58We'll be going live in the AI salon. Tonight is Monday night. Meltdown Mondays we call them. I'm going to be talking about all the
5. 2:09stuff going on. [music] I'm going to be talking about more and more. I'm gonna I'm gonna keep unpacking
6. 2:20this big insight I had, [clears throat] which is uh I realized >> [music] >> last week that uh [music]
7. 2:38[music] I don't know what I just played. [music] I'm ready to melt down.
8. 2:54I had a I had a long day today. I had a an art show to hang for the misses. So, that's done and in. Got a little bit
9. 3:01more work to do on that. Um, had some busy stuff going on at work today. Um, so I wasn't able to pay attention to a lot of the AI stuff. Um,
10. 3:13AI stuff's in a weird place. [music] >> [music]
11. 3:27[music] [music] >> So anyway, we'll be going live in the AI salon at 7:30 Mountain time. So if you're on
12. 3:42LinkedIn or YouTube, the link is in the description. If you're on X or Tik Tok, it's in my profile. Um, just head on
13. 3:49over to the salon. That's where the live stream will be at 7:30. And uh, yeah, I'm in a weird place with AI. I love it, but I'm worn out.
14. 4:04[music] What was the big company today that just hired back 300 people? It was a company that hired back 300
15. 4:15people that they laid off because they had AI. Let's see. Rehired 300. [clears throat] Ford
16. 4:29Motor Company rehired 300 people [music] that they assumed AI could replace.
17. 4:55One of the interesting things about the government doing the the stuff with anthropic and open AI, there's there's a couple of things.
18. 5:04One is they're not messing with any other companies. They're not messing with any of the Chinese models. And this is a yet,
19. 5:12but they're just messing with Open AI and anthropic. So, that's starting to look like Open AI and Anthropic have significant leads.
20. 5:22Xi, XAI hasn't caught up. Meta not even in the conversation. Google's probably close, but they're also probably
21. 5:33being very, very conservative with what they let out right now. Um, discussing the AI revolution and job
22. 5:43market changes. Share your thoughts. So, uh, this is the AI Learning Lab pre-show. We'll be going live. Well, I'm
23. 5:49live now, [clears throat] but the show will go live at 7:30, so in 15 minutes inside the AI salon. So, if you are um
24. 6:00on LinkedIn or YouTube, the the link to the salon is in the in the description below this video. If you're on Tik Tok
25. 6:08or X, it's in my profile. So, just click on the link there. Thank you, Mary, for liking the live. [music]
26. 6:33There's been something baby I've been trying to say [music] for it seems I don't know how [music] it's in a future now surrounding
27. 6:52Surren surrender to whatever cheap thrill can be found. [music] There's been a little [singing] trouble since you came to my rescue.
28. 7:10[music] And if you're like all of the rest, I would have quit you long ago, but I couldn't do that.
29. 7:28Um, meltdown Monday. It is meltdown Monday. I'm not particularly fired up about anything, but I am Well, maybe
30. 7:36that's not true. Maybe that's not true. What I what I'm what I'm a little pissed off about is that
31. 7:56whether whether what the what happened with anthropic whether it was the right move or not, it happened in such a way that it just
32. 8:12it made it made it like shaky like all this AI stuff like a [ __ ]
33. 8:28and I think it effectively slowed down the perception of of AI progressing.
34. 8:44Anthropic got Uno reversed though. You mean you mean Fable? No, they sort of got it reversed. It's not It's not been launched unless I
35. 8:55missed that. No, I would have I would have seen that. It's not back. We don't We don't have access to Fable.
36. 9:05They they made it they made it available for select trusted members of the crew. But anyway, um [music]
37. 9:19there's a bunch of things converging like there isn't going to be enough compute. the fact that we don't have the political leadership
38. 9:34in this country to sell in why we need data data centers and to to sell in why paying attention to AI is a good thing, not a bad thing.
39. 9:49And so the the narrative got was was getting pretty out of hand with the with the anti-AI rhetoric. And I think one of
40. 9:55the things that the the thing with anthropic did was put a little bit of ice water on that on that rhetoric, which is probably not a
41. 10:06bad thing, but what it also does what it what it also communicates to me is businesses and individuals should assume
42. 10:23we're not going to get access to the really powerful models until proven otherwise. We should assume we're not.
43. 10:32And then I think we should also assume we're not going to get access to enough compute. And so that's got my head spinning in
44. 10:43all sorts of ways about what can what can we do as individuals and as small businesses today that ensure we're going to have some leverage
45. 10:58and agency no matter what happens with all that [ __ ] And so that's what I'm going to be talking about tonight. So I'm going to
46. 11:05be talking about that in the AI salon at 7:30. So, in nine minutes, I'm going to be heading over to the AI salon and
47. 11:10we'll start this thing proper. And that's what I'm going to be digging into. I'm going to start talking about the
48. 11:17Dunbar number and why I think that's really important. I think we're about to reimagine marketing.
49. 11:33It's like the CIS admins have root, now the AI companies have root. Hm. Yeah. Yeah. Exactly. [music] Tiny box.
50. 11:46Yeah. I mean, listen, if if George Hots can can get Tiny Box, that's that's George Hotz's thing, right? I mean, he's been talking about, you
51. 11:57know, AI boxes for a while. if he can deliver tiny box in a powerful way. Listen, I don't think it's it's it's not
52. 12:04just about specking a machine there. There's a lot of there's a lot of signals that it's about to get really expensive to be able to
53. 12:16have enough compute. So, I'll be talking about that. So, if you're on LinkedIn or YouTube, in the description of this video of this live
54. 12:24stream is the link to the AI Salon live stream where we're going live in in 8 minutes. And then if you're on X or Tik
55. 12:31Tok, it's right there in my profile. So, just click there and uh click on that AI salon link. If you don't know about the
56. 12:38AI salon, it's a community of really remarkable people that are curious about AI, trying to figure it out, trying to
57. 12:44figure out the future of work, trying to figure out the future of purpose, doing this thing called the great repurpose.
58. 12:52It's people in there creating daily practices around how they use AI. It's about professionals leveling up their
59. 12:58game. It's pretty amazing. Now, AI labs, OpenAI, Anthropic, Google, Chinese labs have root on intelligence itself, the
60. 13:06ultimate substrate. They control the models, the weights, training rounds, the infrate scaling. Well, yeah. I mean,
61. 13:17anyone who can go raise $500 billion can be one, too. [laughter] [music]
62. 13:32I don't I don't I don't think that there's going to be central control. um unless the US does something really
63. 13:40boneheaded like make um open- source models illegal, which wouldn't surprise me if they do. But that's that's a lot
64. 13:49of the stuff I'm going to be talking about tonight is how do you hedge against a future where we may not have access to the latest
65. 14:03models and it may not matter It may not matter if we were to stop AI progress today. Say, say they never released Fable, they
66. 14:15never released Mythos, and then we just take what we have today, Opus 4.8, Open AI 5.5. We we don't even need 5.6.
67. 14:25[ __ ] them. We just start at at that level and we take any open source models that are at that level. We are we probably have a decade
68. 14:41individually as companies and as society before we can even begin to scratch the surface of what the current technology has made possible.
69. 14:54Like not even close. I think what this slowdown probably does is is keeps people employed for a bit longer. Um,
70. 15:09I think it probably quells some of the AI anti-AI rhetoric, but there's a huge opportunity for us to individually reclaim our agency
71. 15:22and get smart about how we can navigate this. That's what I'm gonna be talking about tonight. Whatever. I'm just here for [ __ ] them.
72. 15:31[ __ ] them. [ __ ] him. [music] [music]
73. 15:45[music] All right. So, four minutes. We're going live over in the AI salon. Let's see. Some of us there going to be
74. 16:00millions of people have no idea how to even begin to leverage this. Yes. Correct. This is right now today. Right now today
75. 16:11I created if you go to my LinkedIn and and search for the seven economies um I created a a graphic for this. But
76. 16:20the idea is that we've got AI accelerating and let's let's just say it slows down. Let's say there's an artificial plateau that happens,
77. 16:29but there's been so much advancement in the past two years, past year, and the majority of people still don't
78. 16:37know what AI is. They don't use it. They don't want to use it. They don't haven't heard of it. And so, you've got some
79. 16:44people that are out here on the the bleeding edge of building agentic businesses. And then you've got some people who are
80. 16:52not using AI at all. And you've got a bunch of different levels in between of businesses and individuals who know a little bit about
81. 17:00AI, nothing about AI, smidge, a lot, a whole lot. It's going to take years to just catch up the 80% of people who are not paying
82. 17:12attention to this to what's here, what's here today. And and the government shenanigans with anthropic, that's at some point that'll
83. 17:21work out. They'll figure it out. They'll figure something out. They'll give us dribbs and drabs. What? It doesn't
84. 17:26[ __ ] matter. Doesn't matter. But independent of that, there is so much to learn about what's in our hands today. It's it's insane.
85. 17:37It's remarkable. Narrowed to broad super intelligence on specific test domains, massive capability jump from engineering, data,
86. 17:47and optimization. Have those same people been caught up on what computers can do or how they can leverage basic PC? No, they haven't.
87. 17:56That's it's one of the reasons the AI salon exists. So, we're going to be going live in two minutes in the AI
88. 18:01salon. So, jump over there. Bring these questions. Come on over. So, so if you're on Tik Tok or X, go to my
89. 18:08profile, click on the link. If you're on LinkedIn or YouTube, go to the description. The link's down there. I'll
90. 18:14see you over there in two minutes. Let's keep this conversation going in the uh meltdown Monday. All right, here we go.
91. 18:24Peace out. I'll see you over there.