Coffee With Scott Adams — Knowledge Archive May 24, 2026
Scott Adams Philosophy Archive
Search ideas
Episodes Episode #2945 Segments
NewsReaction Politics as Persuasion

Back to episode — Episode 2945 CWSA 09/01/25

Context —

en essential for China to have a solar plant that's the size of Chicago. Are we going to do that? I feel like we're going to go hard at nuclear, but maybe solar is faster. You could probably put up a solar outdoor facility in say five years with all the permitting and whatnot, but how long would it take you to build a nuclear power plant? Longer than five years is my bet. As you know, there's a m…

← Previous segment →

o you think of the idea of introducing feral cats to deal with the rat problem in New York City?

Uh-huh. Uh-huh. Now, if you don't understand purr talk, that's how cats communicate. I'll translate it for you. Say more. Uh-huh. Uh-huh. Okay.

All right. It turns out Gary is a big fan of Curtis Sliwa and absolutely supports the idea of cats solving all of our problems. Now, you might ask, Scott, how many problems can cats solve? I don't know, but I feel like it could be all of them. I mean, they could end wars. They could make you less lonely. Yeah, they could keep marriages together. I think cats could do all of that.

Meanwhile, in California, the California Highway Patrol is going to team up with local law enforcement to sort of surge against crime. Now, that seems like a reasonably good idea, but do you think that California would do that if Trump had not put the pressure on, you know, in DC and talking about Chicago and talking about going into California? I don't think so.

So while this looks like a win for Newsom because it makes it look like he's dealing with a crime problem, I feel like that's not the message I'm getting. The message I'm getting is that he wasn't going to deal with a crime problem until Trump embarrassed him publicly. Is that what you see? Or do you see this go-getter governor who's all over this problem and he recognizes what the regular people are feeling about crime and so he's activated his resources? I don't feel that at all. I feel like the only way this would have happened and it's probably a good thing is because Trump embarrassed him.

I'm going to give the win to Trump if this works, right? I don't think that's unreasonable to say that this would be a Trump success if Newsom is successful because he wouldn't have done it. I mean, I'm not a mind reader, but really, did you see any movement in this direction? No, not until Trump made it a very big thing.

Here's a story from Ars Technica. Samuel Axon is writing about this. Apparently Microsoft, which as you know is in this deep partnership with OpenAI and uses OpenAI as its own AI as well as what it's producing, weirdly Microsoft separate from OpenAI is developing AI. So does that signal that there's something wrong with the partnership? Did somebody not foresee some problem happening?

Now they're trying to pass it off as these are more specifically trained AIs that would be a little bit more powerful than OpenAI would be because that's more of a general AI. To which I say, really, you couldn't just train the general model which you have some degree of control over? You couldn't just make OpenAI know how to do the specific AI things as well? I don't know. Do you need your own AI for other stuff that OpenAI can't do? I don't know. I don't know.

So I guess what I'm s

Context —

uggesting is there might be this might be telling us something, but I don't know what. So it might be telling us that OpenAI doesn't have a future. Now I'm not seeing evidence of that specifically, but why would they be building alternative models when they have the dominant AI model in the world and it doesn't make sense? So I'm not buying the story of why they have multiple AIs, but it does make…

Next segment → →