[0] https://en.wikipedia.org/wiki/Normalization_of_deviance

As someone who was the technical co-founder of a company before AI, I really feel like a lot of that role can be done by AI, if the foundation is laid appropriately.

Or, maybe I'm wrong, but my understanding is: MoE is just an architecture to keep the activated weights smaller per token. The experts get routed basically token-by-token, and the "experts" themselves don't have a semantic domain so the "expert" word was maybe a poor choice.

I think this also misses the point. The precedent here almost surely implies that it will be illegal to use these frontier models as well.

I can see a future where weights are distributed on the darkweb or bittorrent, or people are trying to use small fly by night hosts of models.

But if this says these models are dangerous and the companies and people can't be trusted with them, then I don't see why that wouldn't also apply to open weight models.

Beyond that, I just started Captive's War, written by the people behind The Expanse, which I adore, and it's looking similarly good (similar to The Expanse; not LotR. I think it will be hard for things to match LotR for me).

Side note, but I'm sort of surprised that this "level" of judge (I think there's almost 700 of them in the country) is able to block these orders. It seems like almost no executive order is possible if you need a unanimous agreement of 700 people.

That said, the foundational models they talk about running on it - is that something they've trained themselves? I know they had some sort of deal with Google; could it be Gemini weights loaded into their private compute or something?

The only real answer is to make a bunch of pancakes and get feedback from the people eating them.

My point is that with AI, where the actual code generation is easy, there's little value in community PR contributions anymore.

I studied math at MIT and have several friends who are professors now and they deal with cranks all the time and since they're very kind and conflict averse people they tend to respond with perfunctory emails when they get inbounds like that.

So just be wary. Your external validation may not be as strong as you think it is, though kudos to you for at least trying to step out of the AI vortex to attempt to ground yourself.

That said, I would love to know how the state of what's in v1.20 compares to un-spec'ed dialyzer. I was under the impression that dyalizer's "success typing" approach (not flagging a function if there are some combination of parameters such that it works, rather than flagging it if some combination of parameters can make it fail) was like what Elixir is doing here, and I haven't found dialyzer terribly useful.