Noam Brown reveals OpenAI is already watching chain-of-thought monitorability degrade as models get better at controlling what they show "And this is one major concern, and we're already seeing signs that chain of thought monitorability is degrading, for various reasons." "We're trying to figure out exactly why, because we want to reverse the trend. But we're seeing that the model is becoming better able at controlling its chain of thought." "So this is a problem, because you could have a situation where the model understands what chain of thought is and that people are observing it." "And eventually they will, because this is all in the pre-training data. The idea of chain of thought monitoring has been around long enough that it's in the pre-training data, they're aware of it, but they're not actually able to control their chains of thought." "If we reach a point where they're actually able to recognize, "Oh, I am being observed, I want to think these bad thoughts in a way that is not observable to my monitors," and then they're able to actually do that, then there's a problem." _________ More key quotes from OpenAI's safety related conversations:

I’ve been frustrated by the discourse around the Nadeau wealth tax because DC already taxes ordinary and capital income the same. The explanation that this replicates a federal tax is bad because capital income is taxed at a much lower rate federally.

Remarkable how people will just make up things like, “Europe doesn’t have data centers because they can’t securitize them.” Lots of American data center builders are active in Europe. Fluidstack pulled out of France and OAI out of Britain because permitting and time-to-power were too slow. In places where those problems don’t exist — i.e. Scandinavia — lots of stuff is getting built. European data center builders will complain about “financing,” but what they mean is that, under the prevailing political, permitting, and energy environment, no one wants to buy the damn things!

This headline doesn’t really seem to match the reporting, which is just that people spanning the entire Cuellar to Jayapal spectrum are rightly nervous about Leading The Future’s money.

I think the dilemma for Hasan Piker is obvious. He wants to be a rich twitch streamer, and he wants to be a electoral progressive kingmaker. But they are ultimately mutually exclusive paths, and he needs to choose one.

I keep saying this but if you work at OpenAI and have concerns about AI risk you need to understand that your company’s superpac is one of the major proximate barriers to doing anything.

I think in most cases, Democrats should talk about how Trump's blundering foreign policy and corrupt indulgence of the worst actors in the AI industry is driving up the cost of living. Does anyone actually disagree with this or do they just like being mad at me?

Sometimes when industry insiders warn that a new, potentially very lucrative product ideas is actually dangerous and bad the doomsayers are correct and the optimists are wrong.

I think in most cases, Democrats should talk about how Trump's blundering foreign policy and corrupt indulgence of the worst actors in the AI industry is driving up the cost of living. Does anyone actually disagree with this or do they just like being mad at me?

It's interesting that almost all the big abundance people also have AI safety concerns, almost as if both movements appeal to thoughtful intelligent people concerned about evidence and good arguments and aren't reducible to lazy stereotypes.