LessWrong 2.0 Reader

My point here is that at the capability level of GPT4, this distinction isn't very important. There's no way to know for sure until it is too late, of course, but it seems pretty plausible that GPT4 isn't cleverly scheming. It is merely human-level at deception, and doesn't pursue any coherent overarching goal with it. It clumsily muddles through with mildly-above-average-for-human convincingness. For most queries (it seems plausible to me) it isn't even adequately coherent to make a crisp distinction between whether it's honestly trying to answer the question vs deceptively trying to make an answer look good; at its level of capability, it's mostly the same thing one way or the other. The exceptions to this "mostly" aren't strategic enough that we expect them to route around obstacles cleverly.

It isn't much, but it is more than I naively expected.

FYI, my update from this comment was:

• Hmm, seems like a decent argument...
• ... except he said "we don't know that it doesn't work", which is an extremely strong update that it will clearly not work.

I think the crux is I think that the important parts of of LLMs re safety isn't their safety properties specifically, but rather the evidence they give to what alignment-relevant properties future AIs have 

[insert standard skepticism about these sorts of generalizations when generalizing to superintelligence]

But what lesson do you think you can generalize, and why do you think you can generalize that?

I think this is a crux, in that I don't buy o1 as progressing to a regime where we lose so much dense feedback that it's alignment relevant, because I think sparse-feedback RL will almost certainly be super-uncompetitive with every other AI architecture until well after AI automates all alignment research.

So, as a speculative example, further along in the direction of o1 you could have something like MCTS help train these things to solve very difficult math problems, with the sparse feedback being given for complete formal proofs.

Similarly, playing text-based video games, with the sparse feedback given for winning.

Similarly, training CoT to reason about code, with sparse feedback given for predictions of the code output.

Etc.

You think these sorts of things just won't work well enough to be relevant?

Most people don't have very fixed ideas about how much a 28% overall increase in bankruptcy happens to be. 

If you would ask most people without a libertarian outlook to rank different factors that lead to an increase in bankruptcy, I would not expect them to be able to accurately compare them and find that sports online gambling only will have such a strong influence. 

I'm aware and I don't disagree. However, in xrisk, many (not all) of those who are most worried are also most bullish about capabilities. Reversely, many (not all) who are not worried are unimpressed with capabilities. Being aware of the concept of AGI, that it may be coming soon, and of how impactful it could be, is in practice often a first step towards becoming concerned about the risks, too. This is not true for everyone unfortunately. Still, I would say that at least for our chances to get an international treaty passed, it is perhaps hopeful that the power of AGI is on the radar of leading politicians (although this may also increase risk through other paths).

The usual plan for control as I understand it is that you use control techniques [LW · GW] to ensure the safety of models that are sufficiently good at themselves doing alignment research that you can then leverage your controlled human-ish-level models to help you align future superhuman models.

I would add that convincing Musk to take action against Altman is the highest ROI thing I can think of in terms of decreasing AI extinction risk.

I would expect, the issue isn't about convincing Musk to take action but about finding effective actions that Musk could take. 

I do get that point that you are making, but I think this is a little bit unfair to these organizations. Articles like Machines of Loving Grace, The Intelligence Age and Planning for AGI and Beyond are implicit public justifications for building AGI.

These labs have also released their plans on "safe development". I expect a big part of what they say to be the usual business marketing, but it's not like they completely ignoring the issue. In fact, taking one example, Anthropic's research papers on safety are often discussed on this site as genuine improvements on this or that niche of AI Safety.

I don't think that money alone would've convinced CEOs of big companies to run this enterprise. Altman and Amodei, they both have families. If they don't care about their own families, then they at least care about themselves. After all, we are talking about scenarios where these guys would die the same deaths as the rest of us. No amounts of hoarded money would save them. They would have little motivation to do any of this if they believed that they would die as the result of their own actions. And that's not mentioning all of the other researchers working at their labs. Just Anthropic and OpenAI together have almost 2000 employees. Do they all not care about their and their families' well-being?

I think the point about them not engaging with critics is also a bit too harsh. Here [LW · GW] is DeepMind's alignment team response to concerns raised by Yudkowski. I am not saying that their response is flawless or even correct, but it is a response nonetheless. They are engaging with this work. DeepMind's alignment team also seemed to engage with concerns raised by critics in their (relatively) recent work [AF · GW].

EDIT: Another example would be Anthropic creating a dedicated team [LW · GW] for stress testing their alignment proposals. And as far as I can see, this team is lead by someone who has been actively engaged with the topic of AI safety on LessWrong, someone who you sort of praised [LW · GW] a few days ago.

It's not clear to me that there was actually an option to build a $100B company with competent people around the world who would've been united in conditionally shutting down and unconditionally pushing for regulation. I don't know that the culture and concepts of people who do a lot of this work in the business world would allow for such a plan to be actively worked on.

This elides the original argument by assuming the conclusion: that countermanding efforts remain cheap relative to the innovations. But the whole point is that significant shifts in costs associated with defense of a certain level can change behaviors and which plans and supply chains are economically defensible a lot.