We are discussing the debate statement: "On the margin[1], it is better to work on reducing the chance of our[2] extinction than increasing the value of futures where we survive[3]". You can find more information in this post.
When you vote and comment on the debate week banner, your comment will also appear here, along with a note indicating your initial vote, and your most recent vote (if your opinion has changed).
However, you can also comment here any time throughout the week. Use this thread to respond to other people's arguments, and develop your own.
If there are a lot of comments - consider sorting by “New” and interacting with posts that haven’t been voted or commented on yet.
Also - perhaps don’t vote karma below zero for low effort submissions, we don’t want to discourage low effort takes on the banner.
- ^
‘on the margin’ = think about where we would get the most value out of directing the next indifferent talented person, or indifferent funder.
- ^
‘our’ and 'we' = earth-originating intelligent life (i.e. we aren’t just talking about humans because most of the value in expected futures is probably in worlds where digital minds matter morally and are flourishing)
- ^
Through means other than extinction risk reduction.
The gap between weak AGI and strong AGI/ASI timeline predictions seems to have ticked up a bit. It doesn't seem like the intra-token reasoning/capabilities is scaling as hard as I'd previously feared. The models themselves are not getting so scarily capable and agentic in each forward pass, instead we are increasingly eliciting those capabilities/agency in context with the models remaining myopic and largely indifferent.
If the new paradigm holds with a significant focus on scaling inference it seems to both be less aggressive (in terms of scaling intelligence) and more conducive to 'passing' safety.
The current paradigm likely places a much lower burden on hard interpretability than I expected ~1 year ago, it feels much more like a verification problem than a full solve. With current rates of interpretability progress (and AI accelerating safety ~inline with capabilities) we could actually be able to verify that a CoT is faithful and legible and that might be ~sufficient.
Agreed, I still think there's a reasonable chance that ML research does fall within the set of capabilities that quickly reach superhuman levels and foom is still on the cards, also more RL in general is just inherently quite scary.
The 9s of safety makes sense from a control perspective but I think there's another angle, which is the possibility of a model that is aligned-enough to actually not want to pursue human extinction.
Potentially, but I think there's still room for scenarios where humans are broadly disempowered yet not extinct - worlds where we get a passing grade on safety. Where we effectively avoid strongly-agentic systems and achieve sufficient alignment such that human lives are valued, but fall short of the full fidelity necessary for a flourishing future.
Still this point has updated me slightly, I've reduced my disagreement.
My model looks something like this:
There are a bunch of increasingly hard questions on the Alignment Test. We need to get enough of the core questions right to avoid the ASI -> everyone quickly dies scenario. This is the 'passing grade'. There are some bonus/extra credit questions that we need to also get right to get an A (a flourishing future).
We don't know exactly which questions will be included or in which section. We also don't know the thresholds for these grades and we are (rightly) focusing the vast majority of our efforts on the expected fundamental questions to maximise our chance of the passing grade.
Relatively to ~1 year ago the 'passing grade' for alignment feels a bit easier and we've got a bit more study time. I've also become aware of just how much more difficult the A grade might be and that a pass might not be very valuable at all - I don't think anything has changed there, I was just somewhat ignorant of risks from gradual disempowerment.
It might make sense to dedicate say 5-20% of our effort to study for questions we expect in the bonus/extra credit section. I think we currently do less than that (perhaps 1-5%). So I think the vast majority of the effort should be spent on avoiding extinction, but I'm less sure about effort at the margin.