Epistemic Note:
Many highly respected community members with substantially greater decision making experience (and Lesswrong karma) presumably disagree strongly with my conclusion.
Premise 1:
It is becoming increasingly clear that OpenAI is not appropriately prioritizing safety over advancing capabilities research.
Premise 2:
This was the default outcome.
Instances in history in which private companies (or any individual humans) have intentionally turned down huge profits and power are the exception, not the rule.
Premise 3:...
I don't know the answer, but it would be fun to have a twitter comment with a zillion likes asking Sam Altman this question. Maybe someone should make one?
I promise I won't just continue to re-post a bunch of papers, but this one seems relevant to many around these parts. In particular @Elizabeth (also, sorry if you dislike being at-ed like that).
...Food preferences significantly influence dietary choices, yet understanding natural dietary patterns in populations remains limited. Here we identifiy four dietary subtypes by applying data-driven approaches to food-liking data from 181,990 UK Biobank pa
Regarding the situation at OpenAI, I think it's important to keep a few historical facts in mind:
Requisite resource levels: The project must have adequate resources to compete at the frontier of AGI development, including whatever mix of computational resources, intellectual labor, and closed insights are required to produce a 1+ year lead over less cautious competing projects.
A Theory of Usable Information Under Computational Constraints
...We propose a new framework for reasoning about information in complex systems. Our foundation is based on a variational extension of Shannon's information theory that takes into account the modeling power and computational constraints of the observer. The resulting \emph{predictive V-information} encompasses mutual information and other notions of informativeness such as the coefficient of determination. Unlike Shannon's mutual information and in violation of the data processing inequality, V-
My reading is their definition of conditional predictive entropy is the naive generalization of Shannon's conditional entropy given that the way that you condition on some data is restricted to only being able to implement functions of a particular class. And the corresponding generalization of mutual information becomes a measure of how much more predictable does some piece of information become (Y) given evidence (X) compared to no evidence.
For example, the goal of public key cryptography cannot be to make the mutual information between a plaintext, and ...
My timelines are lengthening.
I've long been a skeptic of scaling LLMs to AGI *. To me I fundamentally don't understand how this is even possible. It must be said that very smart people give this view credence. davidad, dmurfet. on the other side are vanessa kosoy and steven byrnes. When pushed proponents don't actually defend the position that a large enough transformer will create nanotech or even obsolete their job. They usually mumble something about scaffolding.
I won't get into this debate here but I do want to note that my timelines have lengthe...
There are two kinds of relevant hypothetical innovations: those that enable chatbot-led autonomous research, and those that enable superintelligence. It's plausible that there is no need for (more of) the former, so that mere scaling through human efforts will lead to such chatbots in a few years regardless. (I think it's essentially inevitable that there is currently enough compute that with appropriate innovations we can get such autonomous human-scale-genius chatbots, but it's unclear if these innovations are necessary or easy to discover.) If autonomou...
I'm surprised at people who seem to be updating only now about OpenAI being very irresponsible, rather than updating when they created a giant public competitive market for chatbots (which contains plenty of labs that don't care about alignment at all), thereby reducing how long everyone has to solve alignment. I still parse that move as devastating the commons in order to make a quick buck.
I disagree. This whole saga has introduced the Effective Altruism movement to people at labs that hadn't thought about alignment.
From my understanding openai isn't anywhere close to breaking even from chatgpt and I can't think of any way a chatbot could actually be monetized.
Sometimes I forget to take a dose of methylphenidate. As my previous dose fades away, I start to feel much worse than baseline. I then think "Oh no, I'm feeling so bad, I will not be able to work at all."
But then I remember that I forgot to take a dose of methylphenidate and instantly I feel a lot better.
Usually, one of the worst things when I'm feeling down is that I don't know why. But now, I'm in this very peculiar situation where putting or not putting some particular object into my mouth is the actual cause. It's hard to imagine something more tangibl...
Wait, some of y'all were still holding your breaths for OpenAI to be net-positive in solving alignment?
After the whole "initially having to be reminded alignment is A Thing"? And going back on its word to go for-profit? And spinning up a weird and opaque corporate structure? And people being worried about Altman being power-seeking? And everything to do with the OAI board debacle? And OAI Very Seriously proposing what (still) looks to me to be like a souped-up version of Baby Alignment Researcher's Master Plan B (where A involves solving physics and C invo...
My current perspective is that criticism of AGI labs is an under-incentivized public good. I suspect there's a disproportionate amount of value that people could have by evaluating lab plans, publicly criticizing labs when they break commitments or make poor arguments, talking to journalists/policymakers about their concerns, etc.
Some quick thoughts:
RE 1& 2:
Agreed— my main point here is that the marketplace of ideas undervalues criticism.
I think one perspective could be “we should all just aim to do objective truth-seeking”, and as stated I agree with it.
The main issue with that frame, imo, is that it’s very easy to forget that the epistemic environment can be tilted in favor of certain perspectives.
EG I think it can be useful for “objective truth-seeking efforts” to be aware of some of the culture/status games that underincentivize criticism of labs & amplify lab-friendly perspectives.
RE 3:
Go...
Idea: Daniel Kokotajlo probably lost quite a bit of money by not signing an OpenAI NDA before leaving, which I consider a public service at this point. Could some of the funders of the AI safety landscape give some money or social reward for this?
I guess reimbursing everything Daniel lost might be a bit too much for funders but providing some money, both to reward the act and incentivize future safety people to not sign NDAs would have a very high value.
Sure, I'll try to post here if I know of a clear opportunity to donate to either.
I worked at OpenAI for three years, from 2021-2024 on the Alignment team, which eventually became the Superalignment team. I worked on scalable oversight, part of the team developing critiques as a technique for using language models to spot mistakes in other language models. I then worked to refine an idea from Nick Cammarata into a method for using language model to generate explanations for features in language models. I was then promoted to managing a team of 4 people which worked on trying to understand language model features in context, leading to t...
They would not know if others have signed the SAME NDAs without trading information about their own NDAs, which is forbidden.
From my perspective, the only thing that keeps the OpenAI situation from being all kinds of terrible is that I continue to think they're not close to human-level AGI, so it probably doesn't matter all that much.
This is also my take on AI doom in general; my P(doom|AGI soon) is quite high (>50% for sure), but my P(AGI soon) is low. In fact it decreased in the last 12 months.
Apparently[1] there was recently some discussion of Survival Instinct in Offline Reinforcement Learning (NeurIPS 2023). The results are very interesting:
...On many benchmark datasets, offline RL can produce well-performing and safe policies even when trained with "wrong" reward labels, such as those that are zero everywhere or are negatives of the true rewards. This phenomenon cannot be easily explained by offline RL's return maximization objective. Moreover, it gives offline RL a degree of robustness that is uncharacteristic of its online RL count
Because future rewards are discounted
Don't you mean future values? Also, AFAICT, the only thing going on here that seperates online from offline RL is that offline RL algorithms shape the initial value function to give conservative behaviour. And so you get conservative behaviour.
Several dozen people now presumably have Lumina in their mouths. Can we not simply crowdsource some assays of their saliva? I would chip money in to this. Key questions around ethanol levels, aldehyde levels, antibacterial levels, and whether the organism itself stays colonized at useful levels.
Any recommendations on how I should do that? You may assume that I know what a gas chromatograph is and what a Petri dish is and why you might want to use either or both of those for data collection, but not that I have any idea of how to most cost-effectively access either one as some rando who doesn't even have a MA in Chemistry.
Some philosophy is rubbish. Quite a lot, I believe. And with a statement such as "perceptions are caused by things external to the perceptions themselves", which I find unremarkable in itself as a prima facie obvious hypothesis to run with, there is a tendency for philosophers to go off the rails immediately by inventing precise definitions of words such as "perceptions", "are", and "caused", and elaborating all manner of quibbles and paradoxes. Hence the whole tedious catalogue of realisms.
Science did not get anywhere by speculating on whether there are four or five elements and arguing about their natures.
On an apparent missing mood - FOMO on all the vast amounts of automated AI safety R&D that could (almost already) be produced safely
Automated AI safety R&D could results in vast amounts of work produced quickly. E.g. from Some thoughts on automating alignment research (under certain assumptions detailed in the post):
each month of lead that the leader started out with would correspond to 15,000 human researchers working for 15 months.
Despite this promise, we seem not to have much knowledge when such automated AI safety R&D might happ...
Intuitively, I'm thinking of all this as something like a race between [capabilities enabling] safety and [capabilities enabling dangerous] capabilities (related: https://aligned.substack.com/i/139945470/targeting-ooms-superhuman-models); so from this perspective, maintaining as large a safety buffer as possible (especially if not x-risky) seems great. There could also be something like a natural endpoint to this 'race', corresponding to being able to automate all human-level AI safety R&D safely (and then using this to produce a scalable solution to a...
This seems incredibly interesting to me. Googling “White-boarding techniques” only gives me results about digitally shared idea spaces. Is this what you’re referring to? I’d love to hear more on this topic.
Unfortunately, it looks like non-disparagement clauses aren't unheard of in general releases:
Release Agreements commonly include a “non-disparagement” clause – in which the employee agrees not to disparage “the Company.”
https://joshmcguirelaw.com/civil-litigation/adventures-in-lazy-lawyering-the-broad-general-release
...The release had a very broad definition of the company (including officers, directors, shareholders, etc.), but a fairly reas
AI labs are starting to build AIs with capabilities that are hard for humans to oversee, such as answering questions based on large contexts (1M+ tokens), but they are still not deploying "scalable oversight" techniques such as IDA and Debate. (Gemini 1.5 report says RLHF was used.) Is this more good news or bad news?
Good: Perhaps RLHF is still working well enough, meaning that the resulting AI is following human preferences even out of training distribution. In other words, they probably did RLHF on large contexts in narrow distributions, with human rater...
Bad: AI developers haven't taken alignment seriously enough to have invested enough in scalable oversight, and/or those techniques are unworkable or too costly, causing them to be unavailable.
Turns out at least one scalable alignment team has been struggling for resources. From Jan Leike (formerly co-head of Superalignment at OpenAI):
Over the past few months my team has been sailing against the wind. Sometimes we were struggling for compute and it was getting harder and harder to get this crucial research done.
Even worse, apparently the whole Supera...
Our current big stupid: not preparing for 40% agreement
Epistemic status: lukewarm take from the gut (not brain) that feels rightish
The "Big Stupid" of the AI doomers 2013-2023 was AI nerds' solution to the problem "How do we stop people from building dangerous AIs?" was "research how to build AIs". Methods normal people would consider to stop people from building dangerous AIs, like asking governments to make it illegal to build dangerous AIs, were considered gauche. When the public turned out to be somewhat receptive to the idea of regulating ... (read more)
I think this is the sort of conversation we should be having! [Side note: I think restricting compute is more effective than restricting research because you don't need 100% buy in.
The analogy to nuclear weapons is, I think, a good one. The science behind nuclear weapons is well known -- what keeps them fro... (read more)