3618 karmaJoined Dec 2017Working (0-5 years)Oakland, CA, USAcullenokeefe.com


I am a lawyer and policy researcher interested in improving the governance of artificial intelligence. In May 2019, I received a J.D. cum laude from Harvard Law School. I currently work as a Research Scientist in Governance at OpenAI.

I am also a Research Affiliate with the Centre for the Governance of AI at the Future of Humanity Institute; Founding Advisor and Research Affiliate at the Legal Priorities Project; and a VP at the O’Keefe Family Foundation.

My research focuses on the law, policy, and governance of advanced artificial intelligence.

You can share anonymous feedback with me here.


Law-Following AI
AI Benefits


Topic Contributions

Narrow point: my understanding is that, per his own claims, the Manifund grant would only fund technical upkeep of the blog, and that none of it is net income to him.

Answer by CullenJul 28, 20233

How probable does he think it is that some UAP observed on Earth are aliens? :-)

Super excited about the artificial conscience paper. I'd note that a similar approach be very useful for creating law-following AIs:

An LFAI system does not need to store all knowledge regarding the set of laws that it is trained to follow. More likely, the practical way to create such a system would be to make the system capable of recognizing when it faces sufficient legal uncertainty,[10] then seeking evaluation from a legal expert system ("Counselor").[11]

The Counselor could be a human lawyer, but in the long-run is probably most robust and efficient if (at least partially) automated. The Counselor would then render advice on the pure basis of idealized legality: the probability and expected legal downsides that would result from an idealized legal dispute regarding the action if everyone knew all the relevant facts.

Utilitarianism is much more explicit in its maximisation than most ideologies, plus it (at least superficially) actively undermines the normal safeguards against dangerous maximisation (virtues, the law, and moral rules) by pointing out these can be overridden for the greater good.

Like yes there are extreme environmentalists and that's bad, but normally when someone takes on an ideology like environmentalism, they don't also explicitly & automatically say that the environmental is all that matters and that it's in principle permissible to cheat & lie in order to benefit the environment.

I think it's true that utilitarianism is more maximizing than the median ideology. But I think a lot of other ideologies are minimizing in a way that creates equal pathologies in practice. E.g., deontological philosophies are often about minimizing rights violations, which can be used to justify pretty extreme (and bad) measures.

I would be very curious for Gregory's take on whether he thinks EAs are too epistemically immodest still!

On the Democratic side, challenging Biden is a way to make yourself Very Unpopular with party elites. Challenging Harris, if she is his chosen successor, would be That But Worse.

This seems very wrong to me. Harris is very unpopular.

Thanks, this is a meaningful update for me.

it doesn't seem like a big leap to think that confidence in an ideology that says you need to maximise a single value to the exclusion of all else could lead to dangerously optimizing behaviour.

I don't find this a persuasive reason to think that utilitarianism is more likely to lead to this sort of behavior than pretty much any other ideology. I think a huge number of (maybe all?) ideologies imply that maximizing the good as defined by that ideology is the best thing to do, and that considerations outside of that ideology have very little weight. You see this behavior with many theists, Marxists, social justice advocates, etc. etc.

My general view is that there are a lot of people like SBF who have a lot of power-seeking and related traits—including callous disregard for law, social norms, and moral uncertainty—and that some of them use moral language to justify their actions. But I don't think utilitarianism is especially vulnerable to this, nor do I think it would be a good counterargument if it was. If utilitarianism is true, or directionally true, it seems good to have people identify as such, but we should definitely judge harshly those that take an extremely cavalier attitude towards morality on the basis of their conviction in one moral philosophy. Moral uncertainty and all that.

Load more