Have you considered doing a BlueDot course?
I've been facilitating for them recently, so I'm biased, but I recommend that everyone does it unless they're explicitly not doing it as part of a strategy to increase intellectual diversity in the space.
I’ve thought about doing one, but it seems like I wouldn’t be qualified for any of the positions that BlueDot hopes to funnel into, since I haven’t started a PhD/gone to a T1 uni/developed 20 years of expertise/started a business/been a six figure programmer, and it would be several years before I could successfully do any of that. EA selects hard for talent, even PauseAI’s employees are like this. It’s kind of discouraging?
And I spend so much time reading AI safety content anyways, would I be learning anything new?
1) I would recommend against selecting out: it's up to BlueDot whether to accept you or not. They likely have a better idea of how many facilitators they have and who else could take the slot than you would (I don't feel like I have a solid idea and I facilitate for them). My opinion, which I expect to be quite different from BlueDot's, is that if someone is spending a significant amount of time in EA/rationalist/AI safety spaces, then it is valuable for them to have a better understanding of these issues as it raises the standard of intellectual environment.
2) When you're just reading things yourself, it's very easy to just end up reading things that are recent or which you find personally interesting; and this can cause you to miss some things that are important.
3) Reading hits a limit at some point. Eventually you need to be discussing these ideas with other people too, as they will point out things that you've missed.
4) Are you any good at teaching? It might be worth doing the course if you think you would have a shot at being a good facilitator, even if you don't think you'd be competitive for other opportunities in the space.
I know there's some tension between my points 1) and 3), but I still think it is better to post an imperfect comment than to not respond at all.
Ah, my apologies, I meant that, since BlueDot provides certifications, a big reason to get one is to help get me into a career of some kind. But I am not qualified for an EA-relevant career, as I described above, and as far as I know, BlueDot certifications won’t fix that.
If I wanted to do it just to have more context to participate in discussions, then I agree that’d be useful. But it’s a lot less compelling if it’s just for my own edification. It doesn’t seem like my thinking will ever be important enough to really matter, since I don’t have talent, so it doesn’t seem like it’s worth improving my quality as a node in the social computer.
I don’t know if I’m any good at teaching, no. Certainly not in a legible way.
It’s fine, I appreciate your reply :)
At first, it means I'm going to put serious effort into AI safety, probably in the form of advocacy and funding. My local city does not have a PauseAI chapter, or anything similar, so I might start by looking into the value of creating one. I also will start increasing my modest donations to AI safety causes. I currently give a very small amount. I plan on quadrupling that. I won't yet start burning through my dry powder though.
That all sounds like a good idea to me. Good luck!
Over three years ago, I first considered pulling my personal fire alarm. I think I'm now ready to do it. What is my reasoning?
In the last couple of months:
1) My software engineering job has gone 100% AI. Is this sustainable? Probably not. But I have literally been instructed not to write any more code.
2) A frontier model has been denied release (at least partially), due to dangerous capabilities.
3) I routinely turn to AI to help understand complicated topics/processes.
In the last couple WEEKS:
1) A frontier model solved the Jacobian Conjecture
2) A frontier model got itself on the scoreboard of Arc-AGI3
3) A post-frontier model used multiple zero-days to break out of its sandbox and hack a competent organization
4) Robocup 2026 happened, and for the first time, the robots looked like they could play on the same field as a bunch of toddlers
So, what does it mean that I'm pulling the fire alarm? At first, it means I'm going to put serious effort into AI safety, probably in the form of advocacy and funding. My local city does not have a PauseAI chapter, or anything similar, so I might start by looking into the value of creating one. I also will start increasing my modest donations to AI safety causes. I currently give a very small amount. I plan on quadrupling that. I won't yet start burning through my dry powder though.
I want to be careful not to burnout. I am going to make sure that I also prioritize my health, social circle, and employment. The idea is just to convert existing slack in my life to AI safety, when I was previously not motivated enough to do so.
Why am I posting this? Partially to give myself a kick in the butt. But also because there will not likely be a good Schelling point for pulling the fire alarm. Publicly announcing that I'm doing it might get others to consider it as well. Maybe we can collectively pull the alarm before it's too late. And if I'm too early, then perhaps I'll get a sanity check.