Ben Pace

Karma: 41,507

I’m an admin of LessWrong. Here are a few things about me.

I generally feel more hopeful about a situation when I understand it better.
I have signed no contracts nor made any agreements whose existence I cannot mention.
I believe it is good to take responsibility for accurately and honestly informing people of what you believe in all conversations; and also good to cultivate an active recklessness for the social consequences of doing so.
It is deontologically prohibited to directly work on causing the end of the world. Even if you are fatalistic about what is going to happen.

Randomly: If you ever want to talk to me about anything you like for an hour, I am happy to be paid $1k for an hour of doing that.

(Longer bio.)

Ben Pace 12 Oct 2026 6:10 UTC
2 points
0
on: A desparate plea to be convinced that Roko’s Basilisk won’t torture me
If you make a principled decision not to fall for blackmail, then you don’t get blackmailed so much.
But in general I don’t trust people who tell me that they’re successfully acausally trading with foreign alien minds from other universes or the future. I remember a successful investor guy seriously telling me he believes in aliens who live in the 4th dimension and watch over our world and want us to love each other. He sure believed he knew a lot of obvious bullshit about alien minds. Why don’t you get yourself to believe that instead? It’s at least a nicer fantasy about you being able to talk with future alien minds than your current one, that gets you to behave better.

Ben Pace 11 Oct 2026 17:58 UTC
3 points
0
in reply to: Lukas_Gloor’s comment on: On Kink & Status in the Rationalist Community
“Act with courage and integrity but give weight to possible reputational costs when nothing of high moral value is at stake.”
I find my self-expression to have moral value, and I guess it sounds to me like you don’t.

Ben Pace 10 Oct 2026 21:05 UTC
4 points
2
in reply to: meemi’s comment on: Philip Harker’s Shortform
She says we should instead keep the goal in the periphery while doing things that feel alive and gently leaning in the direction we want to go. Her idea is that this might result in some radical lateral move that cracks AI safety comms.
I don’t think Aella’s model about the impact of plzdontkillus makes sense anymore given the data.^[1] Arguments for a lateral move, and the arguments for partying and creativity make more sense on priors, but not after 20% of the AI safety views comes from a straight shot on goal.
I don’t see how the top AI-safety related video intentionally being an AI-safety video distinguishes these two hypotheses. Are you thinking that the alternative is that videos that aren’t supposed to be about AI safety accidentally become about AI safety?
To me the hypothesis is that, if people can be creative however they wish, and are in a context with lots of AI safety people and arguments and ideas and encouragement, that they will sometimes have very creative AI safety ideas and run directly at them, and succeed. The fact that the best AI safety videos are about AI safety would be true under either hypothesis, so isn’t evidence for one over the other.
(Apparently the best AI safety video was made by someone who exclusively made AI safety videos? That is evidence for one hypothesis over the other.)

Ben Pace 10 Oct 2026 17:07 UTC
21 points
12
in reply to: Lukas_Gloor’s comment on: On Kink & Status in the Rationalist Community
Basically, when you do something that has a risk of reputational damage, do it in support of an actually-important moral principle and not for satisfying your personal sense of humor?
I get triggered by few things as much as people trying to shut down people having a sense of humor for respectability reasons. Feels like one of the Canaries that one’s humanity is not allowed here. If we can cut out jokes because of respectability, then we soon will have no jokes, because there’s so often possible concerns about seeming unserious or offending someone or being confusing.
I have the opposite heuristic, that if it’s done out of a sense of humor, then it’s fine, even if some others didn’t get the joke or many didn’t find it very funny.

Ben Pace 10 Oct 2026 6:15 UTC
12 points
10
in reply to: Trinley Goldenberg’s comment on: On Kink & Status in the Rationalist Community
I’m confused, neither of the things Aella described were getting mad about consent violations?

Ben Pace 9 Oct 2026 1:25 UTC
5 points
10
in reply to: DW11’s comment on: Zach Stein-Perlman’s Shortform
Seems totally normal to me? People in my organization often do drafts of emails for one another, and I expect many folks with lesser epistemic integrity would feel fine with the staff even sending those emails on their behalf.

Ben Pace 3 Oct 2026 18:33 UTC
7 points
7
in reply to: leogao’s comment on: leogao’s Shortform
OpenAI has been a terrible an untrustworthy actor many times over, and so hearing that they might be in a new area, my prior is “Yeah, sounds like them”; but I have not looked into it enough for it to be an additional update to my current prior.

Ben Pace 2 Oct 2026 20:18 UTC
3 points
0
in reply to: Alexander Gietelink Oldenziel’s comment on: Alexander Gietelink Oldenziel’s Shortform
Yep, put it in an LLM block and it won’t be filtered out. That’s the point of them!
Opus 4.6 explains it here in this post.

Ben Pace 2 Oct 2026 8:52 UTC
5 points
0
in reply to: roha’s comment on: Alexander Gietelink Oldenziel’s Shortform
Please put LLM content in the LLM content block! I was skimming this thread and got a rude surprise as I started to smell LLM, then saw the explanation at the top. Here’s what LLM content block looks like:
Just use the / key to open up the menu, and select LLM content block.

Ben Pace 1 Oct 2026 1:18 UTC
3 points
1
in reply to: leogao’s comment on: Wei Dai’s Shortform
I’m somewhat surprised about that, curious if anyone has a link to Leo saying that 5 years ago.

Ben Pace 23 Sep 2026 18:17 UTC
19 points
11
in reply to: jacquesthibs’s comment on: Jensen Huang Says If We Cannot Align AI, Shut Down the AI Labs
That seems plausible, but I still think it helps in communicating with people about the risk that this side is on the record saying that if you cannot align the AI, shut down, and is threatening legal consequences for any damages.

Jensen Huang Says If We Cannot Align AI, Shut Down the AI Labs

Ben Pace23 Sep 2026 16:15 UTC

194 points

13 comments3 min readLW link

Ben Pace 23 Sep 2026 6:30 UTC
4 points
0
in reply to: Austin Chen’s comment on: Elizabeth’s Shortform
I let EA out of my heart a long time ago. I felt the culture was rotting and its leadership abandoning it back in ~2018; I had forbid myself to even look at the EA Forum almost a year before the FTX crash, to help me separate from it. Looking at some of the threads there about this decision of yours this week has just been opening up old traumas, and I quickly closed them again. I would encourage you to focus your efforts elsewhere, I see little good that will come of that place.
I don’t relate to the AI safety community in an especially similar way. It is one of the actors in the play of what is going on, like the news media, or China; with the main difference being that I am more a part of it. I have more ability to steer it, and more responsibility for its impacts.
Not that you asked, but I thought I’d mention it anyway, in case you were interested.

Ben Pace 23 Sep 2026 5:56 UTC
6 points
0
in reply to: Austin Chen’s comment on: Elizabeth’s Shortform
I would say (possibly with the benefit of hindsight) that a piece of writing more shaped like this would be appropriate if you wanted your allies to come to think this was a good decision.
Both the inferential gap here and also the quality-of-the-decision-gap are both far more substantial than usual.

Ben Pace 23 Sep 2026 5:07 UTC
1 point
0
in reply to: Elizabeth’s comment on: Elizabeth’s Shortform
I (sadly) don’t have much time to spend on this thread; but I am here referring to communicating with your trusted allies, whereas I understand respectability politics to be about the signals you send to the broader public.

Ben Pace 23 Sep 2026 4:57 UTC
6 points
2
in reply to: Austin Chen’s comment on: Elizabeth’s Shortform
I mostly wish you had publicly argued more for your belief that a convicted criminal basically didn’t do the crimes. I suspect this would have led me to trust you more (though more likely for you to find out your evidence & arguments were weak).
Perhaps you have and I didn’t see it? But either way it seemed crazy to me to not say something in your announcement post like “I understand to many people I respect, and to many people who use and support Manifund and Mox that I hired someone who is a major convicted felon to a relevantly similar role to where they committed their crimes, and that this seems like a straightforwardly bad decision, but FYI I actually think that they largely didn’t do the crimes they confessed to, so it’s not actually that bad” and (for bonus points) some reasons why you think that.

Ben Pace 22 Sep 2026 15:58 UTC
5 points
0
in reply to: Richard_Ngo’s comment on: Elizabeth’s Shortform
This decision is not only evidence that Austin has poor judgment, but also that he cannot be trusted with leadership roles in-general.
(Why? Well, as a result of this decision I expect I will no longer encourage people to donate to his projects relating to AI and the surrounding ecosystem^[1], and I expect this is true of many other relevant people. Austin could have found this out ahead of time, but evidently didn’t think to check. This means that in future situations where he has power, he will likely continue to make decisions that will cause everyone to lose faith in him and not notice this fact until afterward, and this is a terrible property for someone with a leadership role in an ecosystem.)
These are both important things to discuss; nonetheless framing it primarily around “bad optics” and “PR” is a terribly exploitable tactic and known failure mode.
1. ^
  As an aside, Austin has been an excellent trade-partner, and I expect to still exchange money for goods and services with him.

Ben Pace 22 Sep 2026 15:38 UTC
5 points
−1
in reply to: Dean Valentine’s comment on: Elizabeth’s Shortform
While I disagree with the particulars, I share a sense of moral disgust.

Ben Pace 22 Sep 2026 15:36 UTC
8 points
4
in reply to: Mo Putera’s comment on: Mo Putera’s Shortform
Toby also… crossposted to LessWrong.

Ben Pace 22 Sep 2026 0:57 UTC
17 points
14
in reply to: Elizabeth’s comment on: Elizabeth’s Shortform
This seems much better. Saying something like “I’m not in a position to trust her after the major fraud, and your post is basically not giving me enough information to move past that prior, so basically it seems like you’re doing something insane and terrible and I will act accordingly” is much better than “what you’re doing looks bad to many people and that means it is bad, end of discussion”.

Ben Pace

Jensen Huang Says If We Can­not Align AI, Shut Down the AI Labs

Jensen Huang Says If We Cannot Align AI, Shut Down the AI Labs