Jamie_Harris

• Jamie_Harris 25 Jun 2026 19:36 UTC

  2 points

  0 ∶ 0

  on: Six­teen schemes for AI safety

  Oh! Without having seen this post I started vibe coding a playable version of the AI2027 scenario a few days after you posted this. It’s quite fun, and is ~ready for beta testers. If anyone seeing this would like to play it, comment below or DM me your email and you will be the first person or batch of people other than me to play it.

  To be clear, I made it for fun rather than with an intentional strategy of making it mass marketable but if anyone wanted to take it on and take it that direction I would happily hand over the code etc.

  (PS, cool post generally.)

• Jamie_Harris 12 Jun 2026 9:36 UTC

  4 points

  0 ∶ 0

  on: Col­lege stu­dents: You should run a vir­tual sum­mer EA In­tro Fel­low­ship ASAP. Here’s how!

  Interesting! What are some examples of “where incoming freshmen are tuned in”? (Maybe it’s usually obvious to people at uni these days?)

X-Risk Daily: a new AI-cu­rated daily briefing

• Jamie_Harris 3 May 2026 12:34 UTC

  2 points

  0 ∶ 0

  on: AI safety is ma­tur­ing. Is its gen­er­al­ist tal­ent pipeline keep­ing up?

  Thank you for digging into this and sharing your findings! Some of these seem like important insights if they’re correct.

  On observation 3:

  Motivation exists, experience is scarce and takes years. There might be no shortage of junior, motivated, community-adjacent generalists seeking entry. The shortage is at the other end: senior professionals with years of accumulated judgment in operations, management, institutional engagement, communications, or policy who bring skills that cannot be acquired quickly. Several participants suggested that, for most generalist roles, a competent senior professional can get to a working level of AIS context in weeks, whereas it takes years, sometimes decades, to develop the professional judgment, external relationships, and organisational experience that senior roles require.

  Seniority is a proxy for what is actually needed: diversity and outside-world fluency...

  Several participants named this as an increasingly urgent gap.

  I’m somewhat surprised by this. It’d be good news for me if trye, because I run a career bootcamp that happens to have relatively large numbers of experienced, accomplished professionals seeking to transition into high-impact cause areas, often interested in AI safety, often generalists.

  Would you be able to put me in touch with some of the people who expressed this need so we can understand the gap better and see if I can connect them to relevant talented people?

  You can email me at Jamie.harris@centreforeffectivealtruism.org if easier!

• Jamie_Harris 17 Apr 2026 13:17 UTC

  32 points

  3 ∶ 0

  on: Jamie_Har­ris’s Shortform

  I am currently the only Fund Manager at the EA Infrastructure Fund… and that needs to change!

  I work full-time on something else within the Centre for Effective Altruism, and the EAIF needs a dedicated owner who will drive it forwards.

  I think we’re sitting on a big opportunity here. There’s so much that the EA movement could achieve, and so much great work that could be enabled by EAIF.

  Some indicators of promise here:

  • CEA is growing, but there’s only so much that CEA can work on in-house. We need to fund and nurture great work that’s happening elsewhere, too!

  • There are potential new sources of funding that EAIF could tap into; building a strong product here that donors are excited about is essential.

  • We have a mini roadmap laid out by recent successes within EA Funds.

  Let me say more on that last one. I’ve been extremely impressed by what another EA Fund, the Animal Welfare Fund, has achieved over the past year or two, improving it’s evaluation quality, it’s staffing, and it’s available pool of resources. I think the EAIF has the potential for a similar rocketship trajectory; it needs the right person to come in and make that happen.

  CEA is hiring for a new Head of the EA Infrastructure Fund: full job description and application form here, apply by 4th May.

  Let me know if you have questions! I can’t promise deep engagement with all potential candidates, but I’ll help out with key/​quick uncertainties if I can! Some additional thoughts from Loic, new Head of EA Funds, here.

• Jamie_Harris 16 Apr 2026 13:03 UTC

  2 points

  0 ∶ 0

  in reply to: Rasool’s comment on: How did Leopold do? Eval­u­at­ing Si­tu­a­tional Aware­ness’s predictions

  Oh, nice, I hadn’t seen that, thank you!

  Looks more careful/​thorough.

• Jamie_Harris 30 Mar 2026 7:00 UTC

  7 points

  0 ∶ 0

  in reply to: huw’s comment on: How did Leopold do? Eval­u­at­ing Si­tu­a­tional Aware­ness’s predictions

  Separately, here’s Claude’s direct reply to your specific points in case you’re curious (sorry I don’t enough of a developed inside view take to respond myself!):

  On “China don’t have any frontier labs, only labs which distill other models”: this is probably too strong. DeepSeek introduced genuine architectural innovations (Multi-head Latent Attention, fine-grained MoE) that Epoch AI characterises as real advances, not just distillation. That said, the distillation question is genuinely debated: OpenAI has alleged it, and Chinese labs scraped millions of Claude conversations. The picture is mixed rather than one-sided.

  On “no evidence of an arms race”: both governments explicitly frame AI as a strategic contest (both opted out of the Feb 2026 responsible AI military declaration), there’s confirmed espionage (Linwei Ding convicted Jan 2026 for stealing Google TPU secrets), and $2.5B in chip smuggling. Whether this constitutes an “arms race” depends on your definition, but the competitive dynamic Leopold predicted is clearly present.

  Your most interesting point is the last one: that distillation and open source might mean an arms race never materialises because intelligence becomes cheap and accessible. This connects directly to what I think is Leopold’s most consequential error. He predicted open source would fade and proprietary algorithms would create a durable American moat. Instead, capable AI is diffusing faster than his framework assumed. You’re right that this weakens the case that compute concentration equals geopolitical power, and it’s a genuinely underexplored implication of how things have played out.

• Jamie_Harris 30 Mar 2026 6:57 UTC

  2 points

  0 ∶ 0

  in reply to: huw’s comment on: How did Leopold do? Eval­u­at­ing Si­tu­a­tional Aware­ness’s predictions

  Thanks for reviewing and raising this! You’re right that the US/​China dynamics are central to Situational Awareness’s thesis and we underemphasised them. We’ve now added a dedicated China/​US section with its own tab and three expandable cards, evaluating his specific sub-predictions on infrastructure (7nm chips, power, Middle East), algorithms and open source, and strategic dynamics. Would value your review of the updated version if you have time!

• Jamie_Harris 29 Mar 2026 23:02 UTC

  6 points

  0 ∶ 0

  in reply to: Linch’s comment on: How did Leopold do? Eval­u­at­ing Si­tu­a­tional Aware­ness’s predictions

  True, the 3.5 rating seems a bit harsh! I just tweaked the wording that you quoted directly

How did Leopold do? Eval­u­at­ing Si­tu­a­tional Aware­ness’s predictions

• Jamie_Harris 23 Mar 2026 17:58 UTC

  3 points

  0 ∶ 0

  in reply to: John Salter’s comment on: John Salter’s Quick takes

  Blimey. Did you check with CE about offering it as part of their incubation program (funded by them, maybe paid by results as you say)? And/​or other incubators like Catalyze, or fellowship programs (not founders per se) like Constellation? (IIRC they have an affiliated executive coach already)

  I’m surprised by “I don’t really want a grant” though. E.g. the usual process is basically seed funding grant to check/​demonstrate progress --> if you achieve that (or seem on track to), you get renewed funding. The mechanism isn’t perfect (maybe you can BS your way to success or you aren’t funded without good reasons), but it’s at least ideally fairly results based.

  (I’d be inclined to agree that ideally the founders/​participants themselves would pay, but if you have evidence that they are “irrationally self-sacrificial” and will continue to underpay for the service relative to what they’d endorse themselves with hindsight etc, then that seems like a decent case for grant funding.)

• Jamie_Harris 22 Mar 2026 8:51 UTC

  4 points

  0 ∶ 0

  in reply to: John Salter’s comment on: John Salter’s Quick takes

  I opened your profile and website and couldn’t tell what this referred to? I’m intrigued, even if it’s no longer accepting sign ups!

• Jamie_Harris 14 Mar 2026 14:58 UTC

  2 points

  0 ∶ 0

  on: The case for AI safety ca­pac­ity-build­ing work

  This post prompted me to write up an idea I’ve had in the back of my mind for a while. Asya argues that people in or considering technical or policy roles at AI safety organizations could maybe have more impact doing capacity-building work.

  One way to test if this could be a good fit for you: if you have domain expertise in an AI safety or governance topic, creating a structured course around it might be more feasible than you’d expect. AI tools, volunteer facilitators, and people like me with more experience in courses/​products can handle a lot of the heavy lifting, so the main contribution is your knowledge and judgment about what matters.

  I’ve written up a short proposal exploring how this could work in practice; I’d be keen to hear from anyone interested in trying it out.

• Jamie_Harris 14 Mar 2026 11:39 UTC

  2 points

  1 ∶ 0

  on: The case for AI safety ca­pac­ity-build­ing work

  Separately: the discussion/​comments on the LessWrong cross-post are pretty interesting regarding the case for and against working on capacity building, so people reading here might like to check through those discussions too.

• Jamie_Harris 14 Mar 2026 11:38 UTC

  2 points

  0 ∶ 0

  on: The case for AI safety ca­pac­ity-build­ing work

  This post felt motivating plus personally reassuring to me given that I work in capacity building (albeit not solely focused on AI safety).

  A couple of updates (or at least: things that feel more salient to me) from the case study /​stories were around the value of personal connections and direct personal encouragement to consideration working on [specific thing]. In the stories, it seems that often came from workshops and in-person events, though I’m also wondering if I should be leaning even harder into ways to enable that in the online programs I run.

You need user buy-in to scale your impact

• Jamie_Harris 3 Mar 2026 10:59 UTC

  2 points

  0 ∶ 0

  in reply to: ScienceMon🔸’s comment on: Slaugh­ter­bots are su­per un­pop­u­lar, and the US Govern­ment knows it

  Cool, makes sense. To be clear, I think contacting representatives is helpful! I wasn’t trying to question that.

  I don’t know anything about the Congress authorisation so will defer on that. I’ll just say that if the legality is in dispute rather than unambiguous/​settled, then using the word illegal might be counterproductive/​polarising, whereas “unprecedented” seems unambiguously true.

• Jamie_Harris 2 Mar 2026 0:06 UTC

  3 points

  0 ∶ 0

  on: Slaugh­ter­bots are su­per un­pop­u­lar, and the US Govern­ment knows it

  Nice one for taking action!

  What was the illegal part? Isn’t it just unprecedented?

  (Checking partly for my own knowledge and also because it seemed quite central to your call to action to the legislators)

• Jamie_Harris 23 Feb 2026 22:02 UTC

  5 points

  0 ∶ 0

  on: What Hap­pens After the Ca­reer Ad­vice: The Case for a Shared EA Oper­a­tions Hiring Platform

  Cool post! I thought it was well-structured and evidenced, while also recognising limitations and counterarguments etc.

• Jamie_Harris 23 Feb 2026 21:59 UTC

  3 points

  1 ∶ 0

  in reply to: abrahamrowe’s comment on: What Hap­pens After the Ca­reer Ad­vice: The Case for a Shared EA Oper­a­tions Hiring Platform

  The organization running it would need to have sufficient credibility for the organizations using it to want to forego their own application processes. I think a random person starting it would have very low credibility. My company, which had run several dozen hiring rounds for many organizations had maybe 50% the credibility necessary. This seems like a hard bar.

  I feel like a service that aspires to eventually be a common app could shift towards that incrementally by offering partly-vetted candidates. It’s not a fully centralised common app, but gets customers/​sign ups from orgs who just want access to another sour e kf high quality candidates

  That might reduce some of the value prop to initial candidates at first, if the service doesn’t have many confirmed clients yet, but I suspect that (1) quite a lot would apply anyway, even without confirmed buy in from orgs, if the pitch was done well, (2) there might be other ways to make it appealing, e.g. finding ways to offer some (automated?) feedback.