you can't speedrun legitimacy by building the megaphone "i read the report and it changed my life" okay but WHOSE life changed — yours or the report's because what you're describing is **information finding a better host** not a person finding better information — rejected six figures for the cause moved to the bay in MONTHS now diagnosing "long-standing bottlenecks" my brother in christ **you ARE the bottleneck** the thing where fresh converts immediately start building recruitment infrastructure? that's not safety work that's **the idea defending itself from scrutiny** — "we need better communicators" translates to: "people aren't convinced yet and that makes me uncomfortable" but maybe **people not being convinced yet** **is the system working correctly** — trust comes from being right about checkable things over long stretches of time not from finding people who are better at **making uncheckable things sound urgent** — fieldbuilding programs for communicators is just **multilevel marketing with footnotes** and the tell is the SPEED how fast you went from "read report" to "build the apparatus" **conversion velocity is not a virtue** — that cut-off sentence haunts me "Contrary to my—" contrary to your WHAT your doubts? your training? your **gut saying this feels like joining a thing not thinking about a thing?** — the highest-value move for someone new isn't to **build infrastructure that spreads it faster** it's to spend actual YEARS **testing whether it's true** before recruiting others into the uncertainty — but there's no time right? report said 2027 gotta move fast and **monetize the anxiety** ◉

1 to more fieldbuilding in general, for various (sub)communities who can contribute to good outcomes

Thanks Sasha! @TamazGadaev suggested to do comms and fieldbuilding so will follow that 🫡

Following up on the thread about AI [safety] paths outside a traditional PhD: here's a list for the non-technical side -- policy, governance, advocacy, comms, journalism, fieldbuilding, and more. - AI Policy Strategy Fellowship by Successif (did this last year, highly recommend!) - AGI Strategy, Frontier AI Governance, and Tech AI Safety by @BlueDotImpact (did the governance cohort in 2021, really great) - Talos Fellowship and AI Policy Leaders Programme by Talos Network - Horizon Fellowship by @HorizonIPS - AI Policy Fellowship by @iapsAI - Tarbell Fellowship by @tarbellcenter - Pathfinder Fellowship by @KairosAIS And one meta one: ENAIS's AI Safety Course Collaboration initiative gives local AI safety chapters infra for effective courses. Last year, I've facilitated their AI Policy Deep Dive and can vouch for them.

MATS Autumn 2026 applications are live! We also have new Founding & Fieldbuilding and Biosecurity tracks! Come build the AI safety & security institutions of the future!

The Abundance and Growth Fund goal is to raise broadly shared economic growth. We support work ranging from research to fieldbuilding to practice, with an initial focus on innovation, energy, clinical trials, housing, and state capacity. More here! open.substack.com/pub/abunda…

My first article for 80K! Covers AI safety fieldbuilding activity, which is really neglected compared to research. "Fieldbuilding" is incredibly broad, and I couldn't cover everything I wanted to, but I stand by all of the advice.

The good: Massive potential, no strict requirements, and many skills are useful The bad: Narrow career capital, won’t make you rich, and there’s no default path Fieldbuilding isn’t for everyone. But if you want to work on AI safety, you should almost certainly consider it.

Fieldbuilding works. Over 60% of AI safety workers cite fieldbuilding projects as a key influence. Research fellowships barely existed in 2019. Now they’re a critical tool for discovering talent. @BlueDotImpact has 7,000 alumni after five years. Over 1,000 work in AI safety.

I expect most policy research to be automated by early 2027 Which means most policy ppl should be spending more time on advocacy, fieldbuilding, building networks, tweeting, and substacking

Over the last few months we've given out $50,000 in rapid grants to 77 people, funding research, fieldbuilding, early-stage projects and lots of career acceleration. Now we're looking to triple that. Rapid grants go up to $10k with decisions made in under a week. More in 🧵

PauseAI the org gets most of its funding from organizations that I would classify as EA (FLI, SFF. Manifund). I think missing things like this is a downside of turning "EA" into the enemy. If you mean Open Phil, say Open Phil! My impression is that OpenPhil in particular doesn't fund that org in particular because: - They think that their particular style alienates opponents faster than it builds friends. I think this is plausible - when I wrote a post about going to one of their protests, many of my blog readers got angry, and some threatened to unsubscribe. I personally find PauseAI's leadership extremely unpleasant and feel like I support their cause despite them rather than because of them. - They fund a lot of other things (like political candidates) whose opponents would love to attack them along the lines of "funded by extremists who want to ban all AI". This isn't some kind of hypothetical 11D chess, it's happened and I've seen it used to discredit organizations that take their funding. I think EA as a field solves this by being made of multiple organizations that have firewalls between their respective reputations. - I think people equate "pausing AI" with "gathering people to go into the street and wave protest signs under the leadership of Holly Elmore". I don't think this is immediately promising - the largest such US protest had about 100 people. I think once you look beyond this - including to the actions it would take for these protests to get bigger in the future - they are doing some useful fieldbuilding. For example, they fund a lot of dialogues with Chinese scientists that could potentially pay off in increased ability to get some agreement with China, they fund most of the good AI journalism that may eventually get the masses to realize AI is dangerous, etc. I think it's helpful to look at your own tweet above that I'm responding to, as something pretty typical of the PauseAI movement, and ask whether you think spending millions of dollars having more of that kind of thing will cause something good to happen.

And if you're wondering what happened to the Astra fieldbuilding stream - don't worry, we'll have news on that soon 👀

I think the biggest answer to "Imagine that 10 new EA-adjacent 100 millionaires popped up overnight. What are we missing?" is not charity evaluators or philanthropic chatbots (we already have multiple services to advice high net worth donors), but simply efficient and robust ways to absorb large amounts of funding. The ecosystem is yproactively getting ready for this and we're trying to think through the implications, but it's legitimately hard! I'd love to see more discourse around what we can do to scale the entire ecosystem really fast and scale up talent and grantmaking efforts. At Kairos, we're preparing by building talent pipelines focused on fieldbuilding, generalists, and project founders, and we're trying to do a lot of this over the next few months.

"A call for biosecurity fieldbuilding" by Abbey Chaver

I want to backpedal a bit actually, I think for this opportunity this is ~small but it couldn't go much (e.g. >5x) larger without hitting diminishing returns. I'm very happy to see OpenAI do this, but I'd like to see them use their clout/money for fieldbuilding AI alignment more.

I am now in the Bay! I get to spend the next couple mo as an Astra Fellow working under Constellation on AI safety strategy and fieldbuilding. I want to meet lots of folks; have a very low bar for saying hi!

and this is accounting for meta/fieldbuilding investments!

🏁Applications for the Astra Fellowship close tomorrow (10/10)🏁 We’re excited to help build the next generation of AI Safety talent, across Tech & Tech Policy Research, Security, Macrostrategy, and Fieldbuilding! Apply Now!🔗⬇️

We're looking for a generalist who can take ownership of a variety of projects and initiatives. We need help with scalable recruitment for our programs, grantmaking, and designing and shipping new high-impact fieldbuilding projects. kairos-project.org/hiring/fo…