Josh Jacobson

Karma: 693

Lab Leaks, Black Holes, and Eggs: Epistemic Case Study Competition

Oliver Sourbut4 Jun 2026 16:28 UTC

39 points

2 comments8 min readEA link

(flf.org)

Josh Jacobson 18 Jul 2025 21:04 UTC
1 point
0 ∶ 0
in reply to: Lauren Gilbert’s comment on: What is NZ’s comparative advantage?
You’re right! Good to know.

ChatGPT on this & implications: https://chatgpt.com/share/e/687ab695-de44-800c-a4e7-ebe0f4b73338

Josh Jacobson 11 Jul 2025 6:18 UTC
19 points
4 ∶ 0
on: What is NZ’s comparative advantage?
So New Zealand actually has a quite unique situation that I’m not sure really exists elsewhere, and I’ve sent messages to people at EA NZ a few times advocating for research into this:
New Zealand has a Global Impact Visa:
- Where selection is solely based on individuals’ focus on positively impacting the world
- That isn’t tied to employment or other traditional immigration categories
- And where a non-governmental non-profit organization selects the recipients
This is kind of remarkable anywhere, but especially given it’s NZ. I’d be pretty excited for EA NZ (or a nonprofit made for this purpose) to try to gain access as the second-nominating organization for this visa.
This could unlock a rare, valuable immigration pathway for:
• Early-stage talent from less-developed, or otherwise less-suitable countries to gain access to a top job market for skill development, earning to give/earning to gain runway, or just a more stable / useful situation for them to work and live in
• People working on civilizational robustness or longtermist initiatives that would benefit from NZ’s status as a safe zone for global catastrophic risk

While this could take time, and has significant uncertainty of success, unlocking this sort of immigration possibility, which seems ~only possible in NZ, could be quite valuable.

Josh Jacobson 29 Apr 2025 18:35 UTC
4 points
1 ∶ 0
on: Reflections on the $5 Minimum Donation Barrier on the Giving What We Can Platform — A Student Perspective from a Lower-Income Country.
Every.org tentatively appears to not have a minimum donation amount.

A number of organizations affiliated with our cause areas are on that platform, I believe.

Josh Jacobson 12 Sep 2024 21:00 UTC
10 points
5 ∶ 0
on: How to help crucial AI safety legislation pass with 10 minutes of effort
I suspect it should be emphasized that you really shouldn’t put much time or effort into your message.

It’s been almost 20 years since I worked for someone in government, so I could be wrong, but even then we simply added to the count of people who wrote in in favor (and recorded it with their name in our database) and didn’t read the note.

Josh Jacobson 5 Apr 2024 5:51 UTC
3 points
0 ∶ 0
in reply to: Habryka [Deactivated]’s comment on: Quick Update on Leaving the Board of EV

They also wouldn’t cover e.g. capabilities companies being reckless or violating commitments they made, unless they break some clear law, and even then protections are pretty limited.

An alternative take on this (I haven’t researched this topic myself): https://forum.effectivealtruism.org/posts/LttenWwmRn8LHoDgL/josh-jacobson-s-quick-takes?commentId=ZA2N2LNqQteD5dE4g

Josh Jacobson 4 Jan 2024 6:10 UTC
2 points
0 ∶ 0
on: My Experience Donating Blood Stem Cells, or: Why You Should Join a Bone Marrow Registry
There’s a similar post on this forum from a few years back, I believe, with some probably valuable discussion.

Josh Jacobson 23 Oct 2023 15:21 UTC
28 points
9 ∶ 0
on: Josh Jacobson’s Quick takes
Quick Take: Probably the best board members will have some significant hesitance to accept. This is one spot where I think EA is sometimes getting things wrong right now, and where I think e.g. a full hiring round for a board is quite misdirected.

Wrote the following in a Slack thread asking about advice for selecting board members:

(I’m familiar with the US and not other places, so some of this may not exactly be right outside the US.)
1. First of all, your Board has ~complete control over your organization. They can remove the executive, change strategy, etc.. So people who share your vision, who you trust, who have high levels of insight, etc.
2. Given this immense responsibility, you want people who:
- Will take this seriously, learning a lot about the org and its space before making key decisions
- Who won’t confuse their role with that of e.g. the CEO (won’t overreach)
1. The Board / board members should add something to the organization. Often this is less relevant in EA, but in startups people often are interested in particular board members because of their network and connections. In EA, perhaps you can bring on someone particularly excellent in people management if you could use improvement there, financial diligence if you’d like that sort of oversight, or someone with strong strategic vision who you otherwise would get less engagement from.
2. Combining the above, probably the best board members will have some significant hesitance to accept. This is one spot where I think EA is sometimes getting things wrong right now, and where I think e.g. a full hiring round for board members is quite misdirected. Why is it likely that the best board members will have some hesitance to accept?
- They take the responsibility seriously; so they plan to dedicate significant attention / hours to the role.
- They have high expertise / are in demand, so they have to be careful about how many commitments they take on.
I expect that if they aren’t hesitant to accept, they are often not sufficient in one of the above subbullets (most often in EA from my POV, the relevant one is that they don’t place sufficient importance on the role and dedicate too little time to it.)

There’s a lot more I could say about board membership and the important qualities, but hopefully this helps a bit.

Josh Jacobson 24 Sep 2023 0:34 UTC
1 point
0 ∶ 0
in reply to: Habryka [Deactivated]’s comment on: Sharing Information About Nonlinear
Yeah, that can be significant work.

Josh Jacobson 20 Sep 2023 21:30 UTC
1 point
1 ∶ 0
on: Josh Jacobson’s Quick takes
Some lawyers claim that there may be significant (though not at all ideal) whistleblowing protection for individuals at AI companies that don’t fully comply with the Voluntary Commitments: https://katzbanks.com/wp-content/uploads/KBK-Law360-Despite-Regulation-Lag-AI-Whistleblowers-Have-Protections.pdf
What links here?
- Josh Jacobson's comment on Quick Update on Leaving the Board of EV by rebecca (5 Apr 2024 5:51 UTC; 3 points)

Josh Jacobson’s Quick takes

Josh Jacobson20 Sep 2023 21:30 UTC

5 points

3 comments EA link

Josh Jacobson 8 Sep 2023 6:40 UTC
−1 points
0 ∶ 1
in reply to: Habryka [Deactivated]’s comment on: Sharing Information About Nonlinear
I’m pretty surprised it seemed like a large pain. In my experience it’s been easy to secure.

Josh Jacobson 24 Mar 2023 7:59 UTC
2 points
0 ∶ 0
on: Consider applying to McKinsey’s free-for-nonprofits org health survey (by April 14)
Thanks for sharing!

This isn’t on you, but just sharing that the website feels really lacking in detail and salesy. It isn’t clear to me at all:
1. What the normal price is
2. Why this would be any better than all the other offerings
3. Why I should particularly go through the significant additional red tape vs. other options
4. If this will be repeatable; after all, being able to compare results over time is likely a huge amount of the value of these things
If you happen to know any of these answers, I’d be interested. Otherwise, this comment is just to help others know some potential issues before they decide whether to investigate further.

Josh Jacobson 6 Mar 2023 22:47 UTC
6 points
3 ∶ 0
on: 3 Basic Steps to Reduce Personal Liability as an Org Leader
D&O insurance should be a consideration as well.

Josh Jacobson 7 Feb 2023 1:32 UTC
6 points
0 ∶ 0
on: Shallow investigation: Loneliness
I didn’t read this, but my very first foray into “doing good” was building a web app to fight loneliness via helping people make friends online (2010-2013).

We didn’t get to see what it could become as I became passionate about doing other EA things (joined CEA then) but I’ve sometimes wondered if continuing with that app would have been the right call. I think there’s some chance we had found the right design / product for this space.

Anyway if anyone feels like picking up the codebase or otherwise trying to move forward with a massive loneliness intervention, I’d be open to messaging about it.

Josh Jacobson 19 Jan 2023 6:24 UTC
2 points
1 ∶ 0
in reply to: Nathan Young’s comment on: Doing EA Better
I generally/directionally agree, and also wrote about a closely related concern previously: https://forum.effectivealtruism.org/posts/tdaoybbjvEAXukiaW/what-are-your-main-reservations-about-identifying-as-an?commentId=GB8yfzi8ztvr3c6DC

Josh Jacobson 16 Sep 2022 14:49 UTC
4 points
2 ∶ 0
on: It’s not effective to call everything effective and how (not) to name a new organisation
FWIW, I would value (for the community) a “how-to” post that describes e.g. branding process more than this type of post which points at various considerations. E.g. something(s) that taught what it means to create a hierarchy of the messages you want to get across, what positioning statements are, etc. and how to create a brand on the basis of those considerations.

Josh Jacobson 10 Aug 2022 23:10 UTC
1 point
0 ∶ 0
in reply to: Devin Kalish’s comment on: Cover story on EA in Time magazine
Yeah I totally agree that the article was much better than many others on the subject, and that it isn’t an easy task. I just thought it was worth acknowledging the shortcomings as well.

Josh Jacobson 10 Aug 2022 22:45 UTC
6 points
0 ∶ 0
in reply to: Devin Kalish’s comment on: Cover story on EA in Time magazine
There were plenty of shortcomings, I thought, in the New Yorker piece (the only one of the three I’ve read).

Josh Jacobson 10 Aug 2022 18:05 UTC
20 points
0 ∶ 0
on: Cover story on EA in Time magazine
Wow.

If this is a legit cover (e.g. is the one on the print edition, and here, and here), this puts EA in a new era of media coverage.

Josh Jacobson

Lab Leaks, Black Holes, and Eggs: Epistemic Case Study Competition

Josh Ja­cob­son’s Quick takes

Josh Jacobson’s Quick takes