Really good findings Kristof, and honoured that Joey and I help spark some interest here!

Would be good to see where the LLMs sourced the overstated risks in harm reduction tools, and "manufactured uncertainty" - how much of it was the LLMs' own language versus obtained from a source?

I.e., to what extent could Big Tobacco be running Answer Engine Optimisation campaigns to keep people hooked on their products? If they are, at what point would you think this become a fundable case where the anti-tobacco movement replies with a counter-effort? 

And should it focus on zyns who scored 4.25 on helpfulness (the highest), vapes who although scored 3.25 has a huge TAM, is the bottleneck framing, or is there a better overall strategy?

(acknowledge these are one-run queries and a more sophisticated study is needed, but the overall direction of next steps is what's interesting)

What are your top 3 and bottom 3 picks on this list, and what do you think the right "founder" profile could be for your top 3? (and when will it be you? 😊)

I currently work in a for profit company with middle managers in supply chain / operations that have ultra limited (if any) bandwidth to engage in EA concepts beyond hanging out with me at the water cooler. 

if given time and resources they might be excellent potential fit for the exact kind problem statement outlined in the post.

Do I point them at High Impact Professional / ScaleWise and hope they free up bandwidth, or is there a better solution out there to absorb and onramp time-constrained senior talent?

Hey zenc!

https://forum.effectivealtruism.org/posts/yfpGTWTry9e5pDX4y/have-you-written-a-will-involving-a-bequest-to-charity-do 

Includes: 

• a past initiative from Joey Savoie a decade ago before he founded Ambitious Impact (now defunct)
• A GenerationPledge website for ultra high net worth inheritors (now just some Google Slides, but seems updated as of 2026!)
• UK-specific resources from Giving What We Can
• Survey-style questions from the OP in the comments.

Another guide from 2014 with several defunct links here.

State of the art today (for now): most Effective Giving Organisations have a "Bequests / Last Wills" webpage. To my knowledge there is no "effective giving online will generator" platform (yet).

Current people I'm aware of that are doing this at or above ~0.1 FTE is Anne Schulze (Effektiv Spenden co founder), Fianna Nieuwenstein (Senior Comms professional at Doneer Effectief). 

There are other players who have WIP products but I'm unclear to the extent I can disclose here. I am personally extremely interested in this question and overall subject. We should DM

Ops culture isn’t yet a building culture - the default mindset is “keep the trains running” rather than “ship something”.

real

Should forecasting receive more or less EA funding?

It's not a good use of marginal funding when even one of the best forecasters can't build a strong view around its impact. 

You can say "confidential AI use cases make it valuable". I don't disagree. I disagree that $100M went into this. Why shouldn't it be $10M?

I'm glad as a movement we are able to appreciate this nuance too David. By no means does Anthropic get a free pass, especially since mass foreign surveillance wasnt part of their two non-negociables, as someone who is foreign to the US.

Still want to acknowledge the bold action they took in the face of an autocratic regime

I've never been more proud to be part of the Effective Altruism movement

Knew it was a bit too good to be true 🥲 Meltblown PP rolls and facepiece moulds seem like the key upstream bottlenecks to N95 / reusable respirator production.

More reason to advocate for pre-securing offshore oil rig-style labour contracts + government compliance (NIOSH) for surging PPE production.

Intuitively, advocating for private companies to maximise production and removing the friction to do so feels more tractable than advocating federal governments to stockpile +5 billion additional N95s. Ideally we should still pursue both

500M in N95 mask materials in 28 days from just 2 plants is a really massive number, that's 17M per day!

Quick BOTEC: 50M essential workers in the US demand 10M N95 masks per day, assuming the best N95 masks on the market today have a 5-day lifecycle. Even if they demanded 50M day, if we increased the number of plants from 2 to 10 (assuming the next 8 have 50% the max capacity of the Braskem plants), this might potentially solve the meltblown bottleneck.

Updates me toward thinking meltblown polypropylene may be more capable of surge than I previously assumed.

If similar labour pre-commitments and coordination mechanisms could be secured across other PPE stages before the next pandemic, this could potentially shift the surge vs stockpile ratio more towards surge.