JamesN

Karma: 406

Executive Director of the Swift Centre for Applied Forecasting (led projects with U.K. Gov., Google DeepMind, and on AI security and capability risks).

Co-founder of ‘Looking for Growth’ - a political movement for growth in the U.K.

CTO of Praxis—a AI led assessment platform for schools

Former Head of Policy at ControlAI (co-authored ‘A Narrow Path’)

Former Director of Impactful Government Careers

Former Head of Development Policy at HM Treasury

Former Head of Strategy at the Centre for Data Ethics and Innovation

Former Senior Policy Advisor at HM Treasury, leading on the economic and financial response to the war in Ukraine, and the modelling and allocation of the UK’s ‘Official Development Assistance’ budget.

MSc in Cognitive and Decision Sciences from UCL, my dissertation was an experimental study using Bayesian reasoning to improve predictive reasoning and forecasting in U.K. public policy officials and analysts

JamesN 7 Sep 2026 17:12 UTC
3 points
3 ∶ 0
in reply to: Sudhanshu Kasewa’s comment on: Sudhanshu Kasewa’s Quick takes
I agree with all of this, but I don’t think the gap is talent. It’s in organisations or funders being able to hire, cultivate, or bet on this talent quickly.

Tens of $m’s to a fellowship, talent program, or evaluator seems pretty standard now—where most of that money will go to stipends or compute (read: AI companies). The issue is: what tangible roles aligned to the above are those creating?

Where are the equivalent direct work grants? With $10m I could expand/spin up a team of 30-50 people working directly on advocacy, or communications, or threat modelling and scenario analysis. I’m sure many others could do the same.
I don’t think there much of a talent gap, or a funding gap, but rather a strategic allocation gap.

JamesN 10 Aug 2026 20:22 UTC
3 points
0 ∶ 0
on: Why it’s so hard to get experienced senior managers for EA organisations.
This is an excellent post and the fact it hasn’t got more karma and engagement is evidence of the wider problem within the ecosystem at the moment.

JamesN 14 Jul 2026 9:18 UTC
3 points
0 ∶ 0
in reply to: Sean_o_h’s comment on: JamesN’s Quick takes
Your essay was a good read but it is an incredibly low bar if we count it as “critical of Anthropic”.
My equivalent would be trying to get funding for serious advocacy or policy work that went against Anthropic’s position.

JamesN 13 Jul 2026 15:03 UTC
3 points
1 ∶ 0
in reply to: Mo Putera’s comment on: JamesN’s Quick takes
Those are all 2025, and most are late 2025. Also I’d suggest they are potentially informative to my argument, given one of them is literally about:

“Anthropic is not being consistently candid about their connection to EA”

JamesN 13 Jul 2026 12:01 UTC
0 points
0 ∶ 2
on: JamesN’s Quick takes
In 2023-2024, it seemed a very strong bias from the EA community and the core AI safety funders that Anthropic not to be argued against or funding diverted to anything that was critical of them (directly or indirectly).

I wonder how the community and funders have updated their beliefs and behaviours?

JamesN 17 Jun 2026 7:58 UTC
1 point
0 ∶ 0
in reply to: CalebMaresca🔸’s comment on: Good-tasting alt meat is a scientific problem we can solve: $10M RFP for Alternative Protein R&D
I think the issue is arising from a simple miscommunication.
You seem to be arguing that improving taste has some sort of intrinsic social value in and of itself, that warrants funding. My argument has been, from the start, that if the goal is to maximise uptake of meat alternatives then what is the proof better tasting products is the best method to achieve that and that there is evidence of a market failure there.
Your argument hinges on taste being the value we want here. I reject that as a premise. The value the RFP wants is better uptake of the meat alternatives. And there are big market players that are fundamentally incentivised to solve that (and can raise capital if they prove there will be the returns—I.e. the actual uptake and not just vague “of course if it’s tastier more people will eat it”).
To be as explicit as I can to reply to your argument: your Apple example proves my point above. Your causal chain in that argument is:
More storage → better insect welfare = win
Apple don’t care about insect welfare (I assume). They care about more users and thus more profit. Thus they aren’t incentivised to solve any part of that causal chain.
This RFPs causal chain is not like that, it’s:
Tastier meat alternatives —> more people eat meat alternatives —> less animal suffering = win

Beyond Meat, even if they don’t care about the end goal (which I assume they probably do), do care about solving the middle part, which is the predictor to the less animal suffering end goal. Given that, if tastier meat was the best way to achieve that, they’d be able to raise capital. If they couldn’t, it’d suggest there wouldn’t be enough uptake from such a move—which means bad value for money and thus either we’ve hit the limit of the number of people who will eat meat alternatives, or it is not an efficient way to increase uptake.
Not to get distracted but to avoid being criticised for not answering your point again: Separately, I also disagree with the foundational research point—in that I think the choice of that is a function of this being philanthropic funds, rather than because it’s solving a market problem such as inability to patent the research or keep it secret. Evidence being: all the food manufacturers who spend money making their food taste better and keep their recipes secret.

JamesN 15 Jun 2026 21:28 UTC
3 points
0 ∶ 0
in reply to: CalebMaresca🔸’s comment on: Good-tasting alt meat is a scientific problem we can solve: $10M RFP for Alternative Protein R&D
I fear this is missing the main objective though. Maximising taste is solely for the purpose of maximising uptake of alternative meats. There isn’t an intrinsic good in having tasty meat alternatives.

The logic is: tastier meat alternatives —> more people eat meat alternatives —> less animal suffering.

It’s foundational argument is better tasting meat alternatives increase the number of people eating meat alternatives.
There are companies that exist to sell people meat alternatives. So their incentives are aligned. Profit maximising is just an incentive to do that most efficiently, so they’d do it if taste was the most cost effective way—I.e. if for every $1 spent on making meat alternatives taste better it led to $2 of sales, but $1 spent on marketing led to $1.5 of sales, then it’d be better value to spend on making it taste better.
My issue with this RFP is it presumes the market is failing. My question is where is the proof? It being underfunded by public agencies and private R&D suggests to me either massive vested interest (which may be true for public agencies but less clear why Beyond Meat wouldn’t want more customers if tastier meat alternatives was the best way to do that); or it suggests there is a more efficient method being employed by those with the incentives to find it. Unless you think the meat alternative providers don’t want more customers, or are incompetent (in which case why hasn’t a tastier competitor already appeared).

This whole RFP looks to fix a market failure, which I’m not clear exists. It also tries to do that with $10m, which by their on linked numbers is such a tiny fraction of money spent on tasting research. Unfortunately, this RFP has the hallmarks of a fund that would have to be so amazingly well spent to make a difference that the whole thing seems destined to have little to no impact.

JamesN 13 Jun 2026 21:04 UTC
2 points
0 ∶ 0
in reply to: CalebMaresca🔸’s comment on: Good-tasting alt meat is a scientific problem we can solve: $10M RFP for Alternative Protein R&D
I think you are misunderstanding the market incentives here?

There are companies that explicitly exist to produce meat alternatives. Their incentive (whether profit, reducing suffering, empire building) is to produce the best product to have more people buy it (and realistically eat it given that’s the primary purpose of their product). Taste is assumed (given this RFP) to be a key predictor of that, so those alternative meat companies are incentivised to maximise it.
This RFP explicitly says more people eating alternative proteins would be great for reducing animal suffering. So the market incentives and the desired impact are in alignment. Every $ spent by Beyond Meat to improve the taste, and therefore uptake (as this RFP suggests this relationship is sufficiently linked), is great for animal welfare.
Given the above I struggle to see where the misaligned incentives in the market is and why philanthropic funding of $10m is expected to move the dial here.

JamesN 11 Jun 2026 23:22 UTC
4 points
0 ∶ 0
on: Good-tasting alt meat is a scientific problem we can solve: $10M RFP for Alternative Protein R&D
I’d be interested in understanding why you think there is a market failure here? Why isn’t this a problem that Impossible Foods, Beyond Meat or anyone else is going to be actively solving, and why will $10m shift the funding gap?
How much is taste actually a predictor of uptake?

JamesN 28 May 2026 20:43 UTC
3 points
0 ∶ 0
in reply to: Jared T Peterson’s comment on: Why Forecasting Fails Decision Makers
I think this misunderstands forecasting or mischaracterises decision making.
Your brain absolutely does decide when to make a left-hand-turn at a busy intersection based on a probabilistic estimate. Your brain is just pretty reliable at making said judgments that you assume some binary choice was made. Subconciously you are weighing up the likelihood of various risks based on your senses observing them and concluding that there is a very low (or an appropriately low) probability of it going wrong. The easy way to test this is go to a zip line, one with a drunk operator and one with a sober operator. The zip line may look identically safe but your brain will (hopefully) consider the former more risky—not because the impact of something going wrong is different between the zip lines (both falls would hurt, if not kill you), but because the likelihood of something going wrong is different (as you’d calculate that a drunk operator may not be as cognitively aware of what they are doing). That’s a probabilistic assessment.
The difference is, most decisions we deal with in the world don’t need us to sit down and do a formal prediction process to ensure good decision making. However, most organisational, and especially those on policy and governance etc., do require it. They are concretely two predictions you are making:

1. What will the world be? (through the lens of what you care about, e.g. will there be an oil crisis?)
2. What influence on the world will my policy have? (e.g. if I send Trump a friendly email will he avoid attacking another petrostate?)
Viewing forecasting as just the probabilistic estimate is missing the entire benefit. Forecasting is the entire process. Stating your view probabilistic just allows you to understand your own and others uncertainty (and to provide a purity of accountability and incentive alignment). It only adds decision fatigue if you have not put in the right processes to interpret the result (such as upfront thresholds for action). If you are following an optimal predictive decision making process, you should be making your assumptions and your weighing up of information explicit. This is how you determine what is relevant, what isn’t relevant etc.
Dominic Cummings statement misses that everything he said there IS forecasting. What he’s actually saying is “the point estimate from forecasters was not useful—it was the explicit reasoning about the causal chain etc. from forecasters that was useful”.
However, benchmarking against accuracy ensures the incentives are correct and that decisions aren’t manipulated by elements that decrease the end outcomes efficacy.
If you conclude that it’d be better if decision makers had: 1) the most accurate view of the world; and 2) the most accurate view of how their actions may influence that world towards their objectives. Then I would stand by the statement that it’s sad.

JamesN 28 May 2026 20:21 UTC
2 points
0 ∶ 0
in reply to: Mahi’s comment on: Why Forecasting Fails Decision Makers
So my experience is that identifying/specifying/generating the right questions is at least 50% of the benefit, if not higher. There are lots of reasons organisations struggle with this, from: organisational incentives; incoherent steers and views from senior managers; lack of accountability and ownership; to simply not recognising that they are trying to do a prediction.
This is why forecasting funding that has focused on improving forecasting accuracy is flawed, because it doesn’t matter how accurate you are if your question isn’t of use to the decision making process.
The problem exists at both those “levels”, but the most important one to solve for an organisation is the first one. Issues with resolution criteria etc. decrease accuracy but as long as people’s rationale’s are explicit you can bridge that gap (i.e. I know why one person was higher and another was lower, it was because they both took the resolution to mean something slightly differently). But if the question/problem you are trying to solve in the first instance is wrong, then the whole thing is a waste of time and energy.

JamesN 19 May 2026 10:01 UTC
12 points
0 ∶ 0
on: JamesN’s Quick takes
We do a monthly public forecast at the Swift Centre (swiftcentre.org) - just a chance for us to make predictions and insights on some topical events.

As it’s EAG London month I thought I’d ask what topics/questions are on people’s mind so we can potentially throw them into the mix.
Comment below!

JamesN 12 May 2026 9:26 UTC
1 point
0 ∶ 0
on: Exporters Without Borders: Why You Should Start a Company Instead of Working in Aid
I agree with the premise but we shouldn’t be using philanthropic funds to try to patch over what is a market problem.

The route here should be projects that enable less friction for trade and investment, rather than creating a company that tries to bypass the fundamental issues. Philanthropic funding here should focus on systemic change to have compounded impact.

JamesN 3 May 2026 12:03 UTC
1 point
0 ∶ 0
in reply to: Danila Medvedev’s comment on: Forecasting is Way Overrated, and We Should Stop Funding It
This obviously assumes Marcus has a sufficient level of experience to justify the claims. Which I think, given other comments, can be adequately challenged.

It would be good to know what metric/threshold/examples would be taken as forecasting delivering adequate impact to justify funding. From examples in this thread alone, we can see senior government decision makers in both the U.K. (including Ministerial teams and critical committees) and US, frontier labs safety teams, and philanthropic funds moving tens of millions of dollars a year) have utilised forecasting (either the process or the outputs) to inform their decisions.

The argument of it only shifting a decision 1-2% is totally fair. But to keep consistent I’d expect the same people who make that argument to also be highly sceptical of the vast majority of research funding.

JamesN 30 Apr 2026 8:14 UTC
1 point
1 ∶ 0
on: Technical AI Safety Research Plausibly Increases P(doom) Absent Vastly More Advocacy
(Caveat—I read the premises and skimmed the rest)
Yes—AI research is useful and does help highlight specific advancements or potential risks. However, I fear it is being focused on by many because of personal interest in the topic, rather than the best route to reduce catastrophic and existential risks.

For better or worse, advocacy, policy, and communications are the most likely routes to reduce p(doom) - unless you believe alignment is a plausible and concrete thing.

JamesN 29 Apr 2026 15:59 UTC
4 points
0 ∶ 0
in reply to: Vasco Grilo🔸’s comment on: Why Forecasting Fails Decision Makers
Yeah 40pp, though 40% difference may also be informative depending on the question and distribution.

Regardless, “40” was just a random number. Basically the interesting thing are the areas of greatest difference, not the probability itself.

Bridging the Gap on AI Safety Policy

JamesN29 Apr 2026 14:48 UTC

8 points

0 comments4 min readEA link

JamesN 27 Apr 2026 7:30 UTC
9 points
2 ∶ 0
in reply to: NickLaing’s comment on: Forecasting is Way Overrated, and We Should Stop Funding It
We could “forecast” the likelihood of that haha.
I can’t get into specifics. But if you believe activities like evaluations of models to test for dangerous behaviour etc. is net negative, then that may give credence to your assumption. As an extra data point of whether we’d do work we thought was net negative, I was Head of Policy at ControlAI and co-authored narrowpath.co, and our forecasters have done numerous AI safety focused projects (with and outside of the Swift Centre, including AI 2027).

JamesN 26 Apr 2026 8:28 UTC
4 points
1 ∶ 0
in reply to: Chris Leong’s comment on: Forecasting is Way Overrated, and We Should Stop Funding It
Sort of, but that also doesn’t capture the significant accuracy and efficiency benefits the process of structured reasoning and communication that forecasting enables. There’s substantial risks and issues of “just looking into an issue yourself”—especially when you are more confident in your judgement (because that’s a clear risk of confirmation bias/overconfidence).
The main use of forecasting is in utilising the core scientific benefits it can bring as above into, to help real world decision makers. But fundamentally, that hasn’t been funded—instead we’ve funded tournaments and research.

JamesN 26 Apr 2026 8:10 UTC
58 points
9 ∶ 0
on: Forecasting is Way Overrated, and We Should Stop Funding It
I don’t disagree with some of the fundamentals of this post. Before diving into that, I want to correct a factual error:
“the Swift Centre have received millions of dollars for doing research and studies on forecasting and teaching others about forecasting”
The Swift Centre for Applied Forecasting has not received millions in funding. The majority of our earnings have been through direct projects with organisations who want to use forecasting to inform their decisions.

On your wider argument. I think forecasting has probably received too much funding and the vast majority of that has misallocated on platforms and research. I believe some funding (hundreds of thousands) to maintain core platforms like Metaculus as a public good of information. Though, services like Polymarket can probably fill most of this need in the future (but many useful, informative markets would never reach the necessary volume to be reliable).
Where I think we disagree most is in the application of forecasting and some of the achievements. We’ve worked with frontier AI labs to inform their decisions, are currently advising a U.K. Minister’s team on a central piece of their policy, and are about to start a secondment where I will be advising one of the most influential decision making committees in the country to help improve their scenario analysis and forecasting. Forecasting, and specifically, the science of decision making that it is built on, has the ability to structurally improve decisions in institutions. Significantly better than asking two or three of your smartest friends. That was just never funded, so instead we conclude forecasting is not useful.
What links here?
- Winners of the Manifund Essay Prize by Austin (1 May 2026 2:21 UTC; 29 points)
- Winners of the Manifund Essay Prize by Austin Chen (LessWrong; 1 May 2026 2:21 UTC; 6 points)

JamesN

Bridg­ing the Gap on AI Safety Policy

Bridging the Gap on AI Safety Policy