Will Aldred Email & Phone Number

United Kingdom

Researching ideological fanaticism, especially as it relates to ‘superintelligent’ AI, astronomical suffering risks, and the ‘long reflection’. Receiving mentorship from Polaris’s Research Director, David Althaus.

I am the shield that guards the realm of productive online discourse.

In my first six months, I was a research analyst within a five-person AI forecasting team. During this time, I was responsible for:- Carrying out internal research (on topics like correlation vs. causation in conditional forecasting, to inform Metaculus’ product direction) as well as private research for a grantmaking client (focused on the so-called ‘deployment problem’ in AI safety, especially with respect to forecasting likely US-China dynamics).- Co-organizing, and writing up findings from, a scenario planning workshop on AI existential risk in which several high-profile members of the AI governance community took part.Over the next year or so, I was Metaculus’ AI forecasting lead. My responsibilities included:- Laying out a strategy for Metaculus’ AI forecasting work; getting buy-in from the team and the CEO.- Contributing to Metaculus’ strategic direction more broadly. (I became known as the most prolific commenter in, and writer of, internal documents.)- Running the AI Progress section of the Metaculus site: https://www.metaculus.com/ai/. This included managing question writing contractors as well as ‘resolving’ questions (in the forecasting-speak meaning of the word) and writing close to two hundred forecasting questions myself. I was also active in many forecasting questions’ comment threads, to facilitate better discussions between users/forecasters.- Developing a product roadmap to tie in with my strategy and to help solve particular deficiencies I observed with the Metaculus platform (e.g., skewed forecasting incentives; less-than-ideal inter-forecaster discussions).- Liaising with stakeholders, such as policymakers, grantmakers, and AI governance researchers. Providing private forecasting services for stakeholders—mainly by writing private question series plus forecasting myself on the questions and/or setting up the stakeholder with Metaculus Pro Forecasters.I ended up leaving Metaculus on account of disagreements with the new leadership.

Received a grant from the Effective Altruism Long-Term Future Fund to complete a research project on nuclear weapons risk, and to undertake upskilling and independent research in longtermist AI strategy. During this time, I participated in Cambridge Boston Alignment Initiative’s month-long AI strategy research fellowship (Jan ‘23), held in Berkeley, California.

Cambridge, United Kingdom

Some highlights, in roughly chronological order:- Led three iterations of EA Cambridge’s termly, eight-week-long introductory effective altruism fellowship. Each time, this involved managing around a dozen facilitators; each time, the fellowship had around 60 participants.- Also led some experimental community-building initiatives, such as day-long intensive intro workshops, rationality seminars, and ‘red team a popular paper’ symposiums.- Provided 100+ Cambridge students with 1-1 guidance on how to have a high-impact career.- Participated in ‘Icecone’, a vigorous two-week visiting fellowship in rationality and x-risk run by Lightcone/LessWrong. Held in winter 2021-22 in Berkeley, California.- Facilitated multiple cohorts’ discussion sessions in the AGI Safety Fundamentals seminar programme, governance track. (At the time, AGI Safety Fundamentals was a project of EA Cambridge, directed by another Campus Specialist. It is now part of a spin-out organisation named BlueDot Impact.)- Mentored in Tianxia Academy (a.k.a. EA China)’s research mentorship programme; led a collaboration between Tianxia, EA Cambridge’s ‘China and global cooperation’ cause area group, and Metaculus, a forecasting platform: https://www.metaculus.com/tournament/future-of-china/- Co-directed the Cambridge Existential Risks Initiative (CERI), a summer research fellowship for aspiring x-risk researchers. This included leading the hiring/applications process—we received almost a thousand applications—and managing all twenty-four summer fellows. In addition, I was research principal for two of CERI’s five cause area tracks: nuclear risk and misc. & meta x-risk.- Guided CERI’s vision setting, splitting off from EA Cambridge, and acquisition of fiscal sponsorship and operational support from Rethink Priorities following the ‘22 SRF. (CERI subsequently rebranded as Existential Risk Alliance (ERA). I parted ways with CERI/ERA, at this point, to focus on AI risk and on my own research career.)

Spent ~5 hours/week managing nuclear risk research in the SERI summer ‘22 research fellowship. Also hosted a virtual fireside chat and moderated a talk+Q&A, both with nuclear security experts, at the Stanford Existential Risks Conference.

Collaborated with Rethink Priorities, an Effective Altruism-aligned think tank, under Michael Aird’s supervision, on longtermist cause prioritization research around risks from nuclear weapons.

Cambridge, United Kingdom

San Francisco, California

Comet Labs, a subsidiary of Legend Star – China’s leading early-stage venture capital fund, invests in Silicon Valley/Bay Area artificial intelligence and robotics startups.

Berkeley, California

Ba Hons, Astrophysics & Economics (Double Major), Summa Cum Laude (Top 3% Of Class)

Activities and Societies: NCAA D-I athlete (men’s golf; letterman)

11 A*S (98.9% Avg.) At Gcse, 5 A'S (99.5% Avg.) At As-Level, 4 A*S (99.1% Avg.) At A-Level