Application

AI Safety Law & Policy Research Fellowship

Before You Begin

The application takes about an hour to an hour and a half. Plain, clear answers are fine: we are evaluating your thinking, not your prose. Word limits are maximums, not targets.

Eligibility. Open to all Harvard affiliates. Designed mainly for graduate students, with one or two places for exceptional undergraduates.
Selection. This written application, then interviews for shortlisted candidates.
Stipend. $4,000: $1,000 at the start and $3,000 on completion of the final paper.
Questions. Email [email protected]

Your answers are saved in this browser as you type, so you can come back later.

About You

Full namerequired

Emailrequired

Harvard schoolrequired

Harvard Law School Harvard Kennedy School Graduate School of Arts and Sciences School of Engineering and Applied Sciences Harvard Business School Harvard College (undergraduate) Other

Degree program and expected graduation yearrequiredFor example: J.D., 2028

Link to your CV or résumérequiredGoogle Drive, Dropbox or similar. Set sharing so anyone with the link can view.

Link to a writing samplerequiredSolo-authored legal or policy writing is preferred. Ideally 10–20 pages, though a shorter piece is fine. Set sharing so anyone with the link can view.

LinkedIn or personal website

Background and Motivation

Short, direct answers are best.

Why do you want to do this fellowship, and how does it fit your career plans?required

Which open problems in AI safety law and policy do you think matter most, and why?required

Describe your prior engagement with arguments about catastrophic risks from AI.required

III

Your Proposed Paper

Fellows write an original 6,000–12,000-word paper on AI safety law and policy, with a particular focus on catastrophic risk. Your topic can change after admission. Here we want to see how you frame and scope a question. We strongly encourage you to pitch your own idea, if none of the topics below appeal.

Which of these illustrative topics interest you?Select any.

Antitrust waivers or exemptions for AI labs that jointly negotiate minimum safety standards Designing, structuring and delegating to a federal AI regulatory agency Whether frontier AI companies should be required to report users who show signs of planning mass-casualty attacks A safe harbor protecting independent AI safety testers from CFAA, contract and tort liability Regulating AI “character” traits consistent with the First Amendment Choice of law and liability for AI agents Securities disclosure as a tool for transparency about catastrophic-risk practices at public AI labs Codifying DoD Directive 3000.09’s “appropriate human judgment” standard Other

Working titlerequired

What question will your paper answer, and why does it matter for reducing catastrophic risk from AI?required

Write a hypothetical abstract. What will your paper argue?requiredIt is fine if you are not yet sure of your conclusion.

How would you approach the paper over 12 weeks? Give us a sense of your research process.required

Who should read this paper, and where might it be published?required

What is the strongest case against your argument, or against this paper being worth writing?required

You'll see a confirmation here once it's received.

Application Received

Thank you for applying. We will be in touch about interviews.

Questions: [email protected]