Get Paid $23,000 to Make AI Safer for Humanity

What if you could get paid to help stop AI from going rogue? A London-based program called Pivotal is offering exactly that: a 15-week, fully funded research fellowship focused on AI safety. The next session runs from January 18 to April 30, 2027, and you'll receive a stipend of £15,000–£18,000 (roughly $19,000–$23,000), plus £1,800 per month for housing, and money for food, travel, and computer power. The goal is simple: find smart people who care about AI risks and give them the time, mentorship, and resources to do something about it.

You don't need to be a professional researcher or have a PhD. Past fellows have ranged from first-year college students to experienced professionals without university degrees. What matters most is a strong fit with one of the mentors, who work on technical safety (making AI systems behave reliably), technical governance (rules for AI development), and AI governance (policies and laws). During the 15 weeks, you'll work in person in London with a dedicated mentor, a research manager, and about 40 peers. You'll produce a paper or policy brief, and you'll own the rights to your work.

The program has a strong track record. Across eight previous cohorts and about 160 alumni, fellows have gone on to work at Google DeepMind, the UK AI Safety Institute, and other top organizations. Some have even founded their own safety-focused groups. The fellowship is competitive—acceptance rates at similar programs are 1–3%—but if you're serious about AI safety, it's worth applying. The deadline is November 1, 2026. And if you refer someone who gets accepted, you earn $1,000.