Database

Alparslan Bayrak is an effective altruism community builder in Turkey, founding EA Bilkent and EA Ankara, mentoring in the Open Student Program, and working to launch effective animal advocacy projects.

No summary available yet.

No summary available yet.

Retrospective funding of salary for up-skilling in infrabayesianism prior to start of SERI MATS program

Avinash A. Independent Researcher | Formal Methods & AI Safety, a mathematical researcher specializing in the structural limits of AI alignment. The architect of the Terminal Boundary Systems (TBS) framework, which applies Category Theory to identify fundamental "safety ceilings" in agentic AI. My core work includes the ASE (Absolute Self-Explanation) Impossibility Theorem, a formal proof using Symmetric Monoidal Closed Categories and Lawvere’s Fixed-Point Theorem to demonstrate why total internal transparency is mathematically unreachable. Currently, focused on the Agda formalization of these results to provide a machine-verifiable "Axiomatic Audit" for frontier AI labs. Research aims to bridge the "missing link" between categorical logic and robust, human-centric AI autonomy.

No summary available yet.

Special Projects Manager at the Berkeley Existential Risk Initiative since February 2025. She also works in web and data at Duke University, previously managed educational programs at the Institute for Defense & Business and the Warrior-Scholar Project, and holds a B.A. in Psychology from the University of North Carolina, where she was a Chancellor’s Fellow.

Research to enable transition to AI Safety

Stanford Artificial Intelligence Professional Program tution

No summary available yet.

4 month salary to support an early-career alignment researcher, who is taking a year to pursue research and test fit

No summary available yet.

Systems architect for technological sovereignty, designing frontier R&D programs at the intersection of materials science, synthetic biology, and AI to help states and institutions turn high-uncertainty science into sovereign capabilities.

No summary available yet.

Head of Events at the Safe AI Forum (SAIF). Previously served as Events Producer for The Alan Turing Institute and has held positions with the Australian High Commission in London, the Edinburgh International Culture Summit, Georgetown University, and the British Film Institute, specializing in complex multi‑partner events that tackle global challenges.

6-month support for self study and development in ML and AI Safety. Goals include producing an academic paper while working on the "Inducing Human-Like Biases in Moral Reasoning LMs" project run by AI Safety Camp.

Identifying operational bottlenecks and cruxes between alignment proposals and executable governance.

No summary available yet.

6-month salary to upskill for AI safety

A 501(c)(3) nonpartisan think tank that bridges technology and national security policy, with major programs addressing ransomware, frontier AI security, and the catastrophic risks posed by emerging technologies to nuclear stability.

No summary available yet.

No summary available yet.

No summary available yet.

No summary available yet.

Translating an AI safety report (1k+ downloads) for peer-reviewed publication to formalize "Emergent Depopulation" as a novel systemic risk.

No summary available yet.

Ex-climate tech entrepreneur building a media company to address the Metacrisis

work title: Seductive Machines and Human Agency

No summary available yet.

A nonprofit R&D lab that develops collective intelligence tools and governance models to steer transformative AI development toward better outcomes through democratic public input.

Florian Dietz is an artificial intelligence researcher and data scientist pursuing a PhD in AI at Saarland University's Spoken Language Systems group. Before starting his doctorate, he worked as a consultant and freelance data scientist and built a startup based on an AI system designed to automate software and data science tasks.

No summary available yet.

Drew Spartz is Head of the Incubation Program at Nonlinear. He previously founded Superlinear, an AI safety bounty platform, the Nonlinear Network funding platform, and a digital media company, and has also helped manage a family office. The team bio notes that he is an avid reader and traveler who has visited more than 50 countries.

Princeton University is a leading Ivy League research institution that conducts significant AI safety and AI governance research through several interdisciplinary centers and initiatives.

No summary available yet.

No summary available yet.

No summary available yet.

6-month salary to work on the research I started during SERI MATS, solving alignment problems in model based RL

No summary available yet.

Patrick Stadler works on shaping a safe and flourishing future at the Swiss think tank Pour Demain. He co-founded and chairs the board of the GiveWell-recommended nonprofit New Incentives, which he helped scale to over 100,000 users for its vaccination program in Nigeria. Previously, he served as a political advisor for Switzerland’s economic development agency and worked on strategic communications for peacebuilding and mediation at the United Nations.

Pivotal Research runs a 9-week in-person research fellowship in London for early-career researchers working on AI safety, AI governance, and biosecurity. Fellows work alongside mentors from leading organizations to produce impactful research and launch careers in reducing global catastrophic risks.

AI and neurotech advisor to Lionheart Ventures and Chief Scientific Officer at the Flow Research Collective, where he applies deep learning and distributed machine learning to understanding and training optimal human performance.

Nova DasSarma is the financial director and co‑founder of Hofvarpnir Studios and a systems leader at Anthropic, where she works on large-scale training and infrastructure. Her background includes systems administration at the NIH, engineering roles at several Y Combinator startups, and a BS in Information Systems from the University of Maryland, Baltimore County.

6-month salary for self-study to be more effective at AI alignment research

No summary available yet.

generative.ink is the personal research and creative platform of Janus (also known as "moire" and "@repligate"), a pseudonymous AI safety researcher known for the Simulators framework and the Loom human-AI collaboration tool.

I study the grammar of self-organizing movements: the structural logic that enables them to share risk, generate trust, and spread ecological knowledge in ways that transform financial flows from dissipative to conservative.

Head of AI Safety, Arcadia Impact