Project
Support to create language model (LM) tools to aid alignment research through feedback and content generation
Led byLogan Riggs Smith
Endorsed by
Database
Loading results...
Loading results...
Showing 451-500 of 4527 results
Support to create language model (LM) tools to aid alignment research through feedback and content generation
James Bregan is the co-founder and CEO of Constellation Institute. He previously held senior technology leadership roles at PayPal, where he helped scale the company from roughly 100 to 10,000 employees and served as EVP Engineering during its acquisition by eBay, and he has spent much of his career building startups in both the nonprofit and for-profit sectors.
No summary available yet.
No summary available yet.
Melanie Plaza is CTO at AE Studio with more than 10 years of experience building products and leading teams for early-stage startups and tech companies; she previously served as CTO at To The Tens, co-founded ELIX, worked as a full-stack developer at several LA-based startups, and holds a B.S. from Yale University focused on statistical analysis of ecological systems.
No summary available yet.
Sustaining and Scaling a Grassroots Research Collective for Neural Network Interpretability and Control
Funding for salary and living expenses while continuing to develop a framework of optimisation.
No summary available yet.
No summary available yet.
Leads Lightcone Infrastructure, whose main product is LessWrong, a platform that has significantly shaped discussions on rationality, AGI risk, COVID-19, existential risk, and crypto compared with other similar communities.
Funds to cover speaker fees and event costs for EA community building tied in with my MA course on longtermism in 2022
No summary available yet.
PhD student in the ML4STS Lab working on task-level fairness and fair feature selection in machine learning systems.
Krzysztof Gwiazda is a Polish individual pursuing entry into AI safety research, specifically mechanistic interpretability. In Q3 2024, he received a $5,000 grant from the Long-Term Future Fund to support a two-month period of upskilling in mechanistic interpretability, with the aim of completing two to three projects in the field before exploring adjacent areas. His background appears to be in software and computer science. He represents an early-career pathway into technical AI safety research through self-directed study supported by EA-aligned funding.
No summary available yet.
No summary available yet.
No summary available yet.
Support the growth of an international AI safety research and talent program
Lethal Intelligence is an AI risk awareness media project producing original explainer films, podcasts, and social media content about the existential dangers of advanced AI systems.
A French nonprofit research organization working alongside government institutions to address the security and international coordination challenges posed by general-purpose AI development.
Antony (Ant) Rowstron is a computer systems researcher serving as a senior technology leader at ARIA, after more than two decades at Microsoft Research where he was a Distinguished Engineer. His work has spanned storage, networking, distributed systems, and optical and robotics technologies for cloud data centres.
Nancy Staudt is vice president at RAND and the Frank and Marcia Carlucci Dean of the RAND School of Public Policy, where she is leading efforts to expand the school’s impact, grow its student body, and strengthen its role in training the next generation of policy leaders. A nationally recognized scholar in tax, tax policy, and empirical legal studies, she previously served as dean and the Howard and Caroline Cayne Distinguished Professor at Washington University School of Law and earlier held senior academic leadership roles at the University of Southern California, including vice dean at the Gould School of Law and founding codirector of the Schwarzenegger Institute of State and Global Policy.
A university research lab at the University of Rhode Island directed by Dr. Sarah M Brown, studying how machine learning interacts with complex socio-technical systems, with a focus on fairness of automated decision-making and AI safety evaluation.
No summary available yet.
No summary available yet.
No summary available yet.
No summary available yet.
No summary available yet.
Exploring the feasibility of circuit-style analysis on the level of SAE features (MATS extension)
Rationalist - Member of Pause IA (FRANCE)
Co-Founder, Techplomacy Founation
No summary available yet.
1yr stipend to make videos and podcasts about AI Safety/Alignment, and build a community to help new people get involved
Marc Singer is co-founder and Managing Partner of Osage University Partners, a venture firm investing in university spinouts, where he oversees technology investments and has over 30 years of experience in venture capital.
in the sway of the rainbow serpent
No summary available yet.
6-month salary to interpret neurons in language models & build tools to accelerate this process. The aim is to understand all features and circuits in a model and use this understanding to predict out of distribution performance in high-stake situations.
No summary available yet.
No summary available yet.
Gideon Futerman is a researcher focused on AI safety and existential risk, currently working as a Special Projects Associate at the Center for AI Safety (CAIS) and as a MATS (ML Alignment Theory Scholars) scholar on gradual disempowerment research. He studied Earth Sciences at St Edmund Hall, University of Oxford, where his academic interests first led him into existential risk research through the lens of solar radiation modification (SRM) and climate change. He has been affiliated with the Centre for the Study of Existential Risk (CSER) at Cambridge, contributing to work on how SRM interacts with global catastrophic risk scenarios. His AI safety research spans governance and policy approaches to advanced AI, including analysis of pathways to slowing AI development, international coordination to avoid an artificial superintelligence race, and the systemic risks of gradual human disempowerment as AI capabilities increase. He writes on these topics at his Substack and has published at AI Frontiers, The Oxford Scientist, and co-authored an SSRN paper on escalation pathways in the age of solar geoengineering.
No summary available yet.
Nicky Case is an independent Canadian creator specializing in interactive explorable explanations and educational games. Born in Singapore and raised in Canada, they dropped out of UBC's Computer Science program and worked briefly as a software engineer at Electronic Arts before becoming a full-time independent creator funded through Patreon. They are best known for projects such as "Parable of the Polygons" (a simulation on systemic bias), "The Evolution of Trust" (an interactive game theory explainer), and "Adventures with Anxiety," all of which aim to help people understand complex systems through play. In the AI safety space, they received a Long-Term Future Fund grant for a one-year stipend to create accessible explainers on AI alignment, which resulted in the "AI Safety for Fleshy Humans" series hosted at aisafety.dance, developed in collaboration with Hack Club. In early 2025 they also participated in the MATS AI Safety Research Bootcamp in London.
No summary available yet.
No summary available yet.
A research-focused think-and-do tank that conducts empirical research across animal welfare, global health and development, AI, and other cause areas to uncover high-impact, neglected opportunities for improving the lives of humans and animals.
Brendan Steinhauser is the CEO of The Alliance for Secure AI Action and a partner at the public affairs firm Steinhauser Strategies, where he draws on a long career in political strategy and communications.
