Boston University is a large private research university in Boston, Massachusetts with over 37,000 students, 17 schools and colleges, and more than $554 million in annual research expenditures. It hosts AI safety and alignment student programs and has received Open Philanthropy funding for AI safety-relevant research.
Boston University is a large private research university in Boston, Massachusetts with over 37,000 students, 17 schools and colleges, and more than $554 million in annual research expenditures. It hosts AI safety and alignment student programs and has received Open Philanthropy funding for AI safety-relevant research.
People
Updated 05/18/26Assistant Professor of Linguistics and affiliated faculty in Computer Science
Funding Details
Updated 05/18/26- Annual Budget
- $2,800,000,000
- Current Runway
- -
- Funding Goal
- -
- Funding Raised to Date
- -
Org Details
Updated 05/18/26Boston University is a private research university founded in 1839 by Boston Methodists, originally as the Newbury Biblical Institute in Vermont. It was chartered as Boston University by the Massachusetts legislature in 1869 and has grown into one of the largest private universities in the United States. The university spans three campuses in Boston covering approximately 140 acres, with 17 schools and colleges, more than 300 fields of study, and around 37,500 students. BU employs over 10,600 staff and faculty and generates more than $554 million in annual research expenditures, with sponsored research awards exceeding $645 million. It is a member of the Association of American Universities (AAU), an invitation-only consortium of North America's leading research universities. In the AI safety and governance space, BU hosts several relevant programs. The Boston University AI Safety & Alignment (AISA) community runs introductory technical and policy fellowships covering topics like reward misspecification, mechanistic interpretability, compute governance, and AI regulation. Its Verras Program supports independent research tracks for advanced students. The AI Safety & Policy Lab (AISAP) is a semester-long initiative in which interdisciplinary student teams collaborate with state legislators to translate AI safety research into actionable policy recommendations. These programs are student-led but affiliated with the university. At the institutional level, BU launched the Artificial Intelligence Development Accelerator (AIDA) under President Melissa Gilliam to coordinate AI strategy across academic and administrative functions. The Rafik Hariri Institute for Computing and Computational Science & Engineering houses the AI Research (AIR) Initiative, which focuses on intelligent systems, natural language processing, computer vision, and AI applications in public health and education. BU received an Open Philanthropy grant of $756,396 over two years to support the LLM Research Benchmark project, led by professors Najoung Kim and Sebastian Schuster, evaluating the abilities of large language models at performing academic machine learning research. This grant falls within Open Philanthropy's focus area of potential risks from advanced artificial intelligence.
Theory of Change
Updated 05/18/26As a major research university, BU contributes to AI safety through multiple pathways: training the next generation of AI safety researchers and policymakers via student programs like AISA and AISAP; producing empirical research on AI capabilities and safety (e.g., the OpenPhil-funded LLM benchmark project that helps evaluate how capable AI systems are at consequential tasks); and building institutional infrastructure for responsible AI adoption. The AISAP Lab specifically aims to bridge the gap between technical AI safety research and legislative action by equipping both students and legislators with the knowledge to craft sound AI policy.
Grants Received
Updated 05/18/26Projects
Updated 05/18/26Open Philanthropy–funded research project led by Boston University assistant professor Najoung Kim with co-PI Sebastian Schuster to evaluate the abilities of large language models on challenging tasks.
Discussion
No comments yet. Be the first to share your thoughts.