Skip to content
AI.info

The Pulse

Stanford HAI awards $3.5 million to seven AI research teams

Seven Stanford research teams have received a total of $3.5 million in 2026 Hoffman-Yee grants. Their projects span astronomy, law, workforce skills, medicine and education.

Stanford HAI awards $3.5 million to seven AI research teams

Seven Stanford research teams have received a total of $3.5 million in 2026 Hoffman-Yee grants for AI projects spanning science, law, medicine, education and work, Stanford HAI said on November 16. Each winning team can receive up to $500,000 in the first year, with an opportunity to compete for up to $2 million more over the following two years.

The grants are made possible by a gift from philanthropists Reid Hoffman and Michelle Yee. Stanford's Institute for Human-Centered AI says the program supports interdisciplinary research addressing scientific, technical or societal challenges.

The recipients also include projects on precision medicine, measuring the AI-enhanced classroom, self-improving AI for scientific co-discovery, and a proposed network and platform for infectious disease diagnostics.

An AI collaborator for astronomy

The astronomy team proposes a multi-agent AI co-scientist for research using the Vera C. Rubin Observatory. The researchers say the observatory will issue millions of alerts each night over the next decade as it surveys changes in the sky.

The planned system would help scientific teams preserve context as collaborators change, draw on data from different sources and coordinate follow-up observations. The team says it wants the system to expose its evidence to scrutiny and trace results to data and code.

The team plans to demonstrate the co-scientist on four scientific cases and release the agent, evaluation framework and benchmark suite as open community resources. It also intends to study whether the system accelerates discovery and whether scientists retain agency when using it.

Risa Wechsler is the project's main principal investigator. The co-investigators are Angèle Christin, Emily Fox, Ben Nachman and James Zou.

Law and workforce projects

The law and policy project, led by Dan Ho, focuses on AI reasoning in settings where legal questions may have no single definitive answer. Its proposed work includes methods for reasoning amid ambiguity or conflicting views and for synthesizing information in an interpretable way.

The team also aims to create tools that assist lawyers and policymakers in areas such as housing policy and benefits adjudication. It proposes piloting the systems with government partners while studying their integration into legal work.

A workforce project proposes a regularly updated framework of skills, using AI-based interviewers and analysis of workflows to identify emerging capabilities. Its researchers plan to develop AI training methods that reward collaboration and defer to people on high-stakes decisions.

They also propose models to identify workers' skill gaps and guide personalized training. For evaluation, the team plans a publicly accessible economic observatory updated quarterly, alongside randomized controlled trials in occupations highly exposed to AI.

How follow-on funding works

Teams must compete for funding in the second and third years through a public symposium presentation, a private interview and a progress report, according to Stanford's call for proposals. Only a subset will be selected for subsequent funding.

Six months after receiving funds, each winning team is expected to present its work at an HAI directors meeting. The call also asks funded projects to acknowledge the institute's support in resulting publications, talks, presentations and websites.

Sources

Related stories

Explore

More articles