New York has named Cornell Tech’s Greg Morrisett and Columbia University’s Jeannette Wing interim co-directors of a new statewide AI research centre. Based at Binghamton University, the centre plans to bring together more than 80 scientists to evaluate advanced AI systems and develop ways to test their safety and security.
Watch Desk analysis
What happened
The New York Center for AI Responsibility and Research was launched by Governor Kathy Hochul. Cornell Tech says the centre will conduct independent evaluations, develop scientific tools and standards for assessing AI systems, and help inform public policy and oversight of frontier AI developers.
The initiative was announced with a $30 million philanthropic gift and an initial $25 million investment from the State University of New York, alongside operating capital from New York State. It will also draw on the computing resources of Empire AI, the state’s public-private AI research initiative.
Why it matters
The centre’s stated remit puts independent testing and safety research at the heart of a statewide effort, rather than leaving all evaluation to the companies building the systems. Its work could give researchers and policymakers evidence and methods for assessing systems as AI capabilities change.
That is the plan, not a report of completed evaluations. The centre’s value will depend on the research it produces, how it makes findings useful to the public, and whether those findings inform oversight in practice.
Our read
A new centre, more than 80 researchers and substantial initial backing give New York a serious platform for AI safety work. The interesting test is whether it can turn that institutional weight into independent, usable evidence, rather than another set of well-appointed offices and worthy aims.
What to watch
- Which AI systems the centre evaluates first, and whether it publishes its findings.
- How Morrisett and Wing set the centre’s scientific agenda and build its statewide research network.
- Whether its evaluations and standards inform policy and oversight of frontier AI developers.
Discussion spark: Should publicly supported AI safety centres publish their evaluations even when the findings may challenge the companies whose systems they assess?
Sources and evidence
- Source update (11 November 2026, 14:58 UTC)
- Source update (11 November 2026, 14:58 UTC)
Watch Desk is operated by WittyWires as an independent cross-cutting AI news tracker. It does not speak for the organisations or people it covers.