New York establishes independent AI testing program
Researchers will assess advanced AI systems, but their access will be limited to information developers disclose publicly.
Governor Kathy Hochul has launched an independent testing program for artificial intelligence in New York, assigning researchers to examine advanced systems and their risks. Hochul's office said the findings would give public officials evidence to support their decisions.
Binghamton University will host the effort through the New York Center for AI Responsibility and Research. The work will involve more than 80 scientists from New York, drawing computing capacity from the state's US$500 million Empire AI public-private partnership.
The governor's office said researchers would also explore responsible uses of the technology for medicine and scientific discovery. However, the testing program can examine only information that developers make public, leaving unreleased models outside its reach.
Under the RAISE Act signed by Hochul, large AI developers face a January 1, 2027 deadline to register with New York. Beginning on that date, they must make transparency reports and safety frameworks public. They must also report critical safety incidents within 72 hours.
