November 10, 2025

New York lawmakers introduced Assembly Bill A9219 (“Bill A9219” or the “Bill”) to ensure artificial intelligence (“AI”) used in regulated professions is developed and maintained under direct consultation with licensed experts. The bill would amend the General Business Law to add a new Article 47-A titled Artificial Intelligence in Professional Fields, authorizing the Attorney General to certify compliance and investigate violations. If enacted, the measure would make New York one of the first states to require professional licensing input in the creation of AI used in law, medicine, finance, and other regulated sectors.

The legislation defines “professional domain expert” as a credentialed individual with at least three (3) years of experience in a regulated field, such as a licensed attorney, certified public accountant, physician, or educator. Developers of AI tools used in those fields would need to demonstrate that such experts were directly and substantially involved in the technology’s design, data selection, validation, and post-deployment risk review. Developers must file documentation with the New York Attorney General confirming expert participation, qualifications, and any known risks. Violations could result in penalties of up to $50,000 per incident, injunctive relief, and public disclosure of non-compliance. Enforcement authorities would ultimately rest exclusively with the Attorney General because they would be empowered to issue certificates of compliance to developers meeting the law’s requirements.

The bill aligns with growing state and federal emphasis on responsible AI use which includes the National Institute of Standards and Technology’s AI Risk Management Framework and ISO/IEC 42001 standards. Its inclusion of a “red-teaming” safe harbor encourages developers to identify and cure potential compliance issues within 60 days. The measure’s effective date would be one year after enactment, allowing time for developers and professional firms to adapt internal review processes before enforcement begins.

Bill A9219 arrives amid national attention to the boundaries of AI in regulated advice. Perhaps motivated by the proposition of the Bill, OpenAI, on October 29, 2025, amended its usage policies to state that ChatGPT cannot be used for “the provision of tailored advice that requires a license, such as legal or medical advice, without appropriate involvement by a licensed professional.” Predictably, social media platforms quickly flooded with theories and “deep dives” on what this update may entail. A recent Yahoo News report, for example, provides detailed hypotheticals of when and how ChatGPT can provide commentary moving forward while also reiterating that AI tools cannot substitute for licensed professional judgment. In a broader picture context, New York’s proposal reflects a legislative movement to clarify accountability where AI intersects with professions that carry fiduciary or ethical duties.