Improved AI Governance Required in New York State, According to Report
New York’s Need for Improved AI Governance
Overview of the Report
New York State’s approach to artificial intelligence (AI) governance is under scrutiny. A recent audit from the State Comptroller’s Office has highlighted significant gaps in oversight and guidance for AI use among various state agencies. The report was released just as the state’s first Chief Artificial Intelligence Officer, Shreya Amin, began her tenure.
Key Findings from the Audit
The audit, which covers the period from January 2019 to November 2024, assesses AI implementation across four main agencies:
- Office for the Aging
- Department of Corrections and Community Supervision
- Department of Motor Vehicles (DMV)
- Department of Transportation
It emphasizes that the existing guidance for how these agencies utilize AI technologies is insufficient. According to the report, the lack of centralized oversight raises the risk of irresponsible technology use, causing potential data breaches and service delivery inequities.
Lack of Centralized Guidance
The Comptroller, Thomas DiNapoli, pointed out that New York has only a brief eight-page policy regarding AI. This document provides limited detailed instructions, essentially advising agencies to reference federal policies for more information. This oversight has led to confusion about what constitutes responsible AI use.
Insufficient Training and Oversight
The audit also identifies a critical gap in training for staff managing AI systems. Without adequate training, there is a higher likelihood of noncompliance and unintended consequences resulting from AI applications. The report specifically notes that New York’s current AI policy addresses bias management in only two sentences.
Specific Agency Shortcomings
The audit examined the aforementioned agencies in detail and revealed several shortcomings:
- DMV: The agency has exempted its facial recognition software from oversight, contradicting directives from the Office of Information Technology Services (ITS).
- Department of Corrections: There are no defined processes in place to manage AI risks, nor does the department’s contract with the vendor for inmate phone monitoring address potential biases.
- Transportation Department: Along with the other examined agencies, it has not conducted regular reviews or audits of its AI tools to ensure their accuracy and fairness.
National Context
Other public agencies in the United States are also developing their own AI policies, reflecting a trend of increasing scrutiny and governance around technology use in government. New York’s challenges are not isolated but part of a broader national conversation about responsible AI deployment.
Appointment of the Chief AI Officer
In response to these findings, Shreya Amin has been appointed as New York’s first Chief Artificial Intelligence Officer. With nearly twenty years of experience in data and AI, she aims to establish ethical guidelines and enhance service delivery through technology. Amin has committed to working collaboratively with state agencies and communities to make meaningful improvements.
Moving Forward
The Comptroller’s report calls on the Office of Information Technology Services to strengthen AI governance by developing clearer policies and improving staff training. Furthermore, it recommends that specific agencies create governance structures tailored for AI to align with ITS guidelines.
In particular, the DMV has been advised to reassess its facial recognition system to ensure compliance with state AI policies. This evaluation aligns with the need for continuous adaptation as AI technology evolves.
As the state of New York takes steps to address these deficiencies, the focus remains on utilizing AI responsibly and effectively to serve its residents while minimizing risks associated with the technology.
