Opinion: The window of opportunity for action on AI is closing

Right now, House Republicans are trying to push a provision through the federal budget reconciliation process that would prevent any state-level enforcement of artificial intelligence rules for 10 years. This is just the latest in a string of actions by Republicans that are rolling back key protections and significant advancements in AI policy, like revoking the Biden administration’s 2023 executive order on “Safe, Secure, and Trustworthy Development and Use of Artificial Intelligence,” and inviting Big Tech to contribute money and influence to federal actions. This is a dangerous and shortsighted attempt to preempt state action on AI.

The potential for AI-powered tools to improve the lives of everyday people should not be understated. For instance, the Pew Research Center estimates that already, 1 in 6 U.S. workers are already using AI daily in their workplace. Moreover, AI-powered assistive technologies are improving accessibility for people with disabilities. However, generative AI technology is still far from perfect, and vulnerable populations, including people with disabilities and people of color, are most at risk from harms like algorithmic bias and inaccuracy.

We cannot let Big Tech dictate what safeguards are imposed on the most critical technological advancement since the internet. We’ve already seen a shift away from safety from tech companies driven by a desire to maximize profit over protecting consumers. For example, OpenAI, the creator of ChatGPT, was originally founded with the goal of ensuring that AI benefits all of humanity, but it has removed that sentiment from its mission statement.

History shows us that when it comes to policymaking on emerging technologies, there are finite windows for decisive action. Right now, the window for action on AI is closing. This technology is already being used in high-risk use cases like evaluating job applications, determining insurance coverage for certain procedures and mortgage approvals. If these AI-powered tools hallucinate or make an error in a determination, consumers rarely have meaningful recourse.

While Congress fails to act, state governments have been leading the way. In New York alone, more than 60 bills related to automated decision-making systems are moving through both chambers and onto the governor’s desk. Last year, while the federal government failed to take action on deepfakes, algorithmic bias or facial recognition technology, New York enacted the FAIR Act to regulate the use of deepfakes in elections, banned the use of facial recognition technology by the MTA and instituted guardrails on the state government’s use of automated employment decision-making tools.

There is still an opportunity to shape this technology for good, and we cannot let Congress prevent state and local governments from protecting as many Americans as possible from risky AI development. At the state level, we are continuing the work to realize the promise of AI as a

benefit for all. In New York, I am proud to carry bills like the New York AI Act (S1169), which would ensure AI is applied fairly and developed responsibly in consequential contexts; a proposal that prohibits chatbots from practicing licensed professions like psychiatry; and regulation on how the government uses AI. I also carry bills that promote a positive vision of AI, like a competitive grant program for under-resourced schools and community organizations to participate in AI education, so students can understand the technology and use it in their professional lives.

In response to this proposed moratorium, I joined state legislators across the country on a joint letter and informational campaign on the dangers of leaving AI technology unchecked for a decade and called for our constituents to hold their federal reps accountable. I also organized a letter, signed by more than 50 of my fellow New York state lawmakers, calling on House Republicans to remove the proposed moratorium from the federal budget bill.

Our congressional representatives need to know how AI is already impacting our lives, revoke this dangerous 10-year moratorium and take action by passing comprehensive laws to address AI at the federal level now. The future of AI can be something we can feel proud of and confident about, rather than something we fear. But that requires taking bold action now.