Skip to Content, Navigation, or Footer.
The Tufts Daily
Where you read it first | Wednesday, November 29, 2023

Let’s be smart about regulating AI

Massachusetts is a leader in higher education and scientific discovery, as some of the world’s most prestigious universities and research labs are located in Greater Boston. Companies like Boston Dynamics have been on the forefront of artificial intelligence development since its inception and have been continuously pushing the boundaries of science. Boston’s dominance in the technology world must be coupled with a leading role in responsible use and growth of AI.

The AI most people are familiar with may look like a Tesla car or a Boston Dynamics robot, but the reality is far from the futuristic sci-fi world we see on the silver screen. AI algorithms are everywhere and are making important life decisions for people across the country. The use of AI in fintech, for example, has led to discrimination in consumer lending with people of color having significantly higher mortgage rates compared to their white counterparts. This is but one example of the ways that AI algorithms can perpetuate systemic racism. These very same algorithms are used by the police and government institutions at various levels.

This issue is far from new, but legislators are only now taking notice. Rep. Ted Lieu, D-Calif., introduced the first piece of federal legislation written by AI in late January. His introduction of House Res. 66, which advocates for an increased focus on AI from the U.S. House, was largely symbolic, but it represents a sea change in Congress. Closer to home, Rep. Jake Auchincloss, D-Mass., gave a speech on the floor of the House written by ChatGPT. Meanwhile, here in Massachusetts, state Sen. Barry Finegold, D-Second Essex and Middlesex, introduced the first bill written using AI. Clearly, the political will is there. With a Republican-held House and AI proposals being introduced solely by Democrats, there is little chance for substantive policy on AI for now. However, individual states can — and must — step up and act where Congress cannot.

Here in Massachusetts, the election of a new attorney general, Andrea Campbell, presents an opportunity to lead. The adoption of Sen. Finegold’s bill would grant her office extensive regulatory and oversight powers over AI through algorithmic auditing and by implementing the proper enforcement mechanisms to ensure compliance. While the bill provides an excellent framework, it fails to regulate most algorithms since it is narrowly focused on only “generative artificial intelligence models.” In other words, it centers on platforms like ChatGPT that produce something when prompted. While legislators are almost exclusively focused on ChatGPT, we must go further to regulate all forms of AI algorithms.

To do this, Massachusetts should establish a new independent agency to regulate technological platforms. The current regulatory framework is far too narrow, and the state legislature is far too clunky and lacking in expertise to successfully regulate AI. A new agency made up of experts will be savvy enough to make recommendations to the legislature and the governor, while also having the appropriate rulemaking ability. While the structure of this agency can and should be debated, the importance of such an agency cannot be understated.

AI technology may explode over the next few years, or it may not. We cannot possibly predict the future, but that is why an independent agency is critically important. Twenty years ago, cell phones were little more than devices that could send and receive messages and calls, but today, they’re pocket-sized supercomputers with enough processing power to send someone to the moon.

No other state is better positioned to address this issue than Massachusetts. It is high time that Massachusetts reclaims its mantel as a “city on a hill,” so Beacon Hill must act when Capitol Hill cannot.