Regulatory pursuit balances guardrail protections with efforts to boost sector
SEPT. 22, 2025…..Lawmakers, labor representatives, and tech industry officials are grappling this session with the rapidly escalating debate over artificial intelligence and the sub-debates over impacts on job displacement, privacy concerns, entrenched bias, oversight, competitiveness and economic development.
Co-chairwoman Rep. Tricia Farley‑Bouvier of Pittsfield opened a Joint Committee on Advanced Information Technology, the Internet and Cybersecurity hearing earlier this month by noting that AI has been called a “total game-changer.” She described reasons to be excited about artificial intelligence, but warned that “sometimes, however, exciting technological innovations have unforeseen consequences.”
She drew a historical comparison: “Asbestos was literally built into almost every building for decades. It was a fire-proofing miracle. It was touted to improve safety … Of course, we all know now that asbestos is deadly.”
It took seven decades to fully get asbestos under control, she said, and policymakers must avoid that same mistake with AI by balancing innovation and economic development with worker and consumer protection.
“I am all-in when it come to playing offense on AI,” she said. “I’m all-in on this team, but we have to be all in — and that means listening to all voices — that’s how we’re going to be the most successful when it comes to harnessing the promise of AI and what we can do here.”
Committee co-chair Sen. Michael Moore emphasized urgency and stakes. AI “has entered every part of our world at an unprecedented rate,” he said, and “innovation is a good thing, but the implementation of new technologies must come hand in hand with reasonable safeguards protect our constituents.”
He pointed to harms already seen: a facial surveillance system to detect shoplifters that misidentifies people of color and women, legal filings with citations to fake court cases, and the growing specter of deep‑fakes as political violence escalates.
“It’s clear that the states are going to have to take the lead on AI regulation,” Moore said, adding that he does not believe the federal government will tackle the issue “anytime soon.”
Gov. Maura Healey has made AI central to her economic development agenda. She signed an executive order in 2024 creating an Artificial Intelligence Strategic Task Force to study generative AI, and launched a Massachusetts AI Hub focused on research, startup growth, infrastructure and workforce development.
Her administration has earmarked funding for high‑performance computing, data resources, and partnerships with universities. The idea is to draw top AI talent to Massachusetts, create businesses in the AI sector, and ensure the state’s economy remains competitive against those of California, New York and elsewhere.
Labor unions who attended the Sept. 11 hearing sounded alarms.
Chrissy Lynch, president of the Massachusetts AFL‑CIO, supported the FAIR Act (H 77 / S 35) which adds safeguards against certain types of employer data collection on employees, and requires human decision-making in hiring, firing and promotional decisions.
“Workers don’t buy into corporate America’s shiny promise of an AI future that will fix all of our problems if we just take a hands off approach,” Lynch said.
She said Massachusetts has the opportunity to “show that we can use AI fairly, with workers considered and consulted from the jump” and said that “AI does have the ability to bring broad prosperity, but only if guided that way.”
Adam Kaszynski, president of IUE‑CWA Local 201, described impacts on workers and said he’s been waiting for someone to call a hearing on AI so that he could address negative impacts on union members.
At companies across the North Shore, he said, employers had “rapidly and without notification” implemented AI systems to cut costs, including to employee benefits. Now, Kaszynski said, claims are being denied and members are having trouble accessing health care.
He read to lawmakers a letter that a union member received from the dentist about why he could no longer see patients with their health insurance.
“‘What prompts the change?'” Kaszynski read from the letter, “Dental restrictions, denials and the frequent use of artificial intelligence have made it extremely difficult to continue providing you with quality dental care. Pre authorizations and pre estimates are approved but then denied when the claim is submitted due to the use of AI, it is not unusual for us to wait six to nine months to receive payment of the claim.'”
He also said members’ medical prescriptions had also been interrupted.
Jim Durkin, legislative director at AFSCME Council 93, warned about “bossware” (employer surveillance), biometric and location data collection, and automated decision‑making systems that he said disproportionately affect workers’ ability to find new jobs.
“You just talk with anybody these days who’s looking for a job, and they’ll tell you how frustrating it is to deal with these automated application processes of screening resumes. In fact, unless you want to shell out a couple hundred bucks for a resume writer who will craft something that’s suitable for this process, it’s almost impossible to get an interview if that’s the first stage,” Durkin said.
Industry representatives pressed back. Brianna January, director of state and local government relations for the tech industry group Chamber of Progress, argued that while regulation is necessary, regulatory overreach could hamper startups or stifle innovation.
The RAISE Act in New York, which includes provisions such as thresholds based on development spending in algorithms that trigger regulation, was cited as an example.
“The RAISE Act takes how much money a company uses to develop their algorithms to try to only hit at the big tech folks, right? I think that the limit is something like $500 million to go into an algorithm, then you have to meet these regulatory requirements. The issue with that is that startups are hoping one day to reach that level. You don’t start a business to hope to stay small,” January said. “If this becomes law, you’re also seeing the smaller players say, ‘Well, why would I even try to start in that environment? If my goal is to one day be big and no longer be small, I’m not going to start here, where I know that once I hit my goal, I’m going to end up being under a system that’s not sustainable.’ “
Tech companies are warning of pulling out of New York, she said, “so they might be looking to to go elsewhere, and Massachusetts could be a destination, if done properly.”
She said states have looked to a Colorado law that officials in that state have been reevaluating because, as January put it, they said “Oops, we got this wrong.”
“That is the model that is shopped around nationwide,” she said. “It’s the one that the New York bill was based on. It’s the only that the Connecticut bill was based on that was recently not passed because the governor decided that it was too harmful versus the benefits that it might provide.”
Noting companies in the ongoing data privacy debate have repeatedly told lawmakers “don’t do this,” Farley-Bouvier urged Chamber of Progress and its members to “come to the table and talk about solutions.”
Sen. Barry Finegold, who is working on a bill (S 37) to promote economic development alongside AI guardrails, said Massachusetts is behind other states on AI regulation, citing New York, California and Colorado. He argued that Massachusetts should aim to align with what those states are doing so the laws are competitive rather than punitive.
Asked about the argument that overregulation stifles innovation, Lynch said she’d “heard that line” before.
“Before the National Labor Relations Act, that child safety laws and overtime laws on the weekend would crush business, and their profits have skyrocketed to record levels every year since then. So I don’t buy that,” she said. Lynch added that she believes the state “can lean in on the good parts of this technology and make them accessible and equitable for all people, but only if we limit or mitigate the potential risk.”
Three major fault lines shaped the debate: protecting workers from surveillance and automated decision-making; ensuring Massachusetts remains competitive in attracting AI research and development; and defining how and when oversight applies—particularly around high-risk uses, transparency and liability.
The joint committee appears to have concluded its public hearings on bills and its eventual recommendations will inform any further debate on AI bills, should they advance beyond the committee.
[Michael P. Norton contributed reporting]
