HARTFORD– The Connecticut Senate on Wednesday moved forward with one of the first major legislative proposals in the U.S. to curb bias in artificial intelligence decision-making and protect people from harm, including fabricated videos or deepfakes.
The vote came despite concerns that the bill could stifle innovation, become a burden on small businesses and make the state an outlier.
The bill passed 24-12 after a lengthy debate. It is the culmination of two years of task force meetings in Connecticut and a year of collaboration among a bipartisan group of lawmakers from other states trying to prevent a patchwork of laws across the country because Congress has yet to act.
“I think this is a very important bill for the state of Connecticut. I think it’s also very important for the country as a first step to get a bill like this,” said Democratic Senator James Maroney, the bill’s primary author. “Even if it wasn’t passed into law this year, we worked together as states.”
Lawmakers from Connecticut, Colorado, Texas, Alaska, Georgia and Virginia who worked together on the issue found themselves at the center of a national debate between civil rights-oriented groups and industry over the core components of the legislation. Several lawmakers, including Maroney, took part in a press conference last week to emphasize the need for legislation and how they have worked with industry, academia and advocates to create proposed regulations for safe and trustworthy AI.
But Senate Minority Leader Stephen Harding said he felt Connecticut senators were rushing a vote on the most complicated piece of legislation of the session, which will adjourn on May 8. The Republican said he feared the bill was “full of unintended consequences.” that could harm businesses and state residents.
“I think our voters need to think about this more, pay more attention before we push the button and say this is now going to become law,” he said.
In addition to the opposition from Republican lawmakers, some key Connecticut Democrats, including Gov. Ned Lamont, have expressed concerns that the bill could negatively impact a nascent industry. Lamont, a former cable TV entrepreneur, “remains concerned that this is a rapidly evolving space, and we need to ensure we get this right and don’t stifle innovation,” his spokesperson Julia Bergman said in a statement.
The bill includes protections for consumers, renters, and employees by seeking to address the risks of AI discrimination based on race, age, religion, disability, and other protected classes. In addition to making it a crime to distribute so-called deepfake pornography and deceptive AI-generated media in political campaigns, the bill requires digital watermarks on AI-generated images for transparency.
In addition, certain AI users will be required to develop policies and programs to eliminate the risks of AI discrimination.
The legislation also creates a new online AI academy where Connecticut residents can take classes in AI and ensures AI training is part of state workforce development initiatives and other state training programs. There are some concerns that the bill doesn’t go far enough, with calls from advocates to reinstate the requirement that companies disclose more information to consumers before they can use AI to make decisions about them.
The bill is now awaiting consideration in the House of Representatives.