Gov. Wes Moore (D-MD) argued that states should build their own AI regulatory frameworks in the absence of a federal one.
“The absence of federal leadership on this is staggering, and we need it,” Moore said on Tuesday during a Clinton Global Initiative conference panel on AI.
Politicians across the country are wading into the conversation over whether artificial intelligence should be more heavily regulated, after the heads of frontier AI labs have called for a national slowdown and federal guardrails on model development. The Maryland governor spoke about the state’s hallmark AI regulation bill passed in 2024 and a new AI framework his administration introduced Tuesday, saying his state’s framework could serve as a model for other states seeking to place ethics-based guardrails on the industry.
Moore told the Clinton Foundation panel that he believes AI guardrails should be based around ethical standards, including safeguarding children from predatory algorithms, protecting human jobs, and protecting infrastructure assets. He said there should be a legal framework for the technology while AI develops, although there will continue to be political conversations about evolving AI applications and operations.
“Anyone who’s then working with our state has to understand these are our baseline values that you need to have a good framework of if this work is going to progress and move forward,” Moore said. “And it’s something that we think is also scalable for other states to be able to have as a framework as well.”
When moderator Chelsea Clinton asked Moore if Maryland has “real levers to hold actors accountable,” Moore confirmed and added that his state does so “often.” Maryland passed an initial AI law in 2024, after Moore signed an executive order on AI the same year. The order outlined how the Maryland government can work with AI, including setting up a public inventory of how state agencies use the technology in their processes.
On Tuesday, Moore announced a new framework his administration has made specifically to “mitigate the risks” of frontier AI. As part of the framework built around the ethical standards Moore outlined at the panel, the governor said his administration is working to create legislative recommendations on regulating frontier AI models.
His office listed the following guardrails as tenets of the developing recommendations: “safety frameworks and testing, worker protections, independent third-party audits and evaluations, whistleblower protections, public reporting, and 72-hour incident reporting.”
SPANBERGER UNVEILS ‘MOST AGGRESSIVE’ DATA CENTER REGULATION IN NATION
Moore is the latest governor to jump into the frontier AI conversation with a state framework, following an Anthropic researcher who warned that AI could be existential to humans within the next decade.
In neighboring Virginia, Gov. Abigail Spanberger (D-VA) introduced a framework on state data center regulations last week. She also wrote a letter to congressional leaders urging action on a “meaningful” federal AI framework on Tuesday.
Continue reading...
[ H/T Washington Examiner ]
