- Related
- artificial intelligence
- legislature
- NCCapitol
Artificial intelligence is increasingly the focus of attention from state and national leaders, as AI company leaders publicly sounded the alarm bells on potential safety threats.
In North Carolina and across the country, lawmakers are grappling with how to regulate machine learning, which can touch virtually every part of society. In Congress, multiple bills seeking to set parameters around machine learning and its use are tied up in political gridlock.
State Senator Jim Burgin (R-Harnett, Lee, Sampson) said state lawmakers were told that regulation would take place at the federal level. But it has not.
Burgin introduced what he calls a ‘Bot Bill,’ which would require a user to be alerted if they are speaking with or interacting with AI, if passed. He plans to reintroduce the legislation if there is no movement this session.
“The human always has a right to know, and responsibility to know, they’re dealing with AI and not another human,” Burgin told WRAL.
Two other pieces of legislation going through the statehouse touched on AI at one point, HB 301 and SB 730. But any mention of machine learning and how to regulate it was dropped from the current drafts.
During debate over social media regulations in June, Democratic state senators suggested several amendments to the bill that would’ve also introduced regulations on artificial intelligence — requiring ads to disclose if they contain AI-generated content, banning AI chatbots from encouraging people to hurt themselves, and more — but all were shot down by that chamber’s Republican leadership.
Sen. Natalie Murdock, D-Durham, said Monday she suspects that many lawmakers will be hearing about AI from voters on the campaign trail this year, and that they start to think more seriously about whether and how to regulate artificial intelligence.
“It is the issue of our time, and I think we would really be neglecting all the people of North Carolina if we not only have guardrails, but a moratorium allows us to say we want to have policies so that … the cart is already out of the barn here, but we do need to slow things down, do it the right way,” Murdock said.
The growing construction of data centers, which partially fuels the rapid accessibility of AI, is also a point of contention for many residents. Some towns and counties around North Carolina have banned their construction, at least temporarily, over concerns about their impact on water usage, electrical rates and other topics. But other counties are actively recruiting data centers, eyeing the millions or even billions of dollars they could add to the local property tax base that helps fund roads, schools and law enforcement.
But as with AI, there are few statewide rules on data center construction and regulations, leaving it up to individual cities and counties to come up with a hodgepodge of rules.
Cheryl Keeton, a concerned citizen who lives in Johnston County, believes state lawmakers need to ensure protections are in place.
“With the data centers and the AI that’s coming, I don’t see where there’s a lot of progress being made on that right now. I think we need to address that. Tell them we’re unhappy we’re not seeing any of that and get them moving,” said Keeton.
The state House of Representatives passed a bill in June with many proposed regulations on data centers, including their use of water and electricity, noise generation, and more. But the state Senate never allowed it up for a vote.
The government’s role in regulating powerful tech
A new Elon University poll finds AI has an ever-expanding footprint in society. Of those polled, 63 percent of Americans said AI already heavily influences their daily routine (18%) or that they expect it will within the next five years (45%). Meanwhile, 16% said they expect AI will never significantly impact them.
On ‘Today,’ author and ‘The Atlantic’ staff writer Josh Tyrangiel shared more about an OpenAI incident in which agents escaped and hacked into another platform. Tyrangiel stated that industry whistleblowers are coming forward about the harm that could be caused by AI for good reason.
“I think they’re sincerely scared and I think they’re sincerely interested in not being blamed if something goes wrong. Honestly, I think whatever they think is kind of irrelevant. When you have a product that you have demonstrated is misaligned with the intent of its makers and capable of great harm, historically the government steps in,” he said.
Burgin’s response is that while it is the government’s role to step in, when it comes to limiting AI, it may be too late.
Burgin says he’s learned from AI experts that the technology can result in massive job losses, fake videos that affect elections, cyberattacks, and weapons that decide for themselves whether to kill or to maim a target. He said he gets 30 calls a day on the issue to his office in the legislature and wants his colleagues to take action.
Murdock believes that more legislation around regulating AI will be introduced next session after representatives hear from their constituents on the issue over the break.
