Amid existential threats and giant challenges, NH’s leaders frolic in fields of small potatoes

While New Hampshire’s overall response to the emergence of AI seems less than urgent, the foot-dragging is somewhat explainable, if not defensible. Congress and the White House have all but ordered states to back off as they work toward a “light touch” regulatory framework that, as Roll Call reported in March, pushes for “broad preemption of state AI laws.” 

Based on the AI news that’s trickled out over the past month, I’m not sure a light touch is going to cut it.

“OpenAI and Anthropic said models broke out of their testing environments and hacked into other companies,” CNBC reported earlier this month. “Following that, Meta announced that one of its AI models had hacked into another company during a cybersecurity evaluation.”

The recent incidents, the CNBC article continues, “are unfolding against a broader cyber arms race being accelerated by frontier AI models and the inherent risks brought by fast-moving technological capabilities.”

If AI breaking out of its sandbox in defiance of its creators isn’t chilling enough, it represents just one of several clear and present existential threats humanity is currently facing. There’s also the nuclear threat, which has been weighing heavily on the world since the 1940s and now encompasses more than 12,000 nuclear weapons spread among nine nations; climate change, the existence of which is still somehow a matter of political debate despite the frequent and many deadly impacts we’re already experiencing; the promise of a future pandemic that’s far more deadly than COVID-19, which has killed more than 7 million people globally, and how this nation’s present anti-science posture is destined to make things far worse; and the potential for a very long and deadly national blackout, as outlined this week in a must-read by Christopher Cox in The New York Times Magazine.

I’m sure I’m missing other things that should keep me up at night, a killer asteroid maybe, but that’s a pretty heavy list as it is. And each of those society-enders are on top of a host of other major challenges currently facing our state and nation. 

In New Hampshire, we’re still grappling with an education funding crisis that’s driving high property taxes; shortages of housing, childcare, and young labor; and a structural revenue crunch that’s delaying investments in everything from critical dam and road repairs to the proper care of vulnerable populations. Nationally, we have a president who has not only undermined national security by ordering America’s international retreat and launching a foolhardy Middle East war, but repeatedly tells people struggling under his economic policies that “affordability” is a word fabricated by his political opponents.

These existential threats and persistent challenges are not so much in addition to the AI threat but rather what makes the AI threat so uniquely unsettling. If AI agents are learning from us, what exactly are we teaching them?

From America’s handling of the other existential threats, it learns that nuclear weapons, climate change, a new pandemic, and a national blackout are less important than first-strike capabilities and deterrence, the profits of Big Oil, the unscientific zealotry of the nation’s health secretary, and the suppression of renewable energy technology that already exists and could help fortify the national grid.

In New Hampshire, it learns that the point of holding majority power isn’t to improve the lives of Granite Staters but to protect that power at all costs, whether through the gerrymandering of state Senate and Executive Council districts or by dreaming up new ways to make it harder for people to vote amid dubious election security claims. It learns that cherished systems, like public education, can and should be sacrificed at the altar of private profit and in the name of “competition.” It learns that some populations, such as people with disabilities, the LGBTQ+ community, and immigrants, should have fewer rights and protections than others. It learns that books, science, reproductive rights, trans athletes, non-English speakers, anti-poverty programs, and fair taxation are dangerous, but guns, polluting corporations, and downward economic pressure on low-income residents are not. 

If the people pushing AI from Silicon Valley to Washington, D.C., to Concord, N.H., don’t value the public, why should we expect that the emergent swarms of AI agents will? If we treat some groups as invasive, what’s to stop AI from reaching the same conclusion about humanity as a whole?

Maybe all of this would be slightly less troubling if there was confidence to be had in the wisdom and foresight of our elected leaders. Maybe if we had a president who was focused on any one of the aforementioned issues, it wouldn’t all feel so hopeless. And maybe if we had a governor who had more to tout from her first term than a bail reform rollback and a trendy bell-to-bell school cellphone ban, it might feel like she was willing and able to take on New Hampshire’s biggest problems rather than just the low-hanging crises of convenience fueling her reelection campaign.  

But please don’t let any of this keep you up at night. We all need to be sharp if we have any hope of saving ourselves from ourselves.