During a recent debate regarding Chinese and US AI geopolitical strategy, and how AI could be, should be integrated into society, a discussion of disconnects arose.
The one thing popped into focus as I read through was this: the disconnect between regulatory policy and AI. Let me give an example by copying text from a news story (yesterday’s New York Times, Coral Davenport, 1 May 2026):
“Child Care Chief Seeks to Slash Costs and Rules” “When Alex Adams arrived in Washington late last year as the Trump administrations point man on child care, he was little known outside his home state of Idaho, where he had helped engineer a massive deregulation effort that became the envy of many conservative activists. He made his intentions clear right away. Federal child care regulations, he told his new staff, should ’fit on an index card in my back pocket.’”
Balance this against the old saw “regulations are written in blood” – someone died, before they changed the regulations to make sure that won’t happen again.
Regulations are onerous precisely because they are long and boring. I once read all 732 pages for petroleum well–head explosion safety. Costly, because I was paid a salary to read these. My boss grumbled. He wanted me to read faster. The irony here is that with AI, the need for humans to read and memorize these regulations is sharply diminished. Build that custom AI agent, using ChatGPT, to review your engineering drawings and determine if they meet explosion safety standards. The need for human evaluation is diminished and removed. By using AI, you get explosion risk mitigation “for free”, or at low cost. Workers don’t need to die cause of badly designed electrical circuits. Safety regulations means you can go home alive, to a happy wife, happy children, and be a positive contributor to the economy. Ripping out regulations is an economic death by a thousand cuts: widows and orphans.
The old “regulations are onerous” complaint was coherent ten and twenty and thirty years ago. Today, in the AI world, it is utterly stupid and insane. Use that freakin AI as a basic safety tool. Use it to make sure your factory doesn’t explode, that some line worker doesn’t get an amputation… or in the case of Alex Adams, chief of child care, that some kid doesn’t suffocate on the plastic bag that their lunch came wrapped in. Doesn’t get hospitalized with a plastic straw stuck in his windpipe. The AI, properly integrated into the industrial base, can make sure that toddler lunches don’t cause harm.
Phrase this as a China vs. US competition: if Chinese kids are dying due to toxic food substitutes, well, that’s a shame. If US kids are dying of the same, well, that not only hurts the US economy, but is actively criminal. We’ve got the freakin AI that can be used to improve GDP and happiness and child mortality. Let’s use it. The Alex Adams policy of deregulation is some strange combination of stupid and evil. It’s running in the exact opposite direction of where the world is heading. It’s anachronistic. May as well be asking to go back to the good old days of lead plumbing and leaded gasoline.
— Linas
p.s. I write these screeds precisely cause I have no idea where else to publish. I’m not some youtube influencer, I’m not a journalist. I think insights like the above are valuable, but writing a letter to the editor is anachronistic. If I were to post this on Twitter, I would get five, maybe ten views. Facebook already kicked me off long ago – the above would count as a “personal attack on Alex Adams, a violation of facebook terms of service”.
> On Sat, May 2, 2026 at 09:33 Alvin Wang Graylin <agraylin@stanford.edu> wrote:
> For those interested in a more geopolitical view, here’s a piece that explains the current AI race and why it’s built on a series on misunderstandings. Worth a read if you have 10 minutes.
> https://centerforchinaanalysis.asiasociety.org/p/misdiagnosing-the-uschina-ai-race
Leave a Reply