> Zerzan [the agency’s general counsel] appeared interested mainly in the quantity of regulations that AI could produce, not their quality. “We don’t need the perfect rule on XYZ. We don’t even need a very good rule on XYZ,” he said, according to the meeting notes. “We want good enough.” Zerzan added, “We’re flooding the zone.”

I don’t know what to be more disturbed by: that a taxpayer-funded government wants to cede the work of governing to AI or that said government’s counsel openly admits that neither the quality nor the accuracy of proposed government regulations will matter.

"Judge, the government used a generative language model to draft the regulations, so it was perfectly reasonable for my client to interpret them with the same model and comply programmatically with code from the same model."

As an aside, the unitary executive theory implies a unitary legislature theory, so query whether agencies are constitutionally permitted to legislate like this.