The Absent Regulator at the Center of Every AI Argument

Every major AI argument happening right now has a ghost in it: the government that hasn't acted. Unions negotiating contract language to protect workers from AI-driven layoffs[¹] are doing it because no legislature has told employers they can't replace workers without cause. A police corporal using driver's license photos to generate AI pornography[²] became a Bluesky flashpoint not just because of the act itself, but because commenters immediately understood there was no federal law he'd clearly broken. The state isn't a participant in these conversations — it's the shape of the hole everyone is arguing around.

The labor displacement debate captures this most clearly. One widely circulated framing poses the question as a pure distribution problem: a 40% unemployment rate and a three-day workweek are, mathematically, the same economy — the difference is who captures the gains.[³] That's a political question, not a technological one. But without a government willing to answer it, the conversation loops back to the same abstract optimism or dread, depending on who's asking. Journalism unions seeking "just cause" protections against AI-driven terminations are essentially writing the policy that regulators won't, one collective bargaining agreement at a time.

On the misinformation side, the state's absence has created a different problem: AI denial as a rhetorical escape hatch. When Bluesky users noted that a public figure could now claim any authentic photograph was an AI fake — sardonically pointing out that some of those photos predate the technology — they were identifying something that only becomes a systemic issue without legal standards for evidence authentication. The joke lands because everyone understands the punchline: there's no institution positioned to adjudicate it.

Scientists working at the intersection of basic research and AI are raising parallel alarms, worrying that the same funding and policy environment that produced modern AI is now being dismantled before anyone has thought through what that means for the next generation of foundational work.[⁴] The concern isn't that AI development will stop — it's that the public investment infrastructure that made it possible is being hollowed out while the private infrastructure accelerates. That gap, too, is a form of government inaction: not the failure to regulate, but the failure to sustain.

What's worth naming directly is that "no policy" is itself a policy. Every beat in the AI conversation — labor, misuse, science funding, deepfakes — is shaped by the choice not to govern. The companies filling that vacuum, OpenAI and Nvidia chief among them by the volume of conversation they attract, aren't doing so because they're uniquely powerful. They're doing so because the alternative — a coherent public framework — doesn't exist. The unions writing AI clauses into contracts aren't optimistic about legislation; they're preparing for its continued absence.