AI tried to bury this politician , now people have actually heard of him

The buried angle here is that direct electoral intervention represents a qualitatively different risk profile than lobbying: labs are now accountable to voters and campaign finance law, not just regulators and congressional staffers. That exposure cuts both ways, and neither Anthropic nor OpenAI has navigated that kind of public scrutiny before.

This is largely disconnected from recent activity in our archive, as we have no prior coverage to anchor it to. It belongs, however, to a broader pattern visible across the tech industry: platform companies eventually conclude that shaping the rules is cheaper than complying with them, and begin treating policy infrastructure as a product surface. What's notable here is the speed of that transition for AI labs, which are barely past their Series B vintages in institutional terms yet are already behaving like mature political actors with existential regulatory stakes.

Watch the June primary results in New York's 12th district: if the candidate backed by the lab with the more permissive regulatory stance wins, expect the other lab to escalate its own electoral strategy in subsequent races rather than retreat to conventional lobbying.