Congress is quietly becoming an AI customer — and OpenAI is getting most of the business.
According to TechCrunch, House spending records show that ChatGPT dominates paid AI use on Capitol Hill. Congressional offices are using the chatbot to draft memos, summarize legislation, and help handle constituent communications — the everyday paperwork that consumes much of a congressional staffer's day.
CNBC reports that at least 70 House offices used identifiable AI tools in early 2026, with Democratic offices leading the visible spending. CNBC also cautions that the real footprint is likely larger than the records suggest: much AI use is probably undercounted, since spending disclosures only capture tools that show up as identifiable line items.
That gap matters. Free accounts, tools bundled into other software, and personal use by staff wouldn't necessarily appear in House spending data at all. So the picture these records offer is a floor, not a ceiling.
The timing is what gives the story its edge. As CNBC frames it, this early spending is happening while lawmakers are weighing how to regulate AI. The same members debating rules for the industry are, in practice, becoming users of one company's product — and building institutional habits around it before any framework exists.
None of the reporting alleges wrongdoing. Offices buying commercial software is routine, and drafting help is a plausible fit for the work. But dominance by a single vendor creates a familiar dynamic: the tool that becomes default inside an institution tends to shape what that institution considers normal.
Why it matters: when the people writing AI rules rely daily on one company's AI, the line between regulator and customer gets harder to see.