OpenAI, Anthropic, and Nvidia post record lobbying quarters as AI policy stakes rise

Nvidia, Anthropic and OpenAI each recorded their biggest lobbying quarter to date between July and September, according to federal disclosure filings reported by The Verge. Nvidia spent $2.2 million, Anthropic $2.1 million and OpenAI $1.5 million — record totals for all three, and the clearest evidence yet that AI companies are building permanent policy operations in Washington.

How the numbers compare

For companies with long-established lobbying shops, those figures would hardly register. According to The Verge, Microsoft has spent at least $2 million every quarter since 2021, and Google regularly tops $3 million, including in this one. What stands out is the trajectory. OpenAI had never spent $1 million in a single quarter before this year, and only crossed the half-million mark at the end of 2024. Anthropic's highest quarter last year was just over $1 million. And this is the first quarter in which Nvidia's lobbying spending has exceeded $2 million.

The timing adds context. The quarter covered the run-up to the midterm elections, and it closed shortly after many tech CEOs attended a late-September White House lunch with President Donald Trump that produced an AI safety accord described as "morally binding." According to The Verge, the filings show those public appearances were underpinned by frequent private discussions, arriving just before Trump named his AI czar and task force. With Democrats expected to win back at least partial control of Congress after the midterms, AI companies may see growing reason to engage lawmakers who view cracking down on the industry as a politically rewarding cause.

What the filings show

Anthropic's disclosure is the most detailed of the three. The Verge reports that it lists work on AI-related provisions in the Kids Online Safety Act, export control language in the AI Overwatch Act, Trump's executive order renaming AI "Super Intelligence," and the Ratepayer Protection Act, a bill that would shift energy costs for data centers onto the developers who build them.

OpenAI's filing names only broad topic areas, but the pattern suggests a response to the fallout from the Hugging Face attack. Alongside copyright, the company lobbied Congress, the White House, the Small Business Administration and the Commerce and State Departments on cybersecurity evaluations for frontier models, AI safety oversight frameworks and antitrust protections for information sharing.

Nvidia spread its efforts across Congress, multiple federal agencies and White House offices, weighing in on 6G networks, "responsible AI design," competition policy, quantum computing and export controls.

Apple also set a record in the quarter. According to The Verge, it spent $3.1 million — its first quarter above $3 million — engaging on intellectual property, tariffs, kids online safety, competition and domestic semiconductor manufacturing, including bills that would make app stores responsible for verifying users' ages.

Why it matters

The filings mark a shift in kind, not just degree. Until recently, the leading AI labs treated Washington as an afterthought; they are now institutional players in policymaking, with dedicated budgets and issue portfolios spanning safety, energy, exports and competition. With data center development and AI safety now live concerns for voters, The Verge expects this activity to keep climbing regardless of which party controls Congress next year — meaning the rules governing AI will be written in close consultation with the companies the rules are meant to govern.