Democrat Introduces Bill to Tax AI Companies for Jobs

A Texas congressman just formally introduced legislation betting that taxing AI companies directly is the right response to potential mass job displacement, modeling the approach on one of the most ambitious government job programs in American history. Representative Greg Casar introduced the AI Tax and Work Protection Act, shared first with NBC News, which would tax AI companies while creating a new Work Protection Administration funded by that revenue, according to NBC News's reporting on the bill.

"FDR responded to Great Depression levels of unemployment with a national strategy and a bold plan," Casar said, referencing President Franklin D. Roosevelt's Works Progress Administration, which put millions of Americans to work on public infrastructure projects during the Great Depression.

The bill targets AI companies directly rather than individual consumers of AI products. Companies would pay whichever is higher of two tax structures: a tax on AI tokens, the standard unit AI companies already use to measure and bill for computing power usage, or a tax on the revenue generated from AI products sold, according to NBC News's reporting on the bill's mechanics. Casar explained the reasoning behind using tokens as a tax base in an earlier op-ed for The American Prospect: "AI companies already keep track of their usage; that's how they know what to charge their largest customers... To avoid companies gaming token counts, the tax should measure both the number of tokens and the underlying computing power used to train and use AI models."

Casar has argued this structure creates a self-correcting mechanism tied directly to the scale of potential job losses. "Taxing AI directly ties the solution directly to the problem," he wrote. "If AI use grows quickly, driving layoffs alongside it, the revenue from an AI tax would go up too. Unlike traditional corporate taxes, an AI tax like the one I am proposing works even if employers fire workers before AI companies show a profit."

Casar's bill lands alongside several competing Democratic proposals addressing AI's economic impact, none of which have converged into a single unified party position. Senator Bernie Sanders separately introduced legislation for a $7 trillion AI sovereign wealth fund built on a one-time 50% stock levy on major AI companies, a proposal we covered in detail in our reporting on majority public support for an AI sovereign wealth fund earlier this summer. Senator Elizabeth Warren has proposed taxing the energy consumption of data centers, while Representative Ro Khanna has floated a "Data Center Bill of Rights" giving localities authority to block projects, according to NOTUS's reporting on the genuine policy divide among Democrats.

Critics, including some in the AI industry, argue concerns about widespread AI-driven job losses remain overstated and that taxing AI usage risks slowing innovation and economic growth in a still-developing industry, a tension that mirrors the broader debate we've tracked around restricting Chinese open-weight AI models and its potential unintended consequences for American competitiveness.

This bill is worth watching closely for any AI company or business heavily dependent on AI infrastructure, given the token-based tax structure would directly affect the cost of running AI workloads at scale if it advances. Even if this specific bill doesn't pass, the growing number of competing Democratic AI taxation proposals signals genuine, sustained political pressure building around how AI's economic gains get distributed, a dynamic worth factoring into long-term AI cost planning regardless of near-term legislative odds.

Representative Greg Casar introduced the AI Tax and Work Protection Act, which would tax AI companies based on either token usage or product revenue, whichever is higher, to fund a new federal jobs program modeled on FDR's Works Progress Administration. The bill joins several competing Democratic proposals addressing AI's economic impact, including Senator Bernie Sanders' $7 trillion AI sovereign wealth fund and Senator Elizabeth Warren's data center energy tax, none of which have unified into a single party position. Critics argue the proposal could slow AI innovation and that concerns about mass AI-driven job losses remain unproven.