Schumer proposes $32 billion annual spending under AI road map

Senate Majority Leader Charles E. Schumer and his artificial intelligence working group on Wednesday backed a proposal for the federal government to spend at least $32 billion annually on non-defense related AI systems. 

“No technology offers more promise to our modern world than artificial intelligence,” Schumer, D-N.Y., said in a statement. “But AI also presents a host of new policy challenges. Harnessing the potential of AI demands an all-hands-on-deck approach and that’s exactly what our bipartisan AI working group has been leading.”

That level of spending was first proposed by an independent commission, the National Security Commission on Artificial Intelligence, which was led by former Google CEO Eric Schmidt and issued its final report in 2021. 

Alongside the suggested spending, the Senate AI working group released a “road map” for the technology that outlines proposals for individual congressional committees to take up. Although some measures may be enacted in the current Congress, others could extend to years out, the group, which includes Martin Heinrich, D-N.M., Todd Young, R-Ind., and Mike Rounds, R-S.D., said in a statement.

The road map, called “Driving U.S. Innovation in Artificial Intelligence,” follows a series of forums that Schumer convened last year to brief senators on all aspects of the technology. In those forums, lawmakers heard from industry executives including Alphabet Inc. CEO Sundar Pichai, Tesla Inc. CEO Elon Musk and Open AI CEO Sam Altman, as well as the leaders of civic groups such as Maya Wiley, president of the Leadership Conference on Civil and Human Rights. 

The lawmakers said they support  “a strong comprehensive federal data privacy law to protect personal information,” adding that the “legislation should address issues related to data minimization, data security, consumer data rights, consent and disclosure, and data brokers.” 

The senators also recommended that the Senate Appropriations Committee work with other committees to draw up “emergency appropriations language” to fill the gap between current and proposed spending levels for research and development at the Energy and Commerce departments, as well as at the National Science Foundation, the National Institute of Standards and Technology, and NASA. 

At a news conference Wednesday, Schumer said he plans to meet with House Speaker Mike Johnson soon on the issue “to see how we can make this bipartisan effort … and I think he’s very interested in doing that.” He added that Congress would likely have to appropriate defense expenditures as large as the nondefense spending.

Young said he’s done additional outreach to the House through Rep. Jay Obernolte, R-Calif., the co-chair of the House Task Force on Artificial Intelligence.

“I think this will be highly well received by our colleagues in the House,” Young said.

The road map has already drawn some criticism.

“Schumer’s new AI framework reads like it was written by Sam Altman and Big Tech lobbyists,” said Evan Greer, director at Fight for the Future, a nonprofit digital rights advocacy group. “It’s heavy on flowery language about ‘innovation’ and pathetic when it comes to substantive issues around discrimination, civil rights, and preventing AI-exacerbated harms. The framework eagerly suggests pouring Americans’ tax dollars into AI research and development for military, defense, and private sector profiteering.”

While the goal of the road map is American innovation, Congress is also dealing with the dangers of AI. That includes a set of AI-related bills sponsored by Sen. Amy Klobuchar, D-Minn., chairwoman of the Senate Rules Committee, that the panel is marking up Wednesday.

One would prohibit the distribution of AI generated deepfake messages including audio and visual ads on candidates running for federal office. 

Another would require ad creators and campaigns to disclose whether any publicity material was produced with the aid of AI, and the third would require the Election Assistance Commission to develop voluntary guidelines on the use of AI in elections.