The new U.S. Senate artificial intelligence roadmap is full of ambition but lacks much practical substance. basically, Promoting American artificial intelligence innovation The report is a wish list that its authors hope congressional committees will consider as they address the possible impacts of the emerging field of artificial intelligence.
However, the senators behind it, including Majority Leader Chuck Schumer (D-N.Y.), are convinced that following the roadmap would require Congress to allocate “at least $32 billion per year for (non-defense) purposes as recommended by the NSA.” Artificial Intelligence Innovation” Committee on Artificial Intelligence (NSCAI). As NSCAI leadership explained, “Now is not the time for abstract criticism of industrial policy or concerns about deficit spending to impede progress.” Although Goldman Sachs predicts that private sector spending on artificial intelligence could reach $200 billion by 2025, But the two parties clearly have reached agreement on this point.
The good news is that the roadmap is not an EU innovation-killing AI bill.Unlike the AI Act, the roadmap does not impose any actual new regulations or restrictions on rapidly growing AI applications in the United States
All it does is check off the boxes of issues the task force believes the Senate committee should consider. These include legislation aimed at federal funding for artificial intelligence research and development, voting security and election fraud, worker displacement and training, copyright issues, online child sexual abuse material, bans on the use of artificial intelligence for social scoring, data privacy, international cooperation Legislate cooperation with allies, export controls on critical technologies, defense against chemical, biological, radiological and nuclear threats, and limits on the use of artificial intelligence in warfare.
Unlike the EU’s AI bill, the roadmap does not call for regulations requiring new AI products and services to undergo safety assessments by government officials before being made available to the public.Instead, the roadmap more sensibly suggests developing a “The framework clearly defines the situations that require pre-deployment assessment of AI models. Legislate to prevent the unauthorized use of an individual’s name, image, likeness, and voice by artificial intelligence products and services and to identify novel synthetic content generated by artificial intelligence.
Critics say the roadmap is not enough and not fast enough. “A long list of proposals is not a substitute for enforceable laws,” Amba Kak and Sarah Myers West, co-executive directors of the AI Now Institute, said in a statement. “What we need are stronger rules that hold companies accountable when publishing Proving products before they are safe ensures fair competition in the AI market and protects the most vulnerable in society from the most harmful applications of AI – and that’s just the beginning.
In a press release, Fighting for the Future director Evan Greer declared, “Schumer’s new artificial intelligence framework reads like it was written by Sam Altman and the big tech lobby. It’s full of words about ‘innovation.’ Gorgeous language that seems pathetic when it comes to substantive issues.
Proposed federal deficit spending on AI R&D aside, those of us who support permissionless innovation should be slightly relieved that the roadmap does not, for the most part, call for sweeping federal regulations in the early stages of AI development. . Of course, there’s no guarantee that Congress won’t screw up AI innovation in the future.
