OpenAI wants the government to review its AI models — in exchange for a break from state AI regulations.
On Thursday, the company released a 15-page policy advisory in response to the Trump administration’s request for input, which will inform the administration’s forthcoming AI Action Plan. OpenAI offered to voluntarily let the federal government review its models in exchange for being exempted from state-specific regulations. The company positioned its proposal as a way to counter China’s AI advances because of how it would allow American companies to speed ahead in AI.
Also: Generative AI is finally finding its sweet spot, says Databricks chief AI scientist
“We propose a holistic approach that enables voluntary partnership between the federal government and the private sector,” the company said.
The Trump administration’s forthcoming policy will replace former President Biden’s AI executive order and related efforts, which Trump rescinded on his first day in office. The administration has carried out related firings and funding cuts in recent weeks. While AI policy remains unclear at the federal level, individual states have been exploring their own legislation, which OpenAI’s advisory called “overly burdensome.” Much of this legislation deals with data privacy.
Interestingly, Chris Lehane, OpenAI’s vice president of global affairs, told Bloomberg that he thinks the US AI Safety Institute — created under Biden’s executive order — could be the liaison between the government and private sector. This proposal, if formalized, would change the current course of the Institute, which has been a rumored target for layoffs and funding cuts under Trump in recent weeks. The Biden-appointed head of the Institute, Elizabeth Kelly, stepped down shortly after Trump took office.
Lehane’s comments echo voluntary agreements the Biden administration had previously brokered with AI companies and indicate renewed interest from the private sector in being regulated by the Institute (or an equivalent at the federal level). It’s unclear how that would fit into the Trump administration’s efforts to deregulate AI across the board.
OpenAI also proposed “digitizing government data currently in analog form” in order to make it “machine-readable.” The company said this “could help American AI developers of all sizes, especially those working in fields where vital data is often government-held.”
Also: This new AI benchmark measures how much models lie
“In exchange, developers using this data could work with governments to unlock new insights that help develop better public policies. For example, government agencies can build on the work of the US National Archives and Records Administration in using Optical Character Recognition for text searchability and AI-driven metadata tagging,” the proposal continued.
The proposal also called for changes to US copyright law to “avoid forfeiting our AI lead to the PRC [People’s Republic of China] by preserving American AI models’ ability to learn from copyrighted material.” OpenAI has been sued in many instances over copyright infringement, most notably by publishers like The New York Times, authors, and artists.
On the same day OpenAI made its request, Google published a similar set of requests for lessening copyright law.
Also: Anthropic quietly scrubs Biden-era responsible AI commitment from its website
“For too long, AI policymaking has paid disproportionate attention to the risks, often ignoring the costs that misguided regulation can have on innovation, national competitiveness, and scientific leadership — a dynamic that is beginning to shift under the new Administration,” Google stated. This sentiment is consistent with several tonal and literal changes major AI companies have made away from safety concerns and regulations in recent months.
Both policy proposals follow recent expanded partnerships between AI companies and the US government, including allowing the National Labs to test frontier models and Project Stargate.
Source link
#OpenAI #trade #govt #access #models #regulations
