Among his first actions as president, Donald Trump has rescinded an executive order signed during the Biden administration that directed the U.S. government to develop safeguards for the development of artificial intelligence programs, including generative AI models and their use in the healthcare industry.
The October 2023 order also tasked the Department of Health and Human Services (HHS) with recording reports of unsafe healthcare practices involving AI and pushed the National Institute of Standards and Technology (NIST) to build a U.S. AI Safety Institute as a home for technical guidance in future regulatory efforts.
The NIST institute was also instructed to set standards for “red-team testing” to ensure the safety of AI models before public release and protecting against bias—among other guidelines for employing AI in biotech research and development, such as in the computer-generated design of potential new drug compounds. At the moment, it is unclear whether the organization will continue its work.
Last August, the AI Safety Institute announced framework agreements with Anthropic and OpenAI to establish formal collaborations on safety research, as well as in model testing and evaluation. The institute also held technical dialogues last year with the EU’s AI Office on best practices and risks.
Trump’s other day one actions included revoking Biden-era executive orders related to COVID-19 testing, treatments and pandemic response; new drug pricing models at the Centers for Medicare & Medicaid Services; and policies aimed at expanding health insurance enrollment under the Affordable Care Act.
The new administration also announced plans to withdraw U.S. membership from the World Health Organization and instituted a governmentwide freeze on new hiring. Meanwhile, according to a report from CBS News, Trump is set to announce up to $500 billion in investments into private sector AI infrastructure as part of an OpenAI, Softbank and Oracle joint venture dubbed Stargate.
Earlier this month, HHS published its strategic plan for AI policy—with short- and long-term goals such as establishing a nationwide network of quality assurance labs through the Coalition for Health AI. It also planned to build on challenge initiatives, such as the CMS AI Health Outcomes Challenge and the NIH CRDC AI Data Readiness Challenge, to promote AI innovation and adoption.