Former President Joe Biden’s 2023 executive order on AI has been revoked by President Donald Trump on his first day in office.
The “Executive Order on the Safe, Secure, and Trustworthy Development and Use of Artificial Intelligence” was issued on Oct. 30, 2023, and set forth a series of initiatives for federal agencies to strengthen AI safety and security, avoid harms such as misuse, bias and privacy violations, and strengthen the America’s talent pool of AI professionals.
The order also created requirements for developers of “dual-use foundation models,” which included the most powerful models containing at tens of billions of parameters, to report their plans, cybersecurity measures and red-team testing results to the federal government.
Many of the deadlines set by the executive order for federal agencies to establish policies, develop frameworks and issue reports related to its AI initiatives have already passed; for example, the National Institute of Standards and Technology (NIST) published a series of guidance and testing resources throughout 2023 and 2024, including documents on AI misuse risks and an open-source platform for testing machine learning (ML) models.
However, revoking the order halts requirements for continuous monitoring and reporting related to the EO’s goals, including those related to dual-use foundation models.
“Without a robust framework for governance, we risk the misuse of AI, from weaponized deepfakes to systemic biases that amplify inequalities,” Gabrielle Hempel, customer solutions engineer at Exabeam, said in an email to SC Media. “AI can be a tool for incredible progress, but only if we set clear guardrails that prioritize ethical development, security, and accountability.”
Trump announced shortly after his inauguration that he planned to revoke nearly 80 of Biden’s executive actions during his presidency, according to Reuters. The 2024 Republican Platform also put Biden’s AI executive order on the chopping block, stating: “We will repeal Joe Biden’s dangerous Executive Order that hinders AI Innovation, and imposes Radical Leftwing ideas on the development of this technology. In its place, Republicans support AI Development rooted in Free Speech and Human Flourishing.”
Advocacy groups representing tech giants like Google, Meta and Amazon, criticized the EO when it was first issued, arguing it could stifle innovation and place an undue burden on AI developers, according to FedScoop. Cybersecurity experts, including executives from Bugcrowd and Darktrace Federal, praised the EO’s security focus in previous statements to SC Media.
Abnormal Security Chief Information Officer Mike Britton said in a 2023 column for SC Media that the EO would have benefits for consumer protections, AI talent development and driving demand for AI-native products, but noted it fell short in addressing both adversarial AI and the potential benefits of AI to strengthen cybersecurity defenses.
With a lack of U.S. federal laws broadly addressing AI safety, security and regulation, some states have begun passing their own AI laws, including California, which saw the veto of one major AI safety law but successfully passed more than 17 others over the past year.
“Whether in cybersecurity, legal compliance, product development, or technology strategy, we all have a role in advocating for responsible AI practices. We should be pushing for policies that balance innovation with safety and protect the integrity of our systems and societies,” said Hempel.
