Trump reverses Biden’s executive order on AI safety

Key policy change: Donald Trump eliminated Executive Order 14110, which required AI developers to share safety test results with the federal government and established standards for AI safety testing.

• The rescinded order was part of Trump’s day-one actions that reversed 78 Biden-era executive orders
• Biden’s original order had tasked NIST with developing safety testing standards
• The order had directed federal agencies to assess AI-related risks across various sectors, including cybersecurity and critical infrastructure

Worker and consumer protections eliminated: The rescinded order contained several measures designed to protect American workers and consumers from AI-related risks.

• The original order had commissioned research on AI’s impact on the labor market
• It established guidelines for addressing AI-enabled fraud
• The order included provisions to combat discriminatory algorithms

Tech industry alignment: Trump’s inauguration featured prominent attendance from major technology industry leaders.

• Notable attendees included Jeff Bezos, Elon Musk, Mark Zuckerberg, Tim Cook, Shou Zi Chew, Sundar Pichai, and Sam Altman
• Some tech leaders contributed to Trump’s inauguration budget

Surviving AI initiatives: Some of Biden’s other AI-related policies may remain in effect despite the executive order reversal.

• A new regulatory framework limiting international sharing of AI chips and models remains in place
• An executive order facilitating AI data center development on federal land continues

International implications: The policy shift creates potential tension with international AI regulation efforts.

• The change may lead to conflicts with the European Union’s AI Act, which mandates transparency and restricts certain AI applications

Policy trajectory analysis: The removal of federal AI safety guidelines, combined with strong tech industry support for the new administration, suggests a shift toward less regulated AI development in the United States. This approach could accelerate AI innovation but may also create challenges for international cooperation and raise concerns about consumer protection and algorithmic bias.