MAGA-aligned Groups Push for AI Safety Testing via Executive Order
A coalition of conservative organizations led by Humans First has called on President-elect Donald Trump to issue an executive order requiring mandatory safety testing for frontier AI models before public release, according to a report from The Decoder.
The open letter, dated March 25, 2026, urges Trump to establish federal oversight mechanisms for advanced artificial intelligence systems deemed capable of posing “existential risks.” The groups argue such testing should be conducted by an independent agency to ensure compliance with safety standards before deployment.
“Frontier AI models require rigorous evaluation to prevent unintended consequences,” the letter states, defining frontier models as systems with capabilities exceeding current benchmarks in areas like reasoning, planning, and natural language processing.
The proposal marks a shift for some conservative groups, which historically opposed tech regulation. Humans First, a self-described “pro-human” advocacy organization, has previously criticized both major political parties on AI policy.
If implemented, the policy could impact AI development timelines at companies like OpenAI, Anthropic, and Google DeepMind, which have ongoing discussions with the Biden administration about voluntary safety measures.
