MAGA coalition urges Trump to mandate safety testing for all frontier AI models

A fundamental shift is taking place within the American conservative movement regarding the future of artificial intelligence, as a coalition of high-profile MAGA-aligned organizations and leaders has formally petitioned the White House to establish rigorous federal oversight of the industry. Led by the advocacy group Humans First, the coalition sent a comprehensive open letter to President Donald Trump urging the immediate issuance of an executive order that would mandate safety testing for all "frontier" AI models before they are released to the public.[1] This move marks a significant departure from the traditional Republican platform of deregulation and signals a growing belief among the president’s most loyal supporters that the risks posed by unchecked AI development now constitute an existential threat to national sovereignty and public safety.

The coalition includes a diverse array of influential figures from the populist right, ranging from former White House Chief Strategist Stephen K. Bannon and Tea Party activist Amy Kremer to Brendan Steinhauser of the Alliance for Secure AI and Ryan Girdusky of the 1776 Project PAC.[1] For years, the prevailing conservative sentiment toward the technology sector was characterized by a desire to dismantle the "administrative state" and allow market forces to drive innovation, particularly to maintain a competitive edge over China. However, the Humans First letter argues that the current trajectory of AI development is so rapid and its potential for harm so great that leaving oversight entirely in the hands of private corporations is no longer a viable strategy. The signatories specifically highlight concerns that advanced models could be co-opted for sophisticated cyberattacks, the creation of biological weapons, and the destabilization of critical financial and electrical infrastructure.

This push for regulation is framed not as an endorsement of "big government," but as a necessary component of an "America First" national security policy. The coalition’s rhetoric leans heavily on the analogy of nuclear technology, suggesting that just as the federal government oversees the enrichment of uranium and the operation of power plants, it must also exercise control over the "computational enrichment" required to build the world’s most powerful digital brains. By shifting the focus away from the "woke" algorithmic bias and "social engineering" concerns that dominated conservative AI critiques during the previous administration, the Humans First coalition is attempting to build a new policy framework centered on hard security. They argue that if a model is capable of autonomously navigating complex networks or assisting in the design of pathogens, it cannot be treated as a mere consumer product protected by First Amendment rights or trade secret laws.

The timing of this pressure campaign is closely linked to the recent release and subsequent controversy surrounding Anthropic’s Mythos model. While the tech industry initially celebrated Mythos as a breakthrough in reasoning and problem-solving, internal reports and leaked assessments from the intelligence community have suggested the model possesses capabilities that could significantly lower the barrier for non-state actors to execute high-level cyber operations. The Trump administration, which spent the early months of its second term revoking the previous administration's AI executive orders and dismantling the reporting requirements for large-scale compute clusters, now finds itself at a crossroads.[2] While the White House has publicly favored a "light-touch" approach to ensure American dominance in the global AI race, officials are reportedly reconsidering the necessity of a formal review system.

Industry giants like OpenAI, Google DeepMind, and xAI are watching these developments with high stakes, as the coalition’s demands could lead to a permanent federal vetting process. Under the proposed Humans First framework, the Center for AI Standards and Innovation would be empowered to conduct "pre-deployment audits" to ensure that frontier models do not possess "catastrophic capabilities." The letter even suggests that companies failing these security reviews should be barred from lucrative government contracts and potentially face restrictions on their ability to export their technology. This has created a visible rift in the broader conservative-libertarian alliance; while populist leaders like Bannon are calling for "sovereign control" over AI, other factions—including many in the "Silicon Valley accelerationist" camp who supported the president’s return to power—view mandatory testing as a bureaucratic bottleneck that will only serve to help China close the gap.

The implications for the AI industry are profound, as a shift in the MAGA platform could lead to a rare moment of bipartisan alignment on the need for guardrails. While the motivations differ—with the left often focusing on labor rights, privacy, and equity, and the right focusing on national security, infrastructure, and "human-first" values—the outcome could be a much more restricted environment for the development of foundation models. If President Trump chooses to heed the advice of his allies, the era of "move fast and break things" in AI development may effectively come to an end, replaced by a regime of mandatory red-teaming and government certification. This would represent one of the most significant expansions of federal authority over a specific technology in modern history, reflecting a newfound consensus that the silicon frontier is too dangerous to remain unmapped by the state.

Ultimately, the Humans First movement reflects a growing realization that artificial intelligence is not just a tool for economic growth, but a transformative force capable of reshaping the very nature of human agency and national power. By demanding that the government "put humans first" in the face of autonomous systems, these groups are challenging the tech industry to prove that its products are safe before they are integrated into the fabric of American life. Whether the White House will choose to maintain its course toward total deregulation or adopt the "security-first" posture demanded by its base remains the most pressing question for the future of the American technology sector. The answer will likely determine not only the speed of AI progress but the degree to which the United States government remains the ultimate arbiter of what constitutes a safe and secure digital future.