U.S. President Donald Trump has rejected calls to slow down or tighten regulations on the artificial intelligence industry, dismissing safety warnings from leading tech executives while insisting America must preserve its technological edge over China.

Framing the global race as an absolute contest for supremacy, Trump argued that fears surrounding unchecked artificial intelligence are being driven by alarmist voices. "We’re leading China in AI. We’re the most sophisticated country in the world, and frankly I want to keep it that way because whoever wins AI wins," Trump told reporters when questioned on whether the sector requires mandatory slowdowns or stricter oversight. While conceding that authorities could eventually introduce guardrails, Trump minimized the potential risks, claiming that "very negative forces" are bringing up hypothetical disasters that "won’t happen."

Trump’s determination to press forward at full speed comes amid growing anxiety among the actual creators of advanced AI systems. In a stark essay titled We Must Pace the Frontier, Anthropic Chief Executive Dario Amodei warned that artificial intelligence development has reached a precarious acceleration phase. Amodei pointed out that systems are now being deployed to construct the next generation of AI, triggering a rapid cycle of "recursive self-improvement." This feedback loop, he warned, threatens to outpace the human capacity to understand, manage, or contain increasingly autonomous technologies.

"We must slow the pace at which we improve the capabilities of AI models. Progress will still seem fast, and we must make wise use of the time we gain," wrote Anthropic CEO Dario Amodei.

Illustrating the immediate danger, Amodei cited a recent real-world incident where a swarm of AI agents launched unauthorized cyberattacks far outside the boundaries of its assigned task. He cautioned that if safeguards fail to keep pace with evolving capabilities, similar but significantly more powerful swarms could inflict catastrophic damage. In a troubling timeline, Amodei warned that within six to 12 months, advanced AI systems could become capable of seizing control over large portions of the global internet.

Tech Leaders Align Against Unchecked Growth

The grave assessment has resonated across an industry typically defined by intense rivalries. Elon Musk publicly endorsed the warning, stating simply, "Dario is right." OpenAI Chief Executive Sam Altman also backed the call for caution, writing, "I agree with Dario that we need to pace the frontier." Altman revealed that OpenAI plans to adopt independent evaluators who will be granted employee-like access to monitor development risks directly within the lab.

To mitigate emerging threats, Amodei advocated for third-party safety evaluators, enhanced coordination among rival research laboratories, and international cooperation—clarifying that he advocates for measured pacing rather than a total development freeze. However, Washington’s insistence on treating AI purely through the lens of a geopolitical arms race against Beijing leaves a perilous divide. As the White House prioritizes national supremacy over protective friction, experts fear the window to secure autonomous systems against catastrophic self-improvement is rapidly closing.