Trump Readies AI Order Amid Industry Boom and Regulatory Pushback

A burst of announcements this week underscored how quickly the artificial intelligence sector is advancing even as debates over government oversight intensify. OpenAI revealed that one of its reasoning models had solved a longstanding geometry problem that had eluded mathematicians for decades, a step that points to potential gains in scientific discovery. At the same time, Anthropic reported revenue on track to more than double to 10.9 billion dollars in the current quarter, putting the company on pace for its first profitable period well ahead of earlier forecasts.

These milestones arrived alongside major infrastructure commitments. Anthropic agreed to spend roughly 1.25 billion dollars per month through 2029 on access to SpaceX computing resources, a deal that positions Elon Musk as an increasingly central player in supplying the hardware backbone for advanced models. Nvidia, meanwhile, posted 81.6 billion dollars in quarterly revenue, with its data-center segment accounting for the vast majority, as chief executive Jensen Huang described demand as having gone parabolic.

The pace of commercial progress has sharpened questions about how, or whether, federal policy should shape the field. Anthropic has endorsed measures that would impose new liability rules, disclosure requirements, and restrictions on exporting advanced chips. Critics argue these proposals would raise costs for smaller entrants and effectively lock in advantages for firms already operating at scale. White House AI adviser David Sacks has publicly warned that such efforts amount to regulatory capture dressed up as safety advocacy.

The pattern is familiar from earlier technology waves. Incumbent companies often shift from resisting rules to supporting them once they can turn compliance into a competitive moat. In AI the stakes are higher because the technology touches critical infrastructure, national security, and scientific research. Heavy-handed restrictions risk slowing the very experimentation that produced recent breakthroughs, yet complete hands-off approaches leave open questions about liability, export controls, and concentration of compute resources.

Some analysts have called for moving past the binary framing that pits those most alarmed by AI risks against those focused solely on speed. A more useful lens would examine concrete policy tools that preserve competitive entry while addressing genuine externalities. Targeted public investment in shared research infrastructure, clearer standards for model evaluation, and export policies calibrated to security concerns rather than broad industrial protection could all fit that description. The recent Nvidia results and the OpenAI geometry advance show that technical momentum remains strong. Whether policy choices reinforce or blunt that momentum will depend on resisting the temptation to let any single company write the rules.