How Trump Was Persuaded to Regulate A.I.
TL;DR
President Trump signed a watered-down executive order requiring voluntary disclosure of AI models 30 days before release, marking a stark reversal from his administration's initial deregulatory stance after cybersecurity warnings from Anthropic, Microsoft, and JPMorgan triggered an internal White House battle between libertarian advisors and national security officials.
🏛️ The Regulatory Shift 2 insights
From 'beautiful baby' to acknowledged risk
The administration initially scrapped Biden-era AI safety rules in January, with Trump calling AI a 'beautiful baby' that shouldn't face 'foolish rules,' but reversed course after Anthropic's Mythos model demonstrated AI could autonomously identify software vulnerabilities for weaponized cyber attacks.
Corporate warnings drive policy change
Microsoft and JPMorgan CEO Jamie Dimon directly warned the White House that advanced AI posed immediate threats to critical infrastructure and financial systems, creating political liability fears for Treasury Secretary Scott Bessent and Chief of Staff Susie Wiles.
⚔️ Inside the White House Battle 2 insights
FDA-style proposal sparks industry revolt
The initial draft proposed a mandatory 90-day government review akin to FDA drug approvals, triggering intense lobbying from AI czar David Sacks and tech executives including Mark Zuckerberg who feared it would slam the brakes on innovation in the race against China.
Trump's last-minute reversal
Hours before a scheduled May 21st Oval Office signing ceremony, Trump canceled the order after calling tech executives including David Sacks, who told him directly 'don't sign this,' only to revive a compromise version weeks later after persistent pressure from Bessent and Wiles.
📜 The Compromise and Criticism 2 insights
Voluntary 30-day disclosure requirement
The final June 3rd executive order requires companies to voluntarily share AI models 30 days before public release—down from the proposed 90 days—and explicitly prohibits any mandatory licensing or pre-clearance process to appease industry concerns.
Populists demand stronger action
Populists like Steve Bannon and religious leaders criticize the measure as insufficient, arguing AI threatens working-class jobs and America's 'moral fabric,' and support bipartisan legislation by Senator Josh Hawley for more stringent federal oversight.
Bottom Line
The executive order represents a symbolic but largely toothless first step toward AI oversight, indicating that meaningful regulation will require Congressional legislation rather than voluntary industry cooperation or executive compromise.
More from New York Times Podcasts
View all
Why the Ebola Outbreak Has Been Nearly Impossible to Stop
A rare Ebola strain in the Democratic Republic of Congo has killed at least 250 people with over 1,100 suspected cases, spreading undetected for 2-3 months due to US aid cuts, militia-controlled terrain, and a lack of vaccines or testing kits, while healthcare workers face violent community backlash fueled by conspiracy theories and traditional burial customs.
How Elon Musk Engineered the World’s Biggest I.P.O.
SpaceX is preparing the largest IPO in history, aiming to raise up to $75 billion at a $1.25 trillion valuation that could make Elon Musk the world's first trillionaire, while changing stock market rules to allow unprecedented retail access and rapid index inclusion despite the company currently operating at a $4.3 billion loss.
The Beginner’s Guide To Retinol Products
Dermatologists consider retinoids the gold standard for anti-aging and acne because they stimulate collagen and accelerate cell turnover, but they require careful introduction with small amounts and gradual frequency increases to avoid irritation.
Inside Trump’s Mad Dash to Renovate Washington
President Trump is fast-tracking high-profile renovations across Washington D.C. ahead of the nation's 250th anniversary, utilizing secret no-bid contracts that bypass federal procurement laws designed to protect taxpayer dollars and ensure competitive pricing.