Donald Trump's reversal of Biden's AI policy has changed the regulatory landscape for artificial intelligence in the United States. By striking down Executive Order 14110, Trump removed the first mandatory safety standards for US companies building AI systems and opened a wider gap between the US, the European Union and China.
What Trump removed
Executive Order 14110 was introduced in October 2023 under Biden. Its stated aim was to support the "Safe, Secure, and Trustworthy Development and Use of Artificial Intelligence." The order created safety obligations for US AI companies, including a requirement to share AI safety test results with the government.
The order also addressed AI-generated deception. One part of that effort involved developing standards for digital watermarks, a tool meant to help identify AI-generated content.
Trump struck down Executive Order 14110 in one of his first moves as president. It was among 78 Biden policies Trump reversed upon taking office. The order no longer appears on the White House website, although it remains available through archive sites.
The practical effect is a regulatory gap. The US is now moving away from mandatory national safety requirements for AI development, leaving companies with fewer federal constraints as they build and release new systems.
The global AI race now looks more uneven
The decision places the US on a different path from the European Union. With Biden's order gone, the European Union's AI Act stands as the main regulatory framework for artificial intelligence development worldwide.
China is following a separate model. The source describes China as maintaining tight government control over its AI sector while focusing primarily on rapid development rather than safety standards.
That leaves three sharply different approaches:
- United States: AI development is now largely unregulated at the national level, with a patchwork of conflicting state laws.
- China: AI development is state-led and focused on speed.
- European Union: AI development is governed by comparatively strict safety rules under the EU AI Act.
This split could accelerate competition. US companies may be able to move faster with fewer national safety requirements. Chinese companies remain inside a state-directed system. The EU, meanwhile, is trying to enforce stricter rules while also facing slower development and deployment of large AI models.
Europe may become a tougher market for AI products
The EU's position could now become more complicated. Its rules may carry more weight because they are the main broad regulatory framework left standing, but they may also become a bigger hurdle for foreign companies that want to offer AI products in Europe.
For companies outside the EU, compliance creates a separate burden. Rules that apply inside the EU may not apply in the same way elsewhere, which can make Europe a distinct market with its own product decisions and launch timelines.
Some tech giants have already adjusted their plans. Meta and OpenAI have held back certain AI products and features in the EU, citing regulatory uncertainty.
At the same time, the EU AI Act has not been rejected by the industry as a whole. More than 100 tech companies recently pledged to comply with it. That support creates the possibility that EU rules could influence broader safety standards, although the current political climate makes that outcome unlikely.
Why Elon Musk and xAI matter in this shift
Trump's exact reasoning remains unclear. The source suggests the decision is likely tied to accelerating AI development, particularly as the US competes with China. It could also benefit Elon Musk, a close ally of Trump.
Musk is building large language models through xAI to compete with OpenAI. Without Biden's safety requirements, that competition could move faster and face fewer restrictions.
Musk's public posture on AI is unusual. He frequently warns about AI's dangers and aligns himself with AI doomsayers, yet he also strongly opposes government regulation. The source frames that tension as easier to understand if his warnings are seen as focused on things outside his control.
Other US companies, including OpenAI, could also benefit from a lighter regulatory environment. But the competitive pressure may not affect every company in the same way. Firms that have made safety central to their identity may face pressure to move more quickly if rivals with fewer safety commitments can ship faster.
What changes for AI companies
The end of Executive Order 14110 does not mean AI regulation disappears everywhere. It means the US no longer has those mandatory national safety requirements in place, while the EU continues with its own strict framework and China continues under state control.
For AI companies, the result is a more fragmented operating environment. A model, product or feature may face different expectations depending on where it is developed, where it is released and which rules apply in that market.
The biggest immediate change is strategic. US AI firms now have more room to move quickly at the national level. European rules may continue to shape what companies can offer inside the EU. And the competitive pressure between OpenAI, xAI and other developers could intensify as safety requirements become less uniform across major markets.