Washington, July 26: U.S. President Donald Trump has signed a sweeping executive order barring the use of so-called “woke” artificial intelligence in federal government procurement, putting major technology firms on alert as they navigate an intensifying culture war.
The directive, one of three AI-related orders signed on Wednesday, is part of Trump’s broader plan to counter China’s AI influence and embed “American values” in the development of artificial intelligence technologies. But it marks the first time Washington has explicitly moved to shape the ideological behavior of AI systems.
The order requires companies selling AI tools to the government to demonstrate ideological neutrality, effectively discouraging the inclusion of diversity, equity and inclusion (DEI) principles. It calls out concepts such as critical race theory, transgenderism, unconscious bias, intersectionality, and systemic racism as examples of what it terms “destructive ideologies.”
“The federal government will not be buying WokeAI,” said conservative tech investor and Trump adviser David Sacks, who helped craft the order along with DEI critic Chris Rufo.
While the tech industry has welcomed aspects of Trump’s pro-business AI agenda, the anti-woke provision poses a new regulatory challenge. Industry giants including Google, Microsoft, Meta, Anthropic, and Palantir declined to comment on the order. OpenAI said it is awaiting further guidance but believes its work to ensure ChatGPT’s objectivity aligns with the directive.
Trump’s order does not impose direct content restrictions or output filtering. Instead, it uses federal contracts as leverage, asking companies to disclose internal policies used to guide chatbot behavior. Critics say this will push firms toward self-censorship to retain government business.
“The Trump administration is taking a softer but still coercive route,” said Jim Secreto, a former Commerce Department official under President Biden. “It creates strong pressure for companies to conform.”
The move has drawn comparisons to China’s regulatory approach, though the enforcement mechanisms differ. While Beijing mandates pre-deployment AI audits and censorship of banned content, Trump’s order stops short of such measures.
“There is nothing in this order that says companies must or must not produce certain types of outputs,” said Neil Chilson, former chief technologist at the Federal Trade Commission. “It simply prohibits intentional encoding of partisan or ideological judgments.”
Civil rights groups, however, say the order undermines efforts to eliminate harmful bias in AI systems. “There’s no such thing as ‘woke AI’,” said Alejandra Montoya-Boyer of the Leadership Conference on Civil and Human Rights. “There’s AI that discriminates, and AI that actually works for all people.”
AI models, including those built by Google’s Gemini, Microsoft’s Copilot, and OpenAI’s GPT, are trained on vast internet datasets that reflect society’s many biases. Designing them to be both neutral and equitable is a persistent challenge.
The executive order follows conservative backlash over a February 2024 incident involving Google’s image-generation tool, which displayed racially diverse depictions of U.S. Founding Fathers, leading critics to accuse the company of ideological manipulation.
The White House has not confirmed if certain companies, including Elon Musk’s xAI, will be favored under the new rules. Musk’s chatbot Grok recently drew scrutiny after generating antisemitic content, days before the firm won a $200 million U.S. defense contract.
As the study period preceding the full implementation of the order unfolds, tech firms will be forced to navigate a politically charged regulatory environment while continuing to expand the role of AI in government and society.
