On July 23, 2025, the White House issued an executive order addressing the use of artificial intelligence (AI) in federal agencies. This order raises concerns about ideological biases, particularly regarding diversity, equity, and inclusion (DEI), which could affect the accuracy of AI outputs. It emphasizes a commitment to procure only AI models that adhere to principles of truth-seeking and ideological neutrality, ensuring that these tools deliver reliable and factual information.

The executive order requires that AI models, especially large language models (LLMs), must not include ideological agendas that could distort factual information. Agency heads are directed to adhere to guidelines that promote transparency and compliance with these Unbiased AI Principles, which mandate that AI models prioritize historical accuracy and scientific inquiry. Furthermore, the order stresses that federal contracts must align with these requirements, holding vendors accountable for the AI technology they provide.

In executing this order, the White House aims to improve the integrity of AI systems utilized in government operations while allowing agencies the flexibility to implement these principles. The guidance will take into account technical limitations and permit exceptions in national security circumstances. This initiative builds on prior efforts to encourage trustworthy AI, reaffirming the federal government’s duty to use reliable technology for public service.

Original: article