HHS Permits Employee Use of ChatGPT, Wiht Guardrails
WASHINGTON – The Department of Health and Human Services (HHS) announced today, September 25, 2025, it will allow employees to utilize ChatGPT and other large language model (LLM) artificial intelligence tools, but with strict guidelines to protect sensitive data and ensure responsible implementation. The policy, detailed in America’s AI Action Plan released July 23, 2025, marks a notable shift in federal agency approaches to generative AI, balancing innovation with security concerns.
This decision arrives as federal agencies grapple with the potential benefits and risks of rapidly evolving AI technologies. HHS’s move impacts over 75,000 employees across its operating divisions,including the Centers for Disease Control and Prevention (CDC) and the food and Drug Management (FDA). The policy aims to boost efficiency and accelerate research, but prioritizes safeguarding patient data, intellectual property, and maintaining public trust. Future implementation will likely serve as a model for other federal departments considering similar integrations.
According to the AI Action Plan, HHS will implement a tiered access system. Employees will be required to complete training modules on AI ethics, data privacy, and cybersecurity before gaining access to LLMs. The policy explicitly prohibits entering Protected Health Information (PHI) or Personally Identifiable Information (PII) into public-facing AI tools.
“We recognize the transformative potential of AI to improve public health and human services,” stated an HHS spokesperson. “However, we must proceed cautiously and responsibly, ensuring that these tools are used in a manner that protects the privacy and security of the individuals we serve.”
The guidelines also emphasize the importance of human oversight. AI-generated content will be considered a draft and must be reviewed and validated by a qualified HHS employee before dissemination. The department will continuously monitor the use of AI tools and update the policy as needed to address emerging risks and best practices.
America’s AI Action Plan.pdf (whitehouse.gov)
