Biden Admin Orders AI Safeguards for Agencies by Dec 1

The Biden administration took a major step forward in ensuring the safe and responsible use of artificial intelligence by federal agencies. On Thursday, the Office of Management and Budget released a memo providing new guidance on AI deployment.

Building on an executive order signed by President Biden last October, the policies require agencies to carefully assess the risks of their AI systems, such as bias or privacy concerns. Agencies have until December 1st to put testing safeguards in place, or they’ll need to stop using the AI unless it’s absolutely critical.

The guidance promotes transparency, directing agencies to share their AI models and data publicly when possible without compromising security. Agencies will report annually on their AI use and risk mitigation efforts.

Furthermore, each federal agency must now appoint a chief AI officer to oversee implementation. The administration plans to expand the AI-focused workforce by bringing on at least 100 new professionals this summer.

In developing the policies, input was gathered from a variety of stakeholders. Vice President Kamala Harris highlighted the administration’s aim for the guidance to be an international model, saying: “President Biden and I intend that these domestic policies will serve as a model for global action. We will continue to call on all nations to follow our lead and put the public interest first when it comes to government’s use of AI.”

The goal is to balance managing the risks of AI with encouraging agencies to innovate responsibly. AI has promising potential applications, from disaster response to disease tracking. But much remains unknown about the rapidly evolving technology. These new policies represent the latest White House effort to harness AI’s benefits while reining in the dangers.

For more information, check out the White House Fact Sheet on the new AI guidance.