Vice President Kamala Harris has unveiled a new policy from the White House Office of Management and Budget (OMB) that aims to mitigate risks and maximize the benefits of artificial intelligence (AI) across federal agencies. This policy is a crucial step in fulfilling President Biden's Executive Order on AI, which called for comprehensive measures to strengthen AI safety and security, protect privacy, advance equity, promote innovation, and bolster American leadership in the global AI landscape.

Federal agencies have successfully completed all of the 150-day actions assigned by the Executive Order, building on their previous accomplishments of fulfilling the 90-day actions. This comprehensive approach emphasizes the Biden-Harris Administration's commitment to responsible AI innovation.

The OMB's new policy outlines several key directions for federal agencies. Firstly, it emphasizes the importance of addressing the risks associated with AI usage. The policy mandates that agencies implement concrete safeguards by December 1, 2024, when using AI in ways that could impact the rights or safety of Americans. These safeguards entail reliable assessments, testing, and monitoring of AI impacts on the public. The policy also aims to prevent algorithmic discrimination and ensure transparency in government AI usage across various sectors such as healthcare, education, employment, and housing.

To protect the federal workforce during the adoption of AI, OMB encourages agencies to consult federal employee unions and adhere to the Department of Labor's forthcoming principles on mitigating AI-related harms to employees. This approach involves collaboration with federal employees and labor unions to ensure responsible governance and use of AI.

The OMB's policy also provides guidance for managing AI risks specific to procurement. It acknowledges the challenges associated with federal procurement of AI, emphasizing the need for fair competition, data protection, and transparency. In the coming months, OMB will take action to align agencies' AI contracts with the policy and safeguard the rights and safety of the public.

Transparency of AI usage is another crucial aspect outlined in the policy. Federal agencies are now required to enhance public transparency by releasing expanded annual inventories of their AI use cases, reporting metrics about sensitive use cases withheld from the public inventory, notifying the public of any AI exempted from policy elements, and releasing government-owned AI code, models, and data where it does not pose risks.

The OMB's policy also strives to remove unnecessary barriers to responsible AI innovation within federal agencies. AI technology promises numerous opportunities to address pressing societal challenges, including the climate crisis, public health, and public safety. The policy encourages agencies to responsibly experiment with generative AI while implementing adequate safeguards.

Furthermore, the policy stresses the importance of growing the AI workforce within federal agencies. To achieve this, the Biden-Harris Administration intends to hire 100 AI professionals by Summer 2024 as part of the National AI Talent Surge. This commitment is reinforced by guidance from the Office of Personnel Management, improving pay and leave flexibilities for AI roles.

To ensure effective implementation and oversight of AI usage in the federal government, agencies are required to designate Chief AI Officers to coordinate AI utilization. Additionally, the establishment of AI Governance Boards, chaired by the Deputy Secretary or equivalent, will govern the use of AI across agencies.

Alongside the policy, numerous other measures have been announced to promote responsible AI usage in the government. OMB will initiate a request for information (RFI) on responsible procurement of AI to inform future actions under federal contracts. Federal agencies will expand the 2024 Federal AI Use Case Inventory reporting, enhancing transparency of AI usage within the government. Lastly, the commitment to hiring 100 AI professionals further reflects the administration's dedication to trusted and secure AI implementation.

This policy marks a significant milestone in the Biden-Harris Administration's efforts to establish a robust framework for AI governance that ensures the safety, security, and trustworthy use of AI in the government. By following the Blueprint for an AI Bill of Rights and the NIST AI Risk Management Framework, this policy sets a clear baseline for managing risks, increasing transparency, and advancing responsible AI innovation for the public good.