As we move further into the digital age, artificial intelligence (AI) has become a prominent force across all sectors, particularly in the federal government. Agencies of all sizes are embracing AI to streamline processes and tasks, ultimately making workflows more efficient. The potential of this emerging technology in enhancing agencies’ ability to serve the public and carry out their missions is virtually unlimited.

In light of the increasing prevalence of AI, the Office of Management and Budget (OMB) released a draft memo outlining 10 requirements for the implementation of AI across the federal government. This development came just before the Biden Administration teased an upcoming federal executive order on AI. The memo includes a recommendation to establish an AI governance board to provide essential oversight and guidance on the use of this rapidly evolving technology.

The evolving nature of AI and its rapid implementation pose a challenge to government agencies looking to establish governance protocols. Like the process of adopting cloud technology a few years ago, the adoption of AI goes beyond simply transforming IT; leaders must grapple with myriad questions regarding its potential uses and drawbacks. For agencies venturing into AI adoption, seeking guidance from cross-sector experts can provide a comprehensive understanding of where to begin and what potential pitfalls to anticipate.

Forming partnerships with industry experts, establishing advisory committees, and seeking advice and assistance from those who have expertise in AI is crucial in navigating the complexities of AI governance. What the government leaders driving AI adoption need is a framework that is actionable and implementable, enabling them to swiftly move forward. Consequently, the involvement of federal Chief Information Officers (CIOs) and Chief Information Security Officers (CISOs) is key, as they eagerly await a solution that enables the responsible deployment of AI workloads to fulfill their agencies’ missions.

Supporting the safe adoption of AI are several resources developed by industry experts and academics, such as the NIST AI Risk Management Framework, the MITRE ATLAS project, OWASP’s top 10 security considerations for large language models, and the IEEE program for ambassadors of AI systems. Moreover, efforts by governing bodies like Congress, the White House, and agencies like DHS and NIST to establish standards can be complemented by leveraging existing governance models, including FISMA and FedRAMP ATOs, and creating overlays for AI.

As the federal government prepares to issue guidance on AI and an executive order, it is imperative that safety and security are prioritized. The guidance should focus on promoting the safe and secure adoption of AI, directing the implementation of robust safety and security requirements, and expanding successful cybersecurity programs to cover AI systems. Legislators and government leaders must emphasize mandatory security and safety measures, underpinned by strong regulatory oversight, to ensure the secure adoption of AI across agencies and connected infrastructure.

In conclusion, as AI continues to gain traction within the federal government and across various industries, the need for structured governance and guidance becomes increasingly crucial. By leveraging partnerships, seeking advice from experts, and focusing on safety and security, federal agencies can embrace AI with confidence and ensure that its implementation is in line with the broader public interest.

Source link