Insights

Implementing M-24-10: A Transformative Path for Federal Agencies

By: Brennan Hart

In response to President Biden's Executive Order on "Safe, Secure, and Trustworthy Artificial Intelligence" issued in October, the Office of Management and Budget (OMB) released Memorandum M-24-10, titled "Advancing Governance, Innovation, and Risk Management for Agency Use of Artificial Intelligence." This memorandum is a cornerstone for federal agencies, outlining the steps needed to align with the President's vision for an AI-driven future that is both innovative and ethically grounded. Here's an in-depth look at how M-24-10 is poised to reshape federal agencies and the necessary actions to implement its directives.

The AI Executive Order

The Executive Order laid the groundwork for responsible AI development and use within the federal government. It emphasized the need for AI technologies that are developed and deployed in a manner that is safe, secure, respects privacy, combats discrimination, and maintains trustworthiness. Following this vision, M-24-10 serves as the actionable guidance for federal agencies, detailing the specific measures to adopt in order to realize these overarching goals.

The Role of M-24-10 in Federal Agencies

M-24-10 is not merely a set of guidelines but a transformative policy instrument designed to usher in a new era of AI governance within the federal sector. It establishes a comprehensive framework that agencies must follow, which includes the establishment of Chief AI Officers (CAIOs), the development of AI governance structures, and the implementation of risk management practices tailored to AI's unique challenges.

Establishing Chief AI Officers

One of the memorandum's first mandates is for agencies to appoint CAIOs within 60 days. These officials will spearhead AI strategy, governance, and risk management efforts, ensuring that AI deployments are consistent with federal policies and ethical standards. CAIOs will serve as the central point of coordination for AI-related activities, bridging various departments and fostering an integrated approach to AI innovation and ethics.

Advancing Responsible AI Innovation

Federal agencies are directed to develop enterprise strategies for advancing responsible AI use. This involves identifying opportunities for AI to enhance mission outcomes, while also addressing barriers to AI adoption. Agencies must consider factors such as IT infrastructure, data availability, cybersecurity, and workforce capabilities to create an environment where AI can thrive under the principles of responsibility and equity.

Managing AI Risks

M-24-10 places significant emphasis on risk management, acknowledging the dual nature of AI as a powerful tool for innovation and a potential source of risk. Agencies are tasked with conducting AI impact assessments, testing AI in real-world scenarios, and establishing continuous monitoring mechanisms. These practices ensure that AI systems are deployed safely and effectively, with ongoing evaluations to address any emerging risks or unintended consequences.

Implementing the Guidance

For federal agencies, the journey toward implementing M-24-10 involves several key steps:

1. Appointing CAIOs: Agencies must quickly identify and appoint skilled individuals to fill the CAIO role, ensuring they have the authority and resources needed to effect change.

2. Developing AI Strategies: Crafting a comprehensive AI strategy that aligns with the agency's mission and the broader goals of the Executive Order is critical. This strategy should detail how the agency will address innovation, governance, and risk management in the context of AI.

3. Conducting Impact Assessments and Testing: Agencies need to establish processes for rigorously assessing and testing AI systems before and after deployment, with a clear focus on ethical considerations and public impact.

4. Fostering Collaboration and Sharing: Building a culture of collaboration and sharing within and across agencies will be vital for leveraging AI's full potential while ensuring responsible use.

The issuance of M-24-10 in the wake of the President's Executive Order represents a pivotal moment for federal agencies in the realm of AI. It provides a clear path forward for harnessing AI's capabilities in a manner that is ethical, secure, and beneficial for all Americans. As agencies embark on this path, the guidance from M-24-10 will be instrumental in shaping a future where government operations are more efficient, transparent, and aligned with the values of trust and equity. Implementing these directives is not just about compliance; it's about leading by example in the responsible use of one of the 21st century's most transformative technologies.