The AI leadership imperative: Preparing federal agencies for AI’s impact

A new report underscores the urgent need for federal government leaders to help their executives better understand and embrace artificial intelligence’s rapid emergence to meet the challenges and opportunities that AI’s impact will have on their organizations.

The new report, “Leading Agency Innovation in the Age of AI,” asserts that federal officials must make a broader effort to educate their leadership teams and support a work environment that encourages leaders to identify appropriate AI use cases and lay the groundwork for moving from the possible to the practical.

The report, produced by Scoop News Group and underwritten by Microsoft, highlights the work of one organization — the nonprofit, nonpartisan Partnership for Public Service — that has been working behind the scenes across government to cultivate a growing cohort of senior executives to more fully understand what AI can and cannot do and how to put AI to practical use at their agencies.

The Partnership’s AI Federal Leadership Program brings together qualified senior executives through a six-month course that explores AI’s capabilities and potential impact on government agencies and culminates in executives having to develop an AI project roadmap to implement at their agencies. Since its inception, more than 500 senior executives from 40 agencies across the federal government and over 30 states have completed the program.

“The program not only serves as a model for training government leaders about AI and its impact but is also producing a growing — and much-needed — cohort of senior government executives who are better equipped to guide their agencies through AI transformation,” the report says.

The report highlights several elements that are important for agency leaders to embrace if they are to prepare for the expected impact AI will likely have on their workforce, their operations, and their missions:

Sharing lessons learned: A vital aspect of the program is the opportunity for participants to share their AI aspirations and application lessons with their peers. As Nancy Potok, a former program coach, notes, “The mingling of people who have different levels of technological expertise and experience and come from very different organizational cultures is one of the strengths of the course because they learn from each other.”

Access to AI experts: The program connects participants with technical experts on the front lines of AI development. A chief technology officer from a cabinet-level department emphasizes the value of this access, stating, “The challenge with the federal government is that there are probably [only] about 50,000 people in the world that can engage you about AI and know what they’re talking about.” That makes it difficult, he says, for senior executives to get a first-hand feeling of how AI works at an enterprise level and how to strategize its use.

Focusing on the problem, not just the solution: Patricia Cogswell, a former Homeland Security executive who served as a program facilitator, highlights the importance of defining the problem before selecting an AI use case. She cautions against simply “picking a solution and then finding a problem,” emphasizing the need to identify mission-critical challenges that AI can help address.

The report also includes 12 lessons that participants collectively say they learned from their experience in the program and the pilot projects they later developed at their agencies.

Additionally, it highlights examples of ways federal agency leaders, including Eric Stein, Deputy Assistant Secretary for Global Information Services at the U.S. State Department, applied the lessons they learned from the AI leadership program at their agencies.

If the federal government is to gain genuine traction in adopting AI, agencies — and the government — need to take a more orchestrated approach to training federal workers at all levels, says one cabinet-level chief technology officer in the report. He recommended AI orientation and training programs that focus on the needs of:

• Executives – so that they become familiar with the language of AI and what it can do.
• Middle managers – because they are the people consuming vendor services, they must be able to tell them what’s needed and what to buy.
• Acquisition specialists – because they need to know how to write the contracts.
• End users – so that people can understand how to use AI correctly (and know not to upload data into some public engine.)
• Security specialists – because nobody understands how AI will be misused.

Another federal AI leader cited in the report stressed the importance of experimentation. “Being able to see something that is operational — that may not be the end product but that gets you halfway there and then be able to iterate — is critical to get momentum, to give people an idea of what the art of the possible is,” he said.

Download the full report to learn more about “Leading Agency Innovation in the Age of AI.”

This article was produced by Scoop News Group for FedScoop and underwritten by Microsoft.