Tech Ethics & Policy: Stanford HAI’s AI Fellowship Program Connects Students with Roles in Public Service

To bring technical expertise to the heart of U.S. lawmaking, Stanford HAI’s second cohort of Tech Ethics & Policy fellows headed to Washington, D.C., this summer. Among other projects, the students worked with public interest organizations to develop a public sector chatbot for people in need, finalize a bill on the procurement and use of artificial intelligence by the federal government, and teach AI literacy to policymakers.

As AI continues to exert its influence on every aspect of society, calls are increasing for the creation of federal policies to guarantee its responsible innovation and deployment. However, recent research from Stanford HAI finds that fewer than 1 percent of new AI doctoral graduates currently go into public service. This means that, all too often, those with the most knowledge about how this technology works are absent from those critical policymaking discussions. In response to this problem, Stanford HAI – in conjunction with Stanford’s McCoy Family Center for Ethics in Society – established the Tech Ethics & Policy Fellowship in 2023. The program is led by Rob Reich, HAI senior fellow and Stanford professor of political science, and Marietje Schaake, HAI Policy Fellow.

“As a policymaker who served in elected office, I have seen firsthand how hard it can be for the people passing laws and writing regulations to get access to objective technical expertise,” said Schaake. “Through this fellowship, we are offering Stanford students the opportunity to be part of that pipeline into public service.”

The 10-week program matches tech-savvy graduate fellows with summer jobs at executive branch offices, congressional offices, and think tanks in Washington, D.C. 

“Many students are excited about these opportunities and inspired and ready to go, but it’s not always clear to them how to get there, and the path can be intimidating,” says Drew Spence, policy program manager for Stanford HAI. “This program provides training on the policy world and how to navigate D.C., and gets fellows in the door so they can have an impact.” 

Fellows are required to have completed technical coursework in order to be considered, and once accepted, receive additional policy training. The program also includes an undergraduate track run by the Stanford McCoy Family Center for Ethics in Society.

Julia Lin (MS, Clinical Informatics Management), Jon Ball (PhD, Education Data Science), and Allison Jia (BS, Biomechanical Engineering; MS, Computer Science) were among the most recent cohort of 12 Tech Ethics & Policy (TEP) fellows. Here, they share their summer experiences and what future fellows should expect from the program.

Shaping U.S. Legislation

Allison Jia grew up in Silicon Valley intrigued by both medicine and tech innovation. Her current interest in biosecurity, she says, led to an interest in policy work.

“I was introduced to the idea that beyond technology, there’s policy that can influence people, and benefit or harm them,” she says. “I wanted to try doing something in AI policy to see how it differed from the world of tech and perhaps liaise between the two.”

Jia was placed at the U.S. Senate Committee on Homeland Security and Governmental Affairs (HSGAC), where she worked with a team to finalize the content of S.4495, the “PREPARED for AI Act,” sponsored by HSGAC Chair and Michigan Sen. Gary Peters. Intended to guide the federal government in effectively and responsibly procuring and using AI, the bill resulted in part from a hearing that 2023 TEP fellow Liana Keesing (BS ’23, MS ’24) helped organize.

“My job this summer was to help everyone navigate this bill and push it out of committee,” says Jia. “That involved our team talking and coordinating with over 100 different stakeholders across industry, civil society, and federal agencies, taking in feedback, modifying the bill, and generally working to create the best possible version of the bill.”

Jia provided her own feedback to the team and occasionally acted as a conduit between team members and other technical experts.

“If I had an objection to how something was defined in the bill, I’d do my research, provide evidence, and explain my reasoning,” she says. “They’d listen and change what I thought needed to be changed, so that was rewarding. I was also able to have conversations with experts in the field and then translate those insights into information that could influence the bill’s text.”

Evie Freeman, counsel for HSGAC and Jia’s manager, says the TEP program provides valuable support for policymakers.

“Chairman Peters understands the importance of getting AI policy right at a federal level,” she says. “So at HSGAC, it’s really important to us that we have voices in the room who have hands-on technical expertise. We take that seriously, and we take full advantage of having TEP fellows here.”

For now, Jia plans to continue building her tech expertise – perhaps working on technical solutions for AI governance or exploring medical robotics – but her long-term goal has shifted.

“Tech policy is something I’m newly passionate about,” she says. “I used to think the only way to make a positive impact was through developing technology, but now I see that policy is a lever that can help shape how we develop technology and perhaps align it with a vision the nation and the world may be able to agree on. I didn’t used to think this was a viable path for me, but now I think it’s something I definitely want to do in the future.”

Simplifying Health and Human Services

Julia Lin’s interest in technology and engineering intersects with a strong focus on how it can best be deployed to improve human health. She spent her TEP fellowship at the U.S. Department of Health and Human Services, where she worked with the “Facing a Financial Shock” team to help improve the delivery of benefits to individuals facing financial crises.

Lin also supported her manager in writing a proposal for an AI learning lab to support federal agencies implementing AI, interviewed federal, state, and local government leaders on their AI implementation efforts, and attended multiple congressional hearings.

Among her biggest takeaways: Policymaking and its implementation require tremendous coalition-building, public sector values and motivations differ from those in the private sector, and the D.C. environment isn’t as intimidating as it first seems.

“Until you work with policymakers, it feels very foreign and hard to know where to start,” she says. “I knew I had skills and knowledge; I just didn’t know how to translate it to the public sector. There’s nothing like living an experience to pull all the pieces together.”

Going forward, Lin hopes to pursue healthcare startup ideas, in particular machine translation technology to improve access for immigrant populations.

“I’m going to find a way to include policy in that work,” she says. “For me, the fellowship humanized bureaucracy. Government can seem like a black box, but the reality is often that it’s a small team working on a very large scope, which is amazing to think about.”

Tailoring AI Policy to Different Communities

With interests that include education, translation, and AI, Jon Ball couldn’t resist the opportunity to spend a summer in Washington, D.C., at the American Enterprise Institute (AEI) think tank.

“I wanted to get more involved with HAI and experience a summer in Washington, D.C., that would allow me to network,” Ball says. “This program gave me a wonderful launching pad to work on projects directly related to HAI and the collaboration they’ve had with AEI, but also to speak with Senate and congressional staffers, contacts provided by mentors, and basically as many people as possible in D.C.”

Ball worked with AEI’s workforce development team, where he attended and reported on meetings and collaborated with academics to finalize recent conference proceedings between Stanford and AEI. He provided guidance and AI technical support to researchers analyzing news media coverage and mapping the demand for skilled trades, and gave presentations on AI literacy. He came away from the experience interested in developing AI policy, particularly as it affects students at the state level.

“The opportunities to unite stakeholders in Washington, D.C., especially from different states, and especially in education policy, is fabulous, and something I didn’t see coming,” he says. “I thought it would be just federally focused, but my eyes were opened to how much opportunity there is to bring in people from different areas of the country to talk about their specific issues. A lot of the discourse on AI suggests that one size fits all, but that’s not even mildly accurate. There is so much work to be done tailoring AI policy and regulations to specific states, areas, and communities.”

Ball also came away from the experience satisfied that he and his cohort members were making substantial contributions to their respective agencies.

“The involvement of TEP fellows in planning was much greater than I had expected,” he said. “People were really looking to us to provide expertise and guidance.”

Learn more about the Stanford Tech Ethics & Policy Summer Fellowships.