Students come to Harvard Kennedy School to pursue their passions, learn about complex issues, and gain insight into public policy firsthand. Many students in our two-year programs put their classroom learning into action through internships between their first and second years.
This summer, several students explored how artificial intelligence is reshaping economies and communities around the globe—from rural healthcare systems to workforce development.
Read about their experiences.
- Slavina Ancheva MPP 2026
- Teresa He MPA/ID 2026
- Charlie Kidd MPP 2026
- Michaela Kocher MPP 2026
- Jade Lin MPP 2026
- Arpita Wadhwa MPP 2026
__
Slavina Ancheva MPP 2026
Where did you intern this summer? What projects did you work on?
This summer, I interned at the Organisation for Economic Co-Operation and Development (OECD) in the Centre for Entrepreneurship. My main project was a discussion paper for several OECD countries on AI diffusion and uptake by small and medium enterprises. The project was exciting because it involved collaborating with different experts across the OECD and learning the ins and outs of providing this type of policy expertise to ministries.
What did you learn over the course of the summer?
I learned a lot about how small and medium enterprises are integrating AI into their workflows. The diffusion of AI throughout the economy and its uptake by smaller companies across different sectors will be key to unlocking the promised benefits. Through research, case studies, interviews, and data analysis, I now have a better understanding of the key enablers and challenges to AI adoption by small businesses. Learning this at a micro-enterprise level has been key to understanding the broader impacts across the economy.
What was your favorite part of the experience?
The OECD mission of “better policies for better lives” is what initially drew me to the organization. I am very thankful to have had the opportunity to learn about the OECD’s work and how it functions as an organization. This internship been especially meaningful because my home country of Bulgaria is an accession candidate to the OECD.
How did your summer internship fit into your academic and professional goals?
As someone who wants to stay involved in the AI policy space, it has been incredibly valuable to gain this more international experience. Having previously worked in the EU, the opportunity to work with an organization representing all OECD member states was especially new to me. Learning about the business case studies has also been helpful in understanding the link between the public and private sectors. Overall, this summer has provided me with a solid understanding of where countries stand on AI diffusion and uptake, the key challenges they face, and the policies they are implementing to address them.
Why is AI policy important at this moment in time?
AI is already redefining everyday lives and broader sectors of the economy. The policies we put in place now—either to mitigate risks or promote access to the technology—will have implications for years and decades to come. It’s not just about the technology; it’s about asking ourselves what type of world we want to live in one day.
Teresa He MPA/ID 2026
Where did you intern this summer? What projects did you work on?
I interned at the Agency Fund. As part of my role, I co-hosted the AI Tech Accelerator for nonprofits in collaboration with OpenAI and developed valuation models using discounted cash flow to assess impact in the healthcare, agriculture, and education sectors.
What did you learn over the course of the summer?
I learned how to leverage AI to help the healthcare and education sectors be more efficient and have greater impact.
What was your favorite part of the experience?
I enjoyed connecting with the local entrepreneurs on the ground in Africa and solving problems related to improving their AI model.
How did your summer internship fit into your academic and professional goals?
The experience opened my mind to how advanced technology can amplify the work of nongovernmental organizations in developing markets.
Why is AI policy important at this moment in time?
AI is moving faster than the rules that guide it. The choices we make today will shape whether it benefits everyone or just a few. While AI keeps improving people’s productivity, the productivity gap between developed countries and developing countries continues to broaden. Good policy now means innovation with guardrails—protecting rights, reducing harm, and building trust from the start.
Charlie Kidd MPP 2026
Where did you intern this summer? What projects did you work on?
This summer, I interned as a technical consultant at Catalyst, a nonprofit science and technology innovation hub in Northern Ireland. I used AI tools to build a comprehensive analytics platform that provides real-time data on the startups they support. The platform includes regional economic impact metrics, fundraising and grant data, jobs growth indicators, AI-generated performance insights, and impact assessments for their support programs.
What did you learn over the course of the summer?
I gained a deeper understanding of how regional economic development can be driven by innovation in science and technology. I’ve also significantly improved my data science and coding skills, especially in developing and deploying analytical tools for real-world use.
What was your favorite part of the experience?
Visiting Belfast and spending time with my team was a highlight—both for the professional collaboration and the chance to learn more about the region’s unique history and context. I also loved seeing how quickly my technical capabilities have progressed.
How did your summer internship fit into your academic and professional goals?
After graduation, I hope to work on policies and initiatives that support science and tech entrepreneurship in UK regions beyond the traditional Oxford–Cambridge–London corridor. This internship has given me first-hand experience applying tech to support regional innovation and a clearer vision for how policy can enable that.
Why is AI policy important at this moment in time?
There has never been a more critical time to be tech-literate. Just a year ago, I couldn’t write a line of code—now I’m building tools that help policymakers and entrepreneurs make better, data-informed decisions. The accessibility of AI tools is lowering the barrier to entry for solving complex problems and driving economic growth in places that have traditionally lacked technical talent or venture capital. You no longer need a Silicon Valley software engineer to build tech that solves real-world problems—it’s an extraordinary moment for inclusive innovation.
Michaela Kocher MPP 2026
Where did you intern this summer? What projects did you work on?
I was an intern at Forte Global, a workforce development startup that uses outcomes-based financing to align incentives among employers, training providers, and learners. The Forte model helps individuals access high-demand skills training and career opportunities at no personal cost. This summer, my work focused on two main areas.
- Delivery scalability: I spent time working with the Delivery team in Medellín, Colombia to analyze Forte’s workforce development delivery processes and provide recommendations for how to improve the scalability of these processes.
- AI thought leadership: I researched and published content on the role of AI in workforce development, including strategies for upskilling and ensuring equitable access to emerging technologies, as well as on how to evaluate AI upskilling programs.
What did you learn over the course of the summer?
I learned a lot about the nature of working at a startup—fast-paced decision-making, resource constraints, and the need for adaptable processes.
I deepened my understanding of how workforce development initiatives operate in practice, especially the importance of aligning training programs with employer demand and embedding feedback loops into delivery.
Finally, I saw how solutions that work in one country or region may not translate directly to another. Cultural norms, labor market structures, and educational systems vary widely, so global solutions (such as Forte’s unique outcomes-based financing model) need to be adapted to fit local contexts to be effective and sustainable.
What was your favorite part of the experience?
I enjoyed getting to experience Forte’s work from two very different vantage points. I spent part of the summer in New York City working closely with the CEO and CFO on strategic initiatives, and part of the summer in Medellín, Colombia alongside the team implementing programs on the ground. Seeing the high-level decision-making and the day-to-day realities of delivering workforce development initiatives gave me a richer, more nuanced understanding of how strategy connects to impact.
How did your summer internship fit into your academic and professional goals?
When I started at HKS, my focus was on how technology could be harnessed for social impact. What I didn’t fully appreciate then was how essential workforce development is to translating technological change into real economic mobility. Over the past year, I’ve studied workforce development in multiple classes and applied those lessons to policy projects. This summer at Forte gave me the chance to see those ideas operationalized.
I’ve also long been interested in entrepreneurship, particularly ventures that put social impact at the center of their business model. At Forte, I was able to see a functioning, impact-first business up close and understand how its structure, incentives, and culture differ from what I experienced in government and the private sector. That perspective will be critical as I shape my own career at the intersection of technology, workforce development, and mission-driven entrepreneurship.
Why is AI policy important at this moment in time?
AI and digital transformation are reshaping industries, labor markets, and pathways to opportunity at an unprecedented pace. This creates immense potential to expand social mobility if workforce development systems can adapt quickly enough to equip people with the skills and connections they need. Without intentional policies and program design, these changes risk leaving behind those without access to training, digital tools, or professional networks. However, if we are intentional, we can harness the benefits of technological change to create inclusive career pathways, bridge skills gaps, and ensure that the future of work works for everyone.
Jade Lin MPP 2026
Where did you intern this summer? What projects did you work on?
I interned at the Patrick J. McGovern Foundation. I primarily worked on a project looking at AI’s impact on frontline work, focusing on home care workers.
What did you learn over the course of the summer?
So much! I learned a lot about how philanthropy works, about cutting-edge work on AI for social good, and about some extraordinary businesses and innovations in home care.
What was your favorite part of the experience?
I met a lot of smart, inspiring people who could be making a lot more money in other industries, who are instead working towards AI for social good.
How did your summer internship fit into your academic and professional goals?
My academic and professional goals are to help governments—particularly the Australian government—make life better for their citizens. I’m proud to be from Australia because it has one of the best welfare states in the world, particularly for people with disabilities. It is a place where rich, exciting policy changes are possible.
Spending the summer focused on how AI might change home care in Australia was exciting because it genuinely might make people’s lives better.
I’ve developed a few core these at HKS that this internship touched on:
- Having a good job gives people meaning, which helps hold democratic fabric together.
- Lots of jobs—like home care—are critically important but not very “good”—they often do not pay well, require hard work, and can be unstable.
- There are ways of re-engineering industries, particularly at technological inflection points like AI’s boom, to make jobs better.
- AI is coming to all industries. Actively working through its risks and opportunities is the only way to secure better futures for workers.
Why is AI policy important at this moment in time?
The whole premise of AI for social good is that it can scale impact massively. If you think about tasks that cash- and time-constrained nonprofits do manually that could be automated by AI—think a person watching hours of footage on fishing boats to count how many fish have been brought on—you realize we could do a lot more good with our current resources.
In the case of frontline work, we’re seeing unions and employers have showdowns on bargaining agreements over AI. A lot of those agreements rightly are asking for an employee voice at the table, or restrictions on uses of AI that invade a worker’s privacy, or automatically result in a firing. But there’s also a huge opportunity to buy and build AI tools that impact workers positively in this massive period of change.
A great example is Jolly, which is an incentives platform that gives employees incentives to do things that employers want (e.g., take an undesirable shift or refer a friend). They’re providing win-wins by definition.
Arpita Wadhwa MPP 2026
Where did you intern this summer? What projects did you work on?
This summer I worked with Microsoft Research, where I designed and implemented a randomized controlled trial in rural Rajasthan, India. The study focuses on quantitatively measuring trust and adoption of AI-powered chatbots for Accredited Social Health Activists (ASHAs), India’s one-million-strong, all-women cadre of community health workers who form the backbone of rural health delivery. ASHAs operate in resource-constrained settings, often without immediate access to supervisors, and are frequently tasked with sensitive and time-critical questions from families. An ASHA may need immediate support when a pregnant woman shows warning signs of complications where delays in referral can be life-threatening, when parents seek clarity on new vaccines such as those for COVID-19, or when misinformation in the community fuels reluctance to seek institutional care.
AI-powered chatbots have the potential to provide ASHAs with timely, verified guidance that shortens their turnaround in the field and supports more confident decision-making, thus improving health outcomes. Whether these tools are trusted, however, depends as much on culturally attuned design and framing as on accuracy. The project demonstrates how such design choices can make AI tools socially legitimate and practically usable in public health delivery.
What did you learn over the course of the summer?
A central learning was that trust in technology is never built on accuracy alone; it is shaped by culture.
Much of Western user experience design assumes individuals act in isolation, weighing health risks and benefits on their own. In the settings where ASHAs work, however, decisions about health are relational, shaped by peers, families, and supervisors. I saw that adoption often hinged less on what the chatbot said than on how it was framed, whether it resonated with what colleagues might do or carried the weight of authority. Despite frequent calls for “culturally appropriate” design, it remains a black box in terms of which aspects of culture matter most and how they should be embedded into technology. This summer allowed me to begin unpacking that question and identify the impact of normative framing in building calibrated trust toward new tools.
What was your favorite part of the experience?
The most meaningful part of the internship was working directly with frontline health workers in Rajasthan and seeing design choices translate into real adoption. Their reflections showed that trust stems not only from accuracy but also from perceived legitimacy and resonance.
I also came to see how strong the demand for tools like this is, given the uncertainty and sensitivity of the issues ASHAs face every day. They are often asked difficult questions such as how to respond to a child’s side effects after vaccination, which can fuel parental hesitancy, or how to guide families reluctant to choose institutional delivery for pregnant mothers. Despite the existence of many health information resources, ASHAs have tended to rely on supervisors because those were the voices they trusted. Culturally attuned chatbots provided them with access to credible information in real time, which made their turnaround in handling challenges in the field much quicker and their responses more confident. Seeing this alignment between design theory, lived challenges, and actual adoption was the most rewarding aspect of the summer.
How did your summer internship fit into your academic and professional goals?
This internship allowed me to bring together my interests in evaluation, technology, and governance. I designed and implemented a civic technology intervention and am now analyzing its outcomes and writing a paper that connects experimental evidence to broader policy debates on AI governance. Looking ahead, I aim to work at the intersection of public interest technology and product design, focusing on how digital tools can be built, tested, and scaled responsibly in low-resource contexts. This project was an important step in that trajectory—it gave me the opportunity to approach an emerging technology not only as a research question but also as a product that must earn trust, deliver value, and prove its usability in the field.
Why is AI/tech policy important at this moment in time?
AI is becoming central to how we deliver health, education, and welfare. In resource-constrained contexts, technology can play a critical role in closing gaps that weaken public systems: providing timely and verified guidance when supervisors are unavailable, reducing the delays that make families lose confidence in services, and addressing the leakages that undermine large-scale programs. Yet without thoughtful design and governance, AI can just as easily amplify inequities, spread misinformation faster, or further erode trust.
My work this summer underscored that building credibility in these systems is not simply a technical question of accuracy, but also a social one rooted in norms, relationships, and legitimacy. Embedding these dimensions into product design and policy is essential if AI is to become a true public good—one that improves service delivery and builds trust in institutions, rather than replicating their weaknesses.
Hero image: Adobe Stock. In-line images courtesy of Slavina Ancheva, Teresa He, Charlie Kidd, Michaela Kocher, Jade Lin, Arpita Wadhwa