AI and the Liberal Arts

In NEUR 301, students write an NIH-style grant proposal on a neuroscience topic of their choice. Students use generative AI as a tutor to help explain complicated neuroscience techniques they encounter in their literature searches. In addition, students are encouraged to use GenAI to help explore the critical questions in their topic (for example, mechanisms of cell death in Alzheimer’s disease) and help determine the advantages and limitations of the experimental approaches in their proposals.

In this seminar, students learn theoretical frameworks from both philosophy and computational and data sciences. Students work together to see how knowledge of frameworks from both disciplines serves to enrich our understanding of the ethical issues facing digital technologies.

This course is designed to support the development of student agency and skills in analyzing AI through humanities and media studies. Students examine AI from historical, cultural, and technological perspectives. Readings cover a range of methodologies from media history, media archaeology, and cultural studies to posthumanism, feminist theory, and decolonial approaches. This work encourages students to expand their understanding of what AI is and how it reshapes the cultural order, social relations, and human-machine interaction.

An introductory course open to all students, CS 110 provides a foundation in how computers, the web, and AI systems work while also examining how these technologies shape power, inclusion, bias, privacy, trust, and equity. Students participate in innovative activities such as a drone rescue simulation, confronting both the technical challenges and ethical trade-offs of autonomous systems. Through labs, collaborative projects, and guest lectures, students gain both practical skills and the critical lens needed to understand and influence the role of AI in society.

How has technology impacted religion? How has religion influenced technology? This course explores how digital technologies like the internet, social media, gaming, virtual reality, telecommunications, and AI have changed the way that people think about and practice religion. Throughout this course, students focus on the relationships between religion, digital media, robotics, and popular culture online using both real-world case studies and current research in the fields of religion, anthropology, and science and technology studies.

What is AI, and should humans fear it as one of “our biggest existential threats”? This course discusses the development of the field from the symbolic, knowledge-rich approaches of 20th-century AI to statistical approaches that rely on increasingly large amounts of data, including an overview of contemporary deep learning techniques. Students explore how to apply these techniques in several AI application areas, including robotics, computer vision, and natural language processing, and consider ethical issues around AI in society.

The world’s lack of preparedness for the COVID-19 pandemic revealed how urgently we need effective tools for vaccine and therapeutic development. In summer 2024, Seojean Kim ’26 and Sage Widder ’26 worked at the Marks Lab at Harvard Medical School, which has developed a machine learning model that predicts viral evolution in COVID-19 based on genetic sequences.

In summer 2024, Vicky Lee ’25 bridged technology and policy through two experiences: researching the reliability of large language model (LLM)-generated relevance judgments at the National Institute of Standards and Technology, and exploring national security theory within the context of U.S.-China relations with the Hertog Foundation. While working in D.C., she gained exposure to the intersection of AI and machine learning with public policy, particularly in the rigorous testing that informs safety-focused deployment of AI.

Using AI models for drug development has become a promising strategy in cancer research, as it reduces the time and cost to find possible starter molecules. During summer 2024, Lucia Urreta ’26 worked as an intern with Dr. Al-Lazikani at the MD Anderson Cancer Center, developing code to evaluate 3D molecular generative models using several metrics and analyzing intermolecular interactions between the generated molecules and their protein pockets.

Nina Howley ’27 and Dianna Gonzalez ’27 worked in summer 2025 under the advisement of Professor Eni Mustafaraj on CredBot. CredBot is a Chrome extension powered by LLMS that is capable of annotating articles based on six established credibility signals.

Margot Lang ’26 spent summer 2025 conducting research at the MIT Media Lab centered on leveraging audio data and AI to democratize music education and collaborative composition. She analyzed, processed, and curated a library of sounds to train AI models for an innovative instrument that co-creates a “global symphony.”

Is ChatGPT really a new coworker? Nadine Gibson ’28 spent summer 2025 at MIT Sloan School of Management, researching the similarities between how people engage with AI and engage with humans. A first-gen student, she developed research methods and glimpsed what the discovery, development, and dispute of knowledge looks like at the intersection of technology and society.

This collaborative project brings together researchers, including Sun-Hee Lee, professor of Korean, in corpus linguistics and AI to examine how immigrants are represented in Korean- and English-language media. By integrating large, annotated corpora with fine-tuned AI models such as BERT and GPT, it tracks discourse patterns over time and across regions, showing how language shapes public opinion on immigration. The study establishes a scalable, data-rich model for digital humanities research on migration and diversity.

Professor Orit Shaer and her students study how AI can enhance the team-based decisions made by emergency medical responders. Through interviews, experiments, and immersive VR simulations with EMS professionals, the team will develop a framework for human-centered AI tailored to emergency workflows. The findings will not only advance emergency care but also inform AI design in other high-stakes domains, from disaster response to automated transportation.

Intensive longitudinal studies of suicide offer opportunities to understand how suicidal thoughts unfold in real-world settings using smartphone surveys and wearables; however, this data comes with several challenges that bar the use of traditional machine learning methods. Led by Yaniv Yacoby, assistant professor of computer science, researchers in the MOGU Lab address those challenges by developing a new class of generative models to help better identify types of at-risk patients, forecast when patients are at imminent risk, and help prevent suicide and related behaviors.