Research
Research from The Generator, Babson College's interdisciplinary AI lab. Generator faculty publish in Harvard Business Review, The MIT Press, The Chronicle of Higher Education, and peer-reviewed journals in entrepreneurship, finance, law, writing studies, information systems, computational neuroscience, and leadership.
Featured
AI in Entrepreneurship Education: Interdisciplinary Action on the Frontier
Twenty-two Babson faculty contribute eighteen teaching innovations, organized around the five practices of entrepreneurship education: play, empathy, creation, experimentation, and reflection. The book is written for educators in every discipline, from humanities to engineering.
– From Chapter 1
– From Chapter 1
Distributed AI Leadership: The Generator as a Model for Faculty-Led Innovation
Girdharry and Wynstra examine the Generator as a case of distributed leadership: decisions about AI adoption shared across faculty-led specialty labs and the lab's "Family Conversations." The authors argue the model transfers to other institutions.
– From The Article
Publications & Commentary
Selected publications and commentary by Generator faculty and fellows.
How Generative AI Is Reshaping Venture Capital
Davenport and Noyes find that generative AI has reset the economics of starting up: teams that once needed a dozen engineers now need a few, seed rounds carry companies further, and some startups reach meaningful revenue without venture capital at all. Investors are shifting their attention to what AI cannot supply, including proprietary data, AI-native design, and founders who learn quickly.
Bridging Brains and Machines: Neural Circuits and the Networks Behind Modern AI
Zhu models how biological neural circuits represent stimuli and learn to predict, with direct analogs in the artificial networks used in machine learning. Two papers show how “semi-balanced” network states expand computational power and how homeostatic plasticity lets networks compute prediction errors.
Getting Learning Right: The Promise of Higher Education
Girdharry and co-authors argue that colleges facing public skepticism, political pressure, and AI should organize institutional decisions around how students actually learn. The book documents practices at institutions that do, from listening to students to codesigning courses with them, and draws on firsthand student accounts.
Entrepreneurship Education at the Dawn of Generative Artificial Intelligence
Published within months of the release of ChatGPT, Noyes and co-authors mapped what generative AI would mean for how entrepreneurship is taught; the article remains among the journal’s most-downloaded. They describe “a paradigm shift that mirrors the beginning of a new era not seen since the dawn of the internet.”
From Cheating to Cheat Codes: Integrating Generative AI Ethics into Collaborative Learning
Girdharry adapts the logic of gaming cheat codes, which invite exploration and reduce the fear of failure, into a framework for teaching generative AI ethics in collaborative classrooms.
Measuring and Mitigating Racial Disparities in LLM Mortgage Underwriting
Stein and co-authors submitted real loan applications, with experimentally varied race and credit scores, to leading large language models acting as underwriters. The models recommended more denials and higher interest rates for Black applicants; instructing the models to decide without bias eliminated the approval gap. The paper won two American Real Estate Society prizes and best paper at the New Zealand Finance Meeting.
Human Trafficking in the Global Supply Chain: Using Machine Learning to Understand Corporate Disclosures Under the UK Modern Slavery Act
Nersessian and a co-author apply natural language processing to more than 17,000 corporate statements filed under the UK Modern Slavery Act between 2016 and 2019. They find that compliance with the statutory mandate is limited and that clearer reporting standards are needed for disclosure laws to work.
The Innovation Navigator: Transforming Your Organization in the Era of AI
Fixson and a co-author set out how organizations should structure innovation work in the AI era, covering how AI changes design, product development, and collaborative work, and what that means for the people and processes behind new products. In a companion essay for the Lean Enterprise Institute, Fixson and a co-author argue that human judgment matters more, not less, as AI accelerates product development.
AI Demands a Fundamental Shift in How Higher Ed Organizes Knowledge
In the Chronicle’s report on AI leadership, Girdharry and Noyes argue that AI erodes disciplinary boundaries faster than universities can reorganize, and that knowledge structured in departmental silos cannot answer questions that arrive already interdisciplinary.
Generative AI in Entrepreneurship Research: Principles and Practical Guidance for Intelligence Augmentation
Kim and co-authors set out principles and a four-phase method, Discover, Develop, Discuss, Deliver, for using large language models across the research process, from literature review to peer review.
LEAP: A Method for Programming Education
Li and co-authors analyze 1,500 student queries to ChatGPT and GitHub Copilot in a graduate programming course. Questions about syntax fell from half of all queries to a fifth, while questions about concepts and problem-solving each rose to 40 percent.
Teaching Real-time Object Detection with an Emphasis on Engagement and Inclusiveness
Khachatryan and a co-author describe a method for teaching computer vision to business students who do not code: students collect the images, build and deploy a real-time object-detection model, and test it live. The approach is designed for entrepreneurship and information systems majors.
Ethics and AI Assemblages: A Heuristic Analysis of Undergraduate Business Student Perspectives
Girdharry and a co-author apply assemblage theory to the relationships among business students, educators, and AI tools in the classroom, and draw on student accounts to build a heuristic for teaching AI ethics.
Why Faculty Should Lead the AI Revolution
Noyes and Girdharry argue that faculty, not technology vendors or IT departments, should lead AI integration in business education. “University faculty can’t just rely on IT experts for the answers,” they write.
Utilitarianism in AI-Driven Solutions for Sustainability Development
At Harvard’s Green AI Summit, Li examined the utilitarian assumptions in AI-driven sustainability solutions on a panel covering agriculture, urban planning, and energy. Her research on media, markets, and environmental attitudes includes Environmental Advertising in China and the USA (Routledge).
Using Custom GPTs for Teaching Experiential, Project-Based Courses
Khachatryan and Sims, who lead the Generator’s AI & ML Empowerment and AI & Experiential Learning labs, showed how they configure custom GPTs for project-based business courses and gave attendees hands-on time with the tools their students use.
Six Types of AI Startups, Explained
Generator Fellow Tom Davenport recently published, with a co-author, a typology of AI startups: originators, explorers, infrastructure builders, enhancers, optimizers, and experimenters, the last “by far the largest cohort.”
A Writing Professor’s New Task in the Age of AI: Teaching Students When to Struggle
Girdharry cites research showing that students who used ChatGPT improved their essay scores without measurable gains in knowledge, and argues that teaching students when not to use AI is now part of a writing instructor’s job.
How Ambitious Entrepreneurs Can Use AI to Scale Their Startups
Generator Fellow Tom Davenport recently published, with co-authors, an account of how founders use AI to compress customer discovery and scale operations with small teams.
Not Just Another AI Statement: Modeling Process and Collaboration in Higher Education
Girdharry and a co-author argue that campus AI policies fail when issued as statements and hold up when built through cross-campus collaboration that can revise them as the technology changes.
I Teach AI and Entrepreneurship. Here’s How Entrepreneurs Can Use AI to Better Understand Their Target Customers.
Noyes describes how founders can use AI to analyze customer feedback, test personas, and reach strategic decisions faster.
Why AI Demands a New Breed of Leaders
Generator Fellow Tom Davenport recently published, with co-authors, an argument that the cultural and organizational changes AI requires exceed what most CIOs have the authority to lead, and call for an expanded leadership role.
What If the University Didn’t Fear the Machine?
Girdharry argues for meeting AI with curiosity rather than prohibition and describes what an engaged, critical institutional posture looks like in practice.
Meaningful Writing in the Age of Generative Artificial Intelligence
Girdharry and Khachatryan explain how generative pretrained transformers work, for faculty across disciplines, and propose ways to foster responsible use in writing-intensive courses.
Survey: How Executives Are Thinking About AI in 2026
Generator Fellow Tom Davenport recently published, with a co-author, results from the annual AI & Data Leadership Executive Benchmark Survey. Nearly every data and AI leader surveyed calls AI a high priority, plans to spend more on it, and reports measurable business value.
Engage with The Generator
By submitting this form, you agree to receive communications from Babson College and our representatives about our educational programs and activities via phone and/or email. You can unsubscribe at any time by clicking Unsubscribe from an email.
