Ethan Mollick
Last reviewed
Sources
8 citations
Review status
Source-backed
Revision
v2 · 1,693 words
Improve this article
Add missing citations, update stale details, or suggest a clearer explanation.
Suggest edit
CtrlK
Last reviewed
Sources
8 citations
Review status
Source-backed
Revision
v2 · 1,693 words
Add missing citations, update stale details, or suggest a clearer explanation.
Ethan R. Mollick (born 1975) is an American academic and author who is an associate professor of management at the Wharton School of the University of Pennsylvania and one of the most widely read commentators on the practical use of artificial intelligence in work and education. He holds the named position of Ralph J. Roberts Distinguished Faculty Scholar, is a Rowan Fellow, and is co-director of the Wharton Generative AI Labs. [1][2] Mollick is best known for his 2024 book Co-Intelligence: Living and Working with AI, an instant New York Times bestseller, and for his One Useful Thing newsletter, which translate research and hands-on experimentation with generative AI into accessible guidance for a general audience. [2][3]
Trained as a scholar of entrepreneurship, innovation, and crowdfunding, Mollick redirected much of his work in late 2022 toward how large language models such as ChatGPT change knowledge work, teaching, and organizations. [2] He co-authored a widely cited field experiment, run with Boston Consulting Group, that introduced the idea of a "jagged frontier" of AI capability, and he has become one of the most read voices urging professionals, educators, and students to experiment directly with the technology rather than wait for it to mature. [4][5] In 2024, Time magazine named him to its annual list of the 100 most influential people in artificial intelligence. [6]
Mollick was born in 1975 in Milwaukee, Wisconsin. [1] He earned a bachelor's degree from Harvard University, graduating magna cum laude in 1997. [1] He then completed both of his graduate degrees at the Massachusetts Institute of Technology Sloan School of Management, receiving an MBA in 2004 and a PhD in management in 2010. [1] His doctoral dissertation, titled "Essays on individuals and organizations," was supervised by the economic sociologist Ezra Zuckerman and examined how individual people, rather than firms or markets, shape innovation and organizational outcomes. [1]
At Wharton his official title is Associate Professor of Management, and he is listed as the Ralph J. Roberts Distinguished Faculty Scholar, a Rowan Fellow, and Co-Director of the Generative AI Labs. [2] He has also been named MBA Professor of the Year by Poets and Quants for his work using AI and simulations in the classroom. [2]
For most of his career Mollick studied entrepreneurship and innovation, with particular attention to crowdfunding and to the influence of individuals on the innovation process. His research on platforms such as Kickstarter examined how founders raise money from the public, why some campaigns succeed, and how solo founders differ from teams, and it helped establish crowdfunding as a serious subject of management scholarship. [1][2]
A second strand of his work has been teaching itself. Mollick is a longtime advocate of active, experiential learning, and through Wharton Interactive he designed educational games and simulations, including The Startup Game and an entrepreneurship simulation called ARC, that are used by instructors at many institutions. [2] That combined interest in how new technologies diffuse and in how people learn positioned him to respond quickly when general-purpose AI tools arrived for the public. [2]
Mollick began writing intensively about generative AI in late 2022, shortly after the public release of ChatGPT, on a Substack newsletter called One Useful Thing, whose tagline is "Trying to understand the implications of AI for work, education, and life." [2][8] Written without sponsorship, the newsletter pairs plain-language explanations of fast-moving AI research with concrete examples of using the tools for writing, analysis, teaching, and management. It grew rapidly, and as of 2026 it reports over 452,000 subscribers, which makes it one of the most read publications on the everyday use of AI. [8]
Mollick's most influential research contribution is a 2023 field experiment titled "Navigating the Jagged Technological Frontier." Conducted with Boston Consulting Group and co-authored with researchers at Harvard, MIT, and the University of Warwick, the preregistered study randomly assigned 758 BCG consultants to work with GPT-4, with GPT-4 plus a prompt engineering overview, or without any AI, across 18 realistic consulting tasks. [4] The paper was issued as Harvard Business School Working Paper 24-013 in September 2023 and was later published in the peer-reviewed journal Organization Science. [4]
The headline results, for tasks that fell within AI's competence, are summarized below. [4]
| Outcome (within the frontier) | Effect of using GPT-4 |
|---|---|
| Tasks completed | 12.2 percent more than the control group |
| Speed | 25.1 percent faster |
| Quality of output | About 40 percent higher, on average |
| Effect by skill level | Largest gains for lower-performing consultants |
The gains were largest for lower-performing consultants, whose scores rose the most, which suggested that AI can narrow the gap between weaker and stronger workers. [4][5] On a problem the researchers deliberately chose to sit outside that frontier, however, consultants using AI were correct less often than those working without it, roughly 60 to 70 percent versus 84 percent, because they tended to accept confident but wrong answers. [5]
The study is the source of two ideas closely associated with Mollick. The first is the "jagged frontier": the observation that AI is unexpectedly strong at some tasks and unexpectedly weak at others, with no obvious boundary between the two, so that tasks of seemingly similar difficulty can fall on opposite sides of the line. [5] As Mollick put it, "Everything inside the wall can be done by the AI, everything outside is hard for the AI to do." [5] The second idea is his distinction between "Centaurs," who divide a job cleanly between themselves and the AI, and "Cyborgs," who blend their own effort with the AI's at the level of individual subtasks. [5]
Within the university, Mollick co-founded the Wharton Generative AI Labs with his wife, Lilach Mollick, who serves as a senior fellow and the labs' lead on pedagogy; Ethan Mollick is co-director. [2][7] The labs combine research with prototyping, building and testing practical AI applications and instructional tools meant to change how people work and learn while studying how to deploy them responsibly. [7] In December 2024, the technology entrepreneur Mark J. Pincus pledged a gift to expand generative AI work at Wharton, supporting a Pincus AI Lab within the effort. [7]
On April 2, 2024, Portfolio, an imprint of Penguin Random House, published Mollick's book Co-Intelligence: Living and Working with AI. [3] Written for a general reader, it argues that people should treat AI as a kind of collaborator, a "co-intelligence," and learn its strengths and limits through direct use rather than abstract debate. The book was an instant New York Times bestseller and was named a best book of the year by The Economist and the Financial Times. [3] Its premise begins with the moment generative AI reached the public: "Something new entered our world in November 2022, the first general purpose AI that could pass for a human and do the kinds of creative, innovative work that only humans could do previously." [3]
Co-Intelligence is organized in part around four practical rules that have been widely reproduced by educators and trainers. [3]
| Rule | Meaning |
|---|---|
| Always invite AI to the table | Try AI on every task to learn where it helps and where it does not, since no one yet knows its full range. |
| Be the human in the loop | Keep people responsible for checking, judging, and correcting AI output, especially where errors carry real cost. |
| Treat AI like a person, but tell it what kind of person to be | Interact with it in natural language and assign it a clear role or persona to get better results. |
| Assume this is the worst AI you will ever use | Build habits and processes that will keep improving as the underlying models get better. |
Mollick's clearest influence has been on how working professionals and, especially, educators approach generative AI. Rather than emphasizing restriction, he has argued that teachers and managers should adopt the tools openly, redesign assignments and workflows around them, and share what they learn. Through One Useful Thing, the Wharton Generative AI Labs, and frequent talks, he and Lilach Mollick have published prompt libraries and practical guides for using AI as a tutor, a feedback tool, and a simulation partner in the classroom, material that has been taken up by schools and universities around the world. [2][7]
His vocabulary has entered common use in discussions of AI in education and the future of work: the "jagged frontier," the "Centaur" and "Cyborg" working styles, and the injunction to "always invite AI to the table" are now frequently cited shorthand. [5] As of 2026, Mollick continues to teach at Wharton, to direct the Generative AI Labs, and to write One Useful Thing, remaining one of the most widely followed independent interpreters of what generative AI means for everyday work and learning. [2][8]