About

Andrew W. Lo is the Charles E. and Susan T. Harris Professor at the MIT Sloan School of Management and the director of MIT's Laboratory for Financial Engineering. He is also a Principal Investigator at the Computer Science and Artificial Intelligence Laboratory (CSAIL), an affiliated faculty of the Department of Electrical Engineering and Computer Science, a member of the Operations Research Center (ORC), and the Institute for Data, Systems, and Society (IDSS), all at MIT. Additionally, he is an external faculty member at the Santa Fe Institute. Lo received his AM and PhD in economics from Harvard University, his BA in economics from Yale University, and graduated from the Bronx High School of Science. His academic career began at the University of Pennsylvania's Wharton School, where he served as an Assistant and Associate Professor. His current research spans several areas including evolutionary models of investor behavior and adaptive markets, systemic risk and financial regulation, quantitative models of financial markets, financial applications of machine-learning techniques and secure multi-party computation, healthcare finance, and deep-tech investing such as fusion energy and advanced manufacturing. Lo has published extensively in academic journals and authored the book 'The Adaptive Markets Hypothesis: An Evolutionary Approach to Understanding Financial System Dynamics.' His work has earned numerous awards, including Sloan and Guggenheim Fellowships, the Paul A. Samuelson Award, the Harry M. Markowitz Award, and the James R. Vertin Award from the CFA Institute. He has been recognized as one of TIME’s 2012 '100 most influential people in the world' and has received multiple teaching awards from the University of Pennsylvania and MIT. Lo is also actively involved in various industry and nonprofit ventures, serving as a cofounder and board member of BridgeBio Pharma and Uncommon Cures, a cofounder of AlphaSimplex Group, QLS Advisors, QLS Technologies, Quantile Health, and Rutherford Energy Ventures. He is a board member of GCAR, n-Lorem, and Vesalius, and an investor and advisor to several biotech companies and organizations. His research and contributions have significantly impacted the fields of financial engineering, risk management, and innovative investment strategies.

Research topics

• Economics
• Computer Science
• Artificial Intelligence
• Business
• Finance
• Mathematics
• Statistics
• Physics
• Applied mathematics
• Econometrics
• Statistical physics
• Medicine
• Industrial organization
• Microeconomics
• Pharmacology

Selected publications

• Breaking Bad Financial Habits: How LLM Conversations Correct Financial Misconceptions

  ArXiv.org · 2026-04-29

  articleOpen accessSenior author

  Financial misconceptions carry direct economic costs, from panic selling to equity market avoidance, yet they are notoriously resistant to correction. Traditional financial literacy interventions are constrained by cost, reach, and a persistent gap between knowledge and behavioral change. Across three pre-registered studies, we find that purposefully designed LLMs can durably correct financial misconceptions. Critically, two factors are necessary for this effect. First, corrective intent: LLMs prompted only to discuss a misconception produce corrections no better than unassisted self-reflection, and undirected LLM conversations can actively entrench misconceptions. Second, recipient receptivity: financial concepts are often foreign to the investors who misapply them, and LLM responses pitched below a participant's financial sophistication are judged as less credible and produce substantially weaker corrections. LLMs thus offer a scalable alternative to traditional financial literacy intervention, but only when designed with both factors in mind.

  PublisherOA PDF

• The Impact of AI: From Financial Investments to Scientific Investigations—A Fireside Chat Between Peter Galison and Andrew W. Lo

  Harvard Data Science Review · 2026-01-30

  articleOpen accessSenior author

  In this wide-ranging conversation conducted at Harvard University’s Inaugural AI Summit in October 2025, MIT’s Andrew W. Lo and Harvard’s Peter Galison examine how AI is transforming scientific research, economic decision-making, and collaborative problem-solving. They trace AI’s evolution from automation toward systems that increasingly resemble partners in thought, enabling new forms of understanding across fields such as physics, finance, biology, and economics. Galison highlights how AI and advanced computation are becoming essential for handling massive data sets, including those used to image black holes, while Lo discusses the challenge of embedding ethical and regulatory norms such as fiduciary duty into AI systems. Together, they explore both the promise of AI in accelerating discovery and the risks of misuse, reflecting on how rapid advances in computation may propel science, reshape the economy, and push us closer to artificial general intelligence.

  PublisherDOI

• Quantifying Gender Bias in Large Language Models: When ChatGPT Becomes a Hiring Manager

  ArXiv.org · 2026-03-10

  articleOpen accessSenior author

  The growing prominence of large language models (LLMs) in daily life has heightened concerns that LLMs exhibit many of the same gender-related biases as their creators. In the context of hiring decisions, we quantify the degree to which LLMs perpetuate societal biases and investigate prompt engineering as a bias mitigation technique. Our findings suggest that for a given resumé, an LLM is more likely to hire a female candidate and perceive them as more qualified, but still recommends lower pay relative to male candidates.

  PublisherOA PDF

• One Size Fits None: Heuristic Collapse in LLM Investment Advice

  ArXiv.org · 2026-04-26

  articleOpen accessSenior author

  Large language models are increasingly deployed as advisors in high-stakes domains -- answering medical questions, interpreting legal documents, recommending financial products -- where good advice requires integrating a user's full context rather than responding to salient surface features. We investigate whether frontier LLMs actually do this, or whether they instead exhibit heuristic collapse: a systematic reduction of complex, multi-factor decisions to a small number of dominant inputs. We study the phenomenon in investment advice, where legal standards explicitly require individualized reasoning over a client's full circumstances. Applying interpretable surrogate models to LLM outputs, we find systematic heuristic collapse: investment allocation decisions are largely determined by self-reported risk tolerance, while other relevant factors contribute minimally. We further find that web search partially attenuates heuristic collapse but does not resolve it. These findings suggest that heuristic collapse is not resolved by web search augmentation or model scale alone, and that deploying LLMs as advisors requires auditing input sensitivity, not just output quality.

  PublisherOA PDF

• DORADO: Dynamic Optimization of R&D Options

  SSRN Electronic Journal · 2026-01-01

  preprintOpen access
  PublisherDOI

• Breaking Bad Financial Habits: How LLM Conversations Correct Financial Misconceptions

  arXiv (Cornell University) · 2026-04-29

  preprintOpen accessSenior author

  Financial misconceptions carry direct economic costs, from panic selling to equity market avoidance, yet they are notoriously resistant to correction. Traditional financial literacy interventions are constrained by cost, reach, and a persistent gap between knowledge and behavioral change. Across three pre-registered studies, we find that purposefully designed LLMs can durably correct financial misconceptions. Critically, two factors are necessary for this effect. First, corrective intent: LLMs prompted only to discuss a misconception produce corrections no better than unassisted self-reflection, and undirected LLM conversations can actively entrench misconceptions. Second, recipient receptivity: financial concepts are often foreign to the investors who misapply them, and LLM responses pitched below a participant's financial sophistication are judged as less credible and produce substantially weaker corrections. LLMs thus offer a scalable alternative to traditional financial literacy intervention, but only when designed with both factors in mind.

  PublisherDOI

• One Size Fits None: Heuristic Collapse in LLM Investment Advice

  arXiv (Cornell University) · 2026-04-26

  preprintOpen accessSenior author

  Large language models are increasingly deployed as advisors in high-stakes domains -- answering medical questions, interpreting legal documents, recommending financial products -- where good advice requires integrating a user's full context rather than responding to salient surface features. We investigate whether frontier LLMs actually do this, or whether they instead exhibit heuristic collapse: a systematic reduction of complex, multi-factor decisions to a small number of dominant inputs. We study the phenomenon in investment advice, where legal standards explicitly require individualized reasoning over a client's full circumstances. Applying interpretable surrogate models to LLM outputs, we find systematic heuristic collapse: investment allocation decisions are largely determined by self-reported risk tolerance, while other relevant factors contribute minimally. We further find that web search partially attenuates heuristic collapse but does not resolve it. These findings suggest that heuristic collapse is not resolved by web search augmentation or model scale alone, and that deploying LLMs as advisors requires auditing input sensitivity, not just output quality.

  PublisherDOI

• Quantifying Gender Bias in Large Language Models: When ChatGPT Becomes a Hiring Manager

  arXiv (Cornell University) · 2026-03-10

  preprintOpen accessSenior author

  The growing prominence of large language models (LLMs) in daily life has heightened concerns that LLMs exhibit many of the same gender-related biases as their creators. In the context of hiring decisions, we quantify the degree to which LLMs perpetuate societal biases and investigate prompt engineering as a bias mitigation technique. Our findings suggest that for a given resumé, an LLM is more likely to hire a female candidate and perceive them as more qualified, but still recommends lower pay relative to male candidates.

  PublisherDOI

• Establishing a commercial solution for extremely rare genetic diseases

  Nature Biotechnology · 2026-03-02

  articleSenior author
  PublisherDOI

• Underperformance of Performance Shares

  SSRN Electronic Journal · 2026-01-01

  preprintOpen access
  PublisherDOI

Frequent coauthors

• Kien Wei Siah

  59 shared

• Jiang Wang

  Taiyuan University of Technology

  57 shared

• Richard T. Thakor

  University of Minnesota

  54 shared

• Chi Heem Wong

  50 shared

• Ruixun Zhang

  43 shared

• Shomesh E. Chaudhuri

  39 shared

• A. Craig MacKinlay

  32 shared

• Mila Getmansky

  27 shared

Awards & honors

• Sloan Fellowship
• Guggenheim Fellowship
• Paul A. Samuelson Award
• Harry M. Markowitz Award
• CFA Institute’s James R. Vertin Award