Key Areas of Research
Three Strategic Bets on AI’s Future
This paper examines competition in the consumer AI assistant market using worldwide iOS and Android app-store data from seven major AI assistants from May 2023 through December 2025. Rather than finding a winner-take-all market, we show that major product launches tend to coincide with growth in the overall category, with little evidence of direct cannibalization across leading models. In other words, the “AI war” appears less zero-sum than commonly assumed.
The analysis identifies three distinct strategic positions that currently appear viable to date. ChatGPT is pursuing scale, with by far the largest market share and substantial revenue generated from a very large user base. Google Gemini is pursuing an ecosystem defense strategy, using broad distribution and low direct monetization to support Google’s wider platform. Claude is pursuing a differentiated premium niche strategy, with a much smaller user base but much higher revenue per user.
There are three practical takeaways for managers and investors. First, firms should not assume that AI markets will necessarily converge to a single dominant winner. Second, the right AI strategy depends on structural advantage: scale, ecosystem leverage, or premium differentiation. Third, as market growth slows and capital becomes less abundant, each strategy will face different risks, making monetization quality and strategic fit more important than raw user growth alone.
Maxime C. Cohen, Professor, Desautels Faculty of Management, McGill University
Eddy Hage-Youssef, McGill University
Daniel M. McCarthy, Associate Professor of Marketing, Robert H. Smith School of Business, University of Maryland, College Park
D. Daniel Sokol, Professor, USC Gould School of Law; USC Marshall School of Business
The Public Pension Crisis: Contractual Rights and Constitutional Limits
Cambridge University Press, March 2026
A timely response to the pressing issue of public pension reform, The Public Pension Crisis explores the complex relationship between contract law and government pensions, specifically focusing on the Contract Clause and related state Pension Clauses. Analyzing over a decade of litigation, the book highlights the evolving role of pension contracts in constitutional law and examines more than 70 landmark cases to establish a clear, principled framework for determining when pension benefits qualify as contractual obligations. T. Leigh Anenson presents a unified theory to consistently treat public and private pensions, balancing the interests of employees’ earned benefits with the financial challenges facing governments. Combining legal scholarship with practical policy insights, Anenson not only provides a much-needed legal perspective on pension reform but also calls for a systematic approach to addressing the retirement security crisis.
T. Leigh Anenson, J.D., LL.M., Ph.D.
Setting Higher Referral Targets Increases the Number of Women Recommended: Evidence From the Field and Lab
Journal of Applied Psychology
Women continue to be underrepresented in numerous occupations and in the highest echelons of many organizations. This may be due, in part, to disadvantages they face in referral-based hiring and promotion processes. We propose a low-cost and easily scalable intervention to boost referrals of women in male-dominated contexts: requesting a greater target number of referrals (e.g., at least four instead of at least two referrals). Across six experiments (including two field experiments embedded in an organization’s referrals process), requesting double the number of referrals increased the number of women referred by 17%-88%. Our intervention provides a versatile, low-cost, and low-risk option for managers and leaders looking to recruit from the full range of talent available to them.
Aneesh Rai (Assistant Professor, University of Maryland, College Park); Erika Kirgios (Assistant Professor, University of Chicago); Brian Lucas (Associate Professor, Cornell University); Katherine Milkman (Professor, University of Pennsylvania)
The Caring Machine: Feeling AI for Customer Care
Journal of Marketing
Feeling skills have always been a human domain, but AI is advancing rapidly. This article shows how AI can be used to develop feeling intelligence and build empathetic customer relationships.
Ming-Hui Huang & Roland T. Rust
Evolution of Ride Services: From Ride- Hailing to Autonomous Vehicles
Management Science
In recent years the ride service industry has been evolving rapidly, driven by disruptive technologies such as mobile apps, AI, and autonomous vehicles (AVs). While platform-based decentralized ride hailing companies have gained significant market share, vertically-integrated robotaxi services using emerging AVs are starting to enter the market. In this paper, we aim to provide insights about the evolution and the future of ride services studying these two competing business approaches. We find that in many cases in larger markets the ride-hailing firm surprisingly gains the upper hand in competition, having higher market share and profits as well as lower service delays and higher prices, even if it has a cost disadvantage. Further, entry of the AV firm into a market with a dominant ride-hailing firm may reduce total vehicle supply and increase customer wait costs. We also find that when customers are impatient, the entry of a high cost AV firm may lead to a decrease in social welfare despite introducing competition, suggesting that regulators should be careful about introduction of robotaxi services in a market if they are not sufficiently cost efficient. From a broader perspective, our results demonstrate that platform business models in general may have significant strategic advantages over firms with traditional vertically-integrated models under competition, and platforms' dominance in a market may even result in welfare gains.
Daehoon Noh (Korea University), Tunay I. Tunca (UMD, Smith), Yi Xu (UMD, Smith)
Stochastic Gradients: Optimization, Simulation, Randomization, and Sensitivity Analysis
IISE Transactions, February 2026
Big data and high-dimensional optimization problems in operations research (OR) and artificial intelligence (AI) have brought stochastic gradients to the forefront. This article provides a view of research and applications in stochastic gradient estimation from multiple perspectives, as seminal advances have come from diverse and disparate research fields, including operations research/management science (OR/MS), industrial/systems engineering (ISE), optimal/stochastic control, statistics, and more recently from the computer science (CS) AI machine learning (ML) community.
Michael C. Fu (University of Maryland), Jiaqiao Hu (Stony Brook University, and Katya Scheinberg (Georgia Institute of Technology)
Simulation Optimization and Artificial Intelligence
Foundations and Trends in Optimization, December 2025
With the relentless increase in computing power and the ubiquitous availability of data in many industries, the fields of simulation optimization and artificial intelligence have emerged at the scientific and engineering forefront in their societal impact, manifested in the pervasiveness of technologies such as large language models, chatbots, digital twins, and agent-based systems. We examine cross-fertilization between simulation optimization and artificial intelligence, with a particular focus on reinforcement learning, highlighting research that has been mutually beneficial. This work presents examples of the synergies between the fields, followed by real-world applications and case studies, including some futurist and forward-looking concepts.
Yijie Peng (Peking University), Chun-Hung Chen (George Mason University, and Michael C. Fu (University of Maryland)
Online Learning with Survival Data
February 2026
Decision-makers frequently use adaptive experiments to optimize time-to-event outcomes, such as accelerating healthcare screenings among patients who are not up to date or delaying customer churn. A common choice to run these adaptive experiments is a multi-armed bandit with a dichotomized outcome -- an experimenter sets some threshold (e.g. 1 month) and then uses the algorithm to identify the intervention with better performance on the dichotomized outcome (e.g. which algorithm maximizes the proportion of participants who get up to date on screening within a month of outreach). We introduce "survival bandits," a principled class of algorithms that integrate the Cox proportional hazards model to better learn from time-to-event outcomes. Both theoretically and numerically (with a case study on cervical cancer screening), we show that these new algorithms have the potential to greatly improve adaptive experimentation for decision makers across industries who seek to speed or slow an event of interest.
Arielle Anderer (Assistant Prof, Cornell), Hamsa Bastani (Associate Prof, UPenn Wharton), John Silberholz (Assistant Prof, UMD Smith)
Holding Horizon: A New Measure of Active Investment Management
Journal of Financial and Quantitative Analysis
This article introduces a new holding horizon measure of active management and examines its relation to future risk-adjusted fund performance (alpha). Our measure reveals a wide cross-sectional dispersion in mutual fund investment horizons, and shows that long-horizon funds exhibit positive future long-term alphas by holding stocks with superior long-term fundamentals. Further, stocks largely held by long-horizon funds outperform stocks largely held by short-horizon funds by more than 3%annually, adjusted for risk, over the following 5-year period. We also find a clientele effect: to reduce liquidity costs, long-horizon funds attract more long-term investors through share classes that carry load fees.
Chunhua Lan (University of New Brunswick), Fabio Moneta (Queen's University), and Russ Wermers
Backfiring AI? AI Deployment in Workplace
Management Science
AI in the workplace has the potential to change the competitive dynamics among employees. The AI system can learn from high-performing employees and make that knowledge available to others. In a competitive environment, this can disincentivize high-performing employees and ultimately backfire, leading to a decline in overall firm productivity. Our results suggest that some ostensibly simple solutions, such as guaranteeing or increasing wages for adversely affected employees, may not effectively solve the problem, and firms would have to judiciously choose optimal AI efficacy levels to achieve better outcomes.
Di Yuan (Assistant Professor, Auburn University), Manmohan Aseri (Assistant professor, University of Maryland), Narayan Ramasubbu (Professor, University of Pittsburgh)