Abstract

Inspired by the increasing use of artificial intelligence (AI) to augment humans, researchers have studied human–AI systems involving different tasks, systems, and populations. Despite such a large body of work, we lack a broad conceptual understanding of when combinations of humans and AI are better than either alone. 

In this talk, Michelle will present findings from a preregistered systematic review and meta-analysis of 106 experimental studies reporting 370 effect sizes. . The research reveals unexpected patterns about when human-AI combinations succeed versus fail, how task type fundamentally shapes collaboration outcomes, and why relative performance between humans and AI alone matters more than previously understood.

These findings challenge common assumptions about human-AI collaboration and point to promising new directions for designing systems where humans and AI can truly work better together.

Speaker Bio

Michelle Vaccaro is a PhD candidate in MIT’s Institute for Data, Systems, and Society, where she has been generously funded by the Bose Fellowship and Accenture Fellowship. She studies human-AI interaction and investigates how people and AI should—and should not—work together in organizations. To this end, her research focuses on (i) identifying the conditions for human-AI synergy, when human-AI combinations perform better than both humans and AI alone and (ii) testing and extending existing theories about human behavior to AI and human-AI contexts. Before coming to MIT, Michelle worked at Goldman Sachs in foreign exchange strategy and structuring. She earned her Bachelor’s degree in computer science from Harvard College, where she graduated summa cum laude with highest departmental honors and received the Thomas T. Hoopes Prize for her thesis.