Ada Lovelace and Charles Babbage first imagined an Analytical Engine that could translate codes to perform tasks useful to humans in the middle of the 19th century. Ever since, people have sought to understand the relationship between humans and computational machines. As these analytical engines have evolved into more sophisticated tools over the last 100 years, our interest in this relationship has grown exponentially, making the leap from the fringes of science fiction to the top of our daily newsfeeds.

Many academic disciplines are devoted to this question, ranging from a focus on how individuals interact with machines (Human Computer Interaction) to looking at the impact of computers on societal systems (Cybernetics and Cyber-Physical Systems). Different forms of Crowd-Machine Interaction explore how groups of people work together with computers to achieve shared goals or maximise collaborative efforts. The literature ranges from the more practice-based approaches of Crowdsourcing and Citizen Science to the more academic Human Computation and Social Computing. These terms are all used to describe the aggregation of diverse inputs from a crowd towards a specific goal, respectively: to open up innovation and ideation, contribute to scientific discoveries, solve computational problems or stimulate online social behaviours.

The history of contributions across the field of Computer Science has nevertheless lacked a comprehensive overview of the potential of AI to scale and enhance collective human efforts, particularly with reference to real-world case studies. Notable exceptions include the survey of field in 2015 by Daniel Weld, which mostly focused on the uses of machine intelligence on crowdsourcing platforms, and the more recent report by New York University’s Governance Lab, Identifying Citizens’ Needs by Combining Artificial Intelligence and Collective Intelligence.