Building The Collective Intelligence Of Humans And Machines

Bonisiwe Shabane

-Nov 29, 2025, 7:58 PM

building the collective intelligence of humans and machines

Midlevel leaders are at the heart of every major shift in a business. See how… Learn how Harvard Business Impact shape the best minds in leadership, continuously raising the bar… Midlevel leaders are under more pressure than ever. They’re expected to deliver today and drive… The most successful digital transformation strategies rely on constant coordination between people and technology.

We explore the use of aggregative crowdsourced forecasting (ACF) [2, 42] as a mechanism to help operationalize “collective intelligence” of human-machine teams for coordinated actions. We adopt the definition for Collective Intelligence as: “A property of groups that emerges from synergies among data-information-knowledge, software-hardware, and individuals (those with new insights as well as recognized authorities) that enables just-in-time knowledge... Aggregative crowdsourced forecasting (ACF) is a recent key advancement towards Collective Intelligence wherein predictions (X% probability that Y will happen) and rationales (why I believe it is this probability that X will happen) are... This research asks whether ACF, as a key way to enable Operational Collective Intelligence, could be brought to bear on operational scenarios (i.e., sequences of events with defined agents, components, and interactions) and decision-making,... Collective Intelligence (CI) emerges from new ways of connecting humans and machines to enable decision-advantage, at least in part, by creating and leveraging additional sources of information that decision-makers might otherwise not include [1,... Operational scenarios present a unique challenge for typical collective intelligence approaches.

The wisdom of the crowd, the archetypal collective intelligence, works in classic applications, like predicting the weight of a cow, because the probability distributions of responses in these instances have median estimates centered near... Typical operational scenarios necessitate a different approach because the “crowd” is often heavily biased. For example, in a military setting operators at various levels are privileged to unique knowledge, some of which may alter their predictions in an adversarial reasoning scenario. Moreover, operational scenarios increasingly rely on input from machine intelligence, which decision-makers must integrate with human judgments. Aggregative crowdsourced forecasting (ACF) is a recent, pivotal advancement in CI methods. ACF helps decision-makers from commercial companies, government organizations, and militaries overcome the intractable problem of individual bias by collecting many predictions and integrating the best of machine intelligence into the decision process.

To date, ACF has been used for risk management at a strategic level, focusing on tasks like informing policy, anticipating instability, balancing research portfolios, identifying emerging technology, and serving as an early warning mechanism... This document provides a preliminary exploration and evaluation of the potential use of aggregative crowdsourced forecasting as a mechanism to help operationalize “collective intelligence” for operational scenarios. After operationalizing pertinent terminology and laying out the critical research questions, we identify gaps in the research and propose ways of empirically testing solutions to the open questions. The insight that a group of forecasters may be more reliable than a single individual, colloquially known as the wisdom of a crowd, is far from new. It is in Aristotle’s Politics [30] and was famously championed by the British polymath Sir Francis Galton. (Both use the accuracy achieved by statistical aggregation of forecasters as motivation for democracy; Aristotle waxed poetic about art judgments while Galton, more practically, used the statistical accuracy of a crowd’s guesses of the...

Although digital technology facilitates aggregation of and eases access to crowd wisdom [27], it can also undermine it because crowd wisdom hinges on natural variation in the forecasters’ information. Individual forecasters who use digital technology to change their knowledge of other forecasters’ knowledge may become biased and reflect their new knowledge in their forecasts, ultimately undermining the crowd’s wisdom [37]. Studying the process of crowd decision-making and how the crowd’s wisdom emerges may facilitate mitigation, if not reversal, of digital technology’s undermining effects on the crowd wisdom [32, 52]. Collective intelligence (CI) is a “form of universally distributed intelligence, constantly enhanced, coordinated in real-time, and resulting in the effective mobilization of skills” [32]. We are primarily interested in collective intelligence that emerges from new ways of connecting humans and AI to enable decision advantage in instances of adversarial reasoning. Decision advantages are made possible by connections that allow the collective to create and leverage information sources that decision-makers might otherwise not include in a decision process.

Adversarial reasoning involves “determining the states, intents, and actions of one’s adversary, in an environment where one strives to effectively counter the adversary’s actions.” [28] Images for download on the MIT News office website are made available to non-commercial entities, press and the general public under a Creative Commons Attribution Non-Commercial No Derivatives license. You may not alter the images provided, other than to crop them to size. A credit line must be used when reproducing images; if one is not provided below, credit the images to "MIT." What can we learn about human intelligence by studying how machines “think?” Can we better understand ourselves if we better understand the artificial intelligence systems that are becoming a more significant part of our... These questions may be deeply philosophical, but for Phillip Isola, finding the answers is as much about computation as it is about cogitation.

Isola, the newly tenured associate professor in the Department of Electrical Engineering and Computer Science (EECS), studies the fundamental mechanisms involved in human-like intelligence from a computational perspective. While understanding intelligence is the overarching goal, his work focuses mainly on computer vision and machine learning. Isola is particularly interested in exploring how intelligence emerges in AI models, how these models learn to represent the world around them, and what their “brains” share with the brains of their human creators. We are on the cusp of a new wave of hybrid work where organizations won’t just mix in-person and remote workers—they’ll pair humans and AI agents as co-workers. These AI agents will have the ability to take and act on decisions independently and will not be reliant on detailed user inputs, as today’s mainstream GenAI tools are. For example, they will be capable of interpreting context, adapting dynamically to new information, independently ideating, and even partnering with human colleagues to tackle complex and varied tasks.

AI agents are set to go beyond simply augmenting humans to being true co-workers alongside us. By combining human and AI capabilities, these hybrid teams promise to create new possibilities to deliver competitive advantage far beyond incremental productivity gains. This coming shift also demands thoughtful leadership to balance human workers and AI technologies to ensure the unique strengths of each are maximized. In large global organizations, many workers already find themselves collaborating through Slack or Microsoft Teams with colleagues they have never spoken to, let alone met in-person. Even with close colleagues, these real-time digitalinteractions often outnumber face-to-face meetings. Today, there is another human at the other end of those interactions, providing their expertise or performing a specific task.

While many workers have already begun incorporating GenAI tools, like ChatGPT, to help with targeted analyses and tasks, the increasing maturity of AI will take this relationship a crucial step further: rather than being... This emerging hybrid workforce has been made possible by advances in the natural language processing of large language models (LLMs) that enable humans to communicate with AI agents in the same way they would... The reasoning capabilities of LLMs allow natural language instructions to be translated into action without the need for prescriptive code or detailed instructions, or even well-defined steps. Inputs can be more notional, and the AI coworker can still develop and execute a plan, coming back for feedback as needed. In many ways, the interactions of humans and AI colleagues will be analogous to human passengers in self-driving cars. The cars require a destination, but not specific instructions on when to brake or accelerate.

Self-driving cars plot a course, but also receive new data about their surroundings, processing it to plan and execute actions. AI coworkers will be able to act similarly: interpreting context, interacting with other tools and external systems to develop a plan, and even making certain decisions autonomously. They will also maintain task memory so they can learn and improve on the jobs they do regularly. Jacob Taylor, Thomas Kehler, Sandy Pentland, Martin Reeves Janice C. Eberly, Molly Kinder, Dimitris Papanikolaou, Lawrence D.

W. Schmidt, Jón Steinsson Rosanne Haggerty, Ruby Bolaria Shifrin, Jacob Taylor, Kershlin Krishna, Sara Bronin, Nick Cain, Xiomara Cisneros, Adam Ruege, Henri Hammond-Paul, Jamie Rife, Josh Humphries, Beth Noveck In recent years, we have experienced rapid development of advanced technology, machine learning, and artificial intelligence (AI), intended to interact with and augment the abilities of humans in practically every area of life. With the rapid growth of new capabilities, such as those enabled by generative AI (e.g., ChatGPT), AI is increasingly at the center of human communication and collaboration, resulting in a growing recognition of the... However, there are many unanswered questions regarding how human-AI collective intelligence will emerge and what the barriers might be.

Truly integrated collaboration between humans and intelligent agents may result in a different way of working that looks nothing like what we know now, and it is important to keep the essential goal of... In this special issue, we begin to scope out the underpinnings of a socio-cognitive architecture for Collective HUman-MAchine INtelligence (COHUMAIN), which is the study of the capability of an integrated human and machine (i.e.,... This topic consists of nine papers including a description of the conceptual foundation for a socio-cognitive architecture for COHUMAIN, empirical tests of some aspects of this architecture, research on proposed representations of intelligent agents... Keywords: Artificial intelligence; Collaboration; Collective intelligence; Human–AI; Human–machine teaming. Midlevel leaders are at the heart of every major shift in a business. See how…

Learn how Harvard Business Impact shape the best minds in leadership, continuously raising the bar… Midlevel leaders are under more pressure than ever. They’re expected to deliver today and drive… The most successful digital transformation strategies rely on constant coordination between people and technology.

Building The Collective Intelligence Of Humans And Machines

People Also Search

Midlevel Leaders Are At The Heart Of Every Major Shift

We Explore The Use Of Aggregative Crowdsourced Forecasting (ACF) [2,

The Wisdom Of The Crowd, The Archetypal Collective Intelligence, Works

To Date, ACF Has Been Used For Risk Management At

Although Digital Technology Facilitates Aggregation Of And Eases Access To