Join our daily and weekly newsletters for the latest updates and exclusive content on industry-leading AI coverage.
Ever imagined our brains linked together into a global “hive mind”? This idea has been around since Olaf Stapledon’s 1930 novel, “Last and First Men,” where humans evolve to communicate telepathically and form intelligent collective minds. In reality, this goal is called Collective Superintelligence, and it doesn’t rely on telepathy. Instead, it uses new generative AI technologies to connect large groups of people in real-time, helping us solve tough problems using our combined knowledge, wisdom, and insights.
This has been my research focus for the past decade. I believe it can create superintelligent systems that prioritize human values, morals, and interests. While it might seem strange or unnerving to think of large groups making decisions together, nature shows us it’s possible and effective. Many social species, like fish, bees, and birds, use what biologists call Swarm Intelligence. This allows them to make rapid and smart decisions that surpass the capabilities of any single member.
Fish schools are particularly fascinating. They can make complex decisions without a leader, relying on each fish’s limited knowledge and a special organ called the lateral line. This organ helps them sense the speed and direction of their neighbors, enabling fast and effective collective movement even when most fish are unaware of threats around them.
This natural ability to make quick, effective decisions as a group is something humans can benefit from, especially as our problems are more complex and multifaceted. Traditionally, effective human conversations work best in small groups of four to seven people. Larger groups often lose coherent debate and turn into disjointed monologues. But recent advancements in generative AI have provided new solutions.
Enter Conversational Swarm Intelligence (CSI). This technology allows groups, regardless of size, to discuss complex issues in real-time and quickly converge on intelligent solutions. It starts by breaking a large group into small subgroups, each engaging in their own dialog. The key innovation is the use of AI agents that act as “conversational surrogates,” sharing insights across these subgroups. This ensures information spreads and integrates seamlessly, much like fish schools.
Studies support the effectiveness of CSI. For instance, a 2023 study at Carnegie Mellon University found that groups using CSI had more coherent and productive conversations, with individuals contributing significantly more. Another study in 2024 showed that a group of people with average IQs of 100 could achieve a collective IQ of 128 using the CSI platform Thinkscape.
While these studies used text, CSI can be applied to teleconferencing or VR meetings, enabling large groups to communicate effectively and solve problems with enhanced group intelligence. This could revolutionize fields like enterprise collaboration, market research, and civic engagement.
In the long run, CSI could help build superintelligent systems aligned with human values and wisdom, enabling global collaboration to solve our toughest challenges. This approach, blending human and AI capabilities, offers a safer path than relying solely on artificial superintelligence, ensuring that human interests remain central.
Louis Rosenberg is a researcher in AI, collective intelligence, and mixed reality, and the CEO and chief scientist of Unanimous AI. He encourages experts and data enthusiasts to join the VentureBeat community at DataDecisionMakers, where cutting-edge ideas and data insights are shared. You might even consider contributing your own articles!