158 - Guest: Louis Rosenberg, Human/AI Hybrid Intelligence Expert, part 1
Jun 26, 2023
auto_awesome
Louis Rosenberg, Human/AI Hybrid Intelligence Expert, discusses the concept of hybrid swarm intelligence and its potential applications. We explore AI-powered collective decision-making in fields like sports prediction and medical diagnosis. The use of AI in assisting radiologists and swarm intelligence in political polling is also highlighted. The hosts celebrate their podcast's third year and discuss news about robots writing their own code.
Swarm intelligence allows human groups to make better decisions using real-time AI algorithms, resulting in more accurate forecasts and solutions.
SWARM technology has been successfully employed in various domains, such as sports prediction, medical diagnosis, and financial trading, enhancing accuracy and decision-making.
Deep dives
AI and Swarm Intelligence
Louis Rosenberg discusses the concept of swarm intelligence, inspired by nature's ability to make decisions for the benefit of the whole population. He explains that swarms, such as fish schools and bird flocks, become significantly smarter together than alone. Drawing on this idea, Rosenberg developed a system called SWARM that allows human groups to make better decisions using real-time AI algorithms. Numerous studies have shown that when individuals collaboratively interact in the system, they converge on answers that are more accurate than individual predictions. This real-time process amplifies commonality and facilitates the discovery of solutions that best align with the collective sentiments of the group.
Applying SWARM to Various Fields
Rosenberg shares how SWARM technology has been used in different domains. For example, groups of sports fans using SWARM technology predicted the outcomes of English football matches with 70% accuracy. Similarly, doctors working together as a swarm reduced diagnostic errors by 26%. Financial traders improved their accuracy by 30% when collaborating in a SWARM. These examples demonstrate that when human groups connect in real time and AI is used to infer their sentiments, the SWARM system combines their expertise and convictions to achieve more accurate forecasts and decisions.
The Power of Real-Time Interaction
Rosenberg explains the real-time nature of SWARM interactions, where individuals continuously adjust their behavior to convey their levels of conviction. The AI algorithms observe these behaviors and guide the SWARM towards a collective answer. As the SWARM moves, participants change their behavior, influencing the AI's understanding of their sentiments. This dynamic process allows groups to find solutions that may not be anyone's first choice but best align with the collective sentiment. The SWARM's ability to amplify commonality and focus on solutions that benefit the whole distinguishes it from traditional polling or voting methods. It encourages convergence and minimizes polarization, enabling groups to make better collective decisions.
What do honeybees have to teach us about AI? You'll find out from Louis Rosenberg on this episode. He's been working in AR and VR starting over 30 years ago at Stanford and NASA. 1992 he developed the first mixed reality system at Air Force Research Laboratory. In 2004 he founded the early AR company Outland Research which was acquired by Google in 2011. He received a PhD from Stanford, was a tenured professor at California State University, and has been awarded over 300 patents.
He's currently CEO and Chief Scientist of Unanimous AI, a company that amplifies human group intelligence using AI technology based on the biological principle of Swarm Intelligence, which is where the bees come in. The UNU Swarm Intelligence that he created has an extraordinary record of making predictions like Oscar winners.
We talk about how AI can help humans cooperate instead of conflict, and we talk about threats to privacy, and the convergence of AI and AR/VR technology like Apple's new headset.
All this plus our usual look at today's AI headlines.