Ben Goertzel: Artificial General Intelligence | Lex Fridman Podcast #103

In this episode of the Lex Fridman Podcast, Ben Goertzel explores the multifaceted landscape of artificial general intelligence with the depth and nuance that has made him a central figure in the AGI community. The conversation begins with philosophical questions about intelligence and consciousness, examining whether intelligent beings might exist all around us in forms we don't recognize. Ben draws on literary influences, particularly Dostoevsky, to discuss the nature of human consciousness and morality.

Reflecting on his Russian heritage and early life, Ben recounts how he fell in love with artificial intelligence through early encounters with computers and philosophical exploration. He shares his perspective on fundamental human questions, including whether humans are inherently good or evil, touching on themes that span from personal philosophy to the development of artificial minds.

The episode then shifts to more concrete discussions about building AGI. Ben outlines the difference between narrow AI and general intelligence, explaining why achieving AGI requires fundamentally different approaches than current deep learning methods. He discusses OpenCog, his ambitious AI framework designed to incorporate multiple paradigms including logic, probability, and neural networks, to create a more comprehensive approach to machine intelligence.

A significant portion of the conversation focuses on SingularityNET, Ben's project to create a decentralized marketplace for AI services. This venture represents his belief that AI development should be distributed rather than concentrated in the hands of a few large corporations. Ben explains how blockchain technology and decentralized mechanisms could democratize access to AI capabilities.

The discussion also covers Sophia the robot, Ben's work as Chief Scientist of Hanson Robotics, and what it means to create machines that can interact naturally with humans. He discusses the limitations and possibilities of current robotics and AI integration.

Beyond technical topics, the conversation explores deeper existential questions. Ben reflects on the nature of life and death, whether immortality would be desirable, and what gives life meaning. These discussions connect to broader concerns about humanity's future, including the possibility of colonizing Mars and navigating existential risks.

Throughout the episode, Ben demonstrates a unique ability to move fluidly between technical AI discussions and philosophical inquiry. He speaks to the importance of the AGI community maintaining focus on safety and beneficial development, rather than pursuing capabilities for their own sake. The conversation concludes with Ben considering what questions he would ask an AGI system, reflecting his ongoing curiosity about how artificial minds might perceive reality.