Ben Goertzel of SingularityNET: What AGI Means Now, Open Source AI, and Compute Constraints

This episode is sponsored by Your360 AI. Get 10% off through January 2026 at https://Your360.ai with code: INSIDE. Ben Goertzel, founder and CEO of SingularityNET, joins us to argue for an open path to AGI. We unpack what “open source” should include beyond code, how access to training data, pipelines, and compute shapes who can actually participate, and why “open weights” can still fall short. We also ask who needs convincing, what incentives could move the market, and how decentralized infrastructure fits into his vision for building advanced AI outside a handful of mega labs. Note: Time codes subject to change depending on dynamic ad insertion by the distributor. CHAPTERS: 0:00 - Start 2:13 - How the definition of AGI has shifted over time, and why the term has become distorted 7:58 - The “thousand narrow AIs” future, and who builds the next specialized system 12:40 - Revisiting Goertzel’s past AGI timeline predictions, what changed by late 2025, and is 2029 still feasible for AGI? 18:42 - Whether today’s AI investment is flowing to the right bets (LLMs vs world models and beyond) 25:43 - Why Goertzel argues LLMs cannot reach AGI, and what evidence could change his mind 31:44 - What it would take to build a truly open-source path to AGI 37:18 - Open weights vs real openness: Goertzel’s take on what’s missing from current releases 43:56 - Who needs to be convinced for open AI to succeed, and how that persuasion happens 50:20 - Advice for college students preparing for an AI-shaped career landscape Learn more about your ad choices. Visit megaphone.fm/adchoices