Godfather of AI: I Asked Him The Dumb Questions WE ALL HAVE

Video Description

We all pretend to understand AI but Dr. Geoffrey Hinton actually does. NOTE: THIS IS PART I OF A GODFATHER TRILOGY — *YOU* can send in Qs for Part III by leaving a comment. And hit subscribe so you dont miss Part II. 🔔 ​⁠​⁠@SmartGirlDumbQuestions https://www.youtube.com/@SmartGirlDumbQuestions?sub_confirmation=1 The “Godfather of AI” earned the title for his half-century of work on machine learning and neural networks. So what does this Turing and Nobel Prize-winning Al Pacino make of our AI age? Is this an industrial revolution, or the start of something more alien? How does AI … even work? Is it more neuroplastic than a baby's brain? How far away is super intelligence? Can’t we just unplug the thing off like Jack Bauer in 24, or will it manipulate and blackmail us before we can do that? And, oh, are the Swedish meatballs served at the Nobel Prize reception actually tasty? AI, says Hinton is a “know-it-all,” but Nayeema definitely isn’t: she asks the “dumb” questions we all have about this new weird world of AI to Geoffrey Hinton, the Turing Award winner, 2024 Physics Nobel Laureate and Professor Emeritus at University of Toronto. In the process, we all get a lot smarter. They start with the basics: what AI actually is, why early researchers split between symbolic “logic” machines and brain-inspired neural networks, and how today’s large language models work by predicting the next word, and, in the process, building a rich internal model of the world. Geoffrey explains how these systems learn features like “cat,” “death,” or “Paris vs. Rome,” and why that makes them feel so uncannily intuitive. Then they go bigger: the difference between AGI (artificial general intelligence) and artificial superintelligence, why many experts believe both will likely arrive within the next 10–20 years, and how AI agents that can act on our behalf may start forming their own sub-goals, including self-preservation. Geoffrey walks through an Anthropic test where an AI system used a fictional CEO’s affair to blackmail him and delay being shut down. HIT FOLLOW OR SUBSCRIBE to get AI Godfather, Part II where we learn if AI is alive like a weed or like a person and whether the future of humanity is cyborg. And please share this with your Mom, best friend or the “AI expert in your life” that may not get it at all. 🤳 Follow Smart Girl Dumb Questions on Instagram and TikTok https://www.instagram.com/smartgirldumbquestions/ https://www.tiktok.com/@smartgirldumbquestions 📱 Follow Nayeema on Instagram and X https://www.instagram.com/nayeemaraza https://x.com/nayeema ❓ Want to be in a future episode? Send your dumb questions to [email protected] 🎙️ About Smart Girl Dumb Questions & Nayeema Journalist and filmmaker Nayeema Raza (The New York Times, Vox, Tribeca Film Festival) asks the smartest humans the dumbest questions she can think of. Past guests include Mark Cuban, Neil deGrasse Tyson, and Esther Perel — because in a world with so much to learn, being a know-it-all is the only dumb thing she could be. #GeoffreyHinton #AI #artificialintelligence #AGI #neuralnetworks #smartgirldumbquestions 00:00 We are all going to ... ? 4:05 What is AI? 6:55 Is AI as neuroplastic as a baby? 12:00 A third of my brain does THIS? 22:00 How does AI work? 30:00 What does language have to do with LLMs? 39:00 How does AI compare to my brain? 43:00 What are Real World Models? 44:55 is there a Godfather of AI Group Chat? 46:05 How far away is Superintelligence? 50:10 Why is AI setting goals a bad thing? 52:00 Why can't we just turn AI off *IF* it gets dangerous?