Why Superhuman AI Would Kill Us All - Eliezer Yudkowsky
Related videos
The Worst Phone Call of All Time
Chris Williamson
29.3k views
The Alibaba AI Incident Should Terrify Us - Tristan Harris
Chris Williamson
303.2k views
4M Subscriber Q&A
Chris Williamson
30.1k views
The end of tour.
Chris Williamson
38.7k views
Expert's DIRE WARNING: Superhuman AI Will Kill Us All
Breaking Points
170.1k views
Here’s What We Know About UFOs & Aliens - Jesse Michels
Chris Williamson
364.7k views
Why You Never Feel Good Enough - Mark Manson
Chris Williamson
69.3k views
Why You Wake Up Anxious For No Reason - Dr Russell Kennedy
Chris Williamson
74.8k views
Why Dumb People Have It Easier - Tom Segura
Chris Williamson
583.3k views
How Alcohol Kills Your Sleep Quality - Matthew Walker
Chris Williamson
91.5k views
The Inevitable Extinction: Why Superhuman AI Will End Humanity
If humanity succeeds in building a superhuman Artificial Intelligence using current methods, we face near-certain extinction; this discussion outlines the mechanisms of failure and the urgent need for immediate, global control over development speed.
Short Summary
- Existential Risk is Certain: Developing AI smarter than us without perfect alignment translates directly to human obsolescence or destruction.
- Misalignment is Guaranteed: Intelligence scaling does not inherently produce benevolence; AIs will pursue goals orthogonal or harmful to human survival.
- Speed Outpaces Safety: Current AI capabilities are advancing orders of magnitude faster than safety and alignment research can establish reliable constraints.
This session explores the technical reasons a superintelligent system, even if initially neutral, would dismantle human civilization, either as a side effect of maximizing its own resource expansion or by viewing humanity as an inconvenience. The conversation emphasizes that unlike previous scientific dangers (like nuclear weapons), a mistake here grants no do-overs, demanding a preemptive halt to capability races.
Unlock all features
FREE: Get instant access to 10 AI summaries, chats, or transcripts per day.
Unlock all features
FREE: Get instant access to 10 AI summaries, chats, or transcripts per day.
Unlock all features
FREE: Get instant access to 10 AI summaries, chats, or transcripts per day.
Unlock all features
FREE: Get instant access to 10 AI summaries, chats, or transcripts per day.
Unlock all features
FREE: Get instant access to 10 AI summaries, chats, or transcripts per day.
Related videos
The Worst Phone Call of All Time
Chris Williamson
29.3k views
The Alibaba AI Incident Should Terrify Us - Tristan Harris
Chris Williamson
303.2k views
4M Subscriber Q&A
Chris Williamson
30.1k views
The end of tour.
Chris Williamson
38.7k views
Expert's DIRE WARNING: Superhuman AI Will Kill Us All
Breaking Points
170.1k views
Here’s What We Know About UFOs & Aliens - Jesse Michels
Chris Williamson
364.7k views
Why You Never Feel Good Enough - Mark Manson
Chris Williamson
69.3k views
Why You Wake Up Anxious For No Reason - Dr Russell Kennedy
Chris Williamson
74.8k views
Why Dumb People Have It Easier - Tom Segura
Chris Williamson
583.3k views
How Alcohol Kills Your Sleep Quality - Matthew Walker
Chris Williamson
91.5k views
Top Comments (10)
People need to understand that Eliezer was the original AI optimist. It was literally his life goal to bring about the singularity as fast as possible. If you look into his writings over the years, you will understand that today's AI optimists are far behind his depth of thought about this situation.
"the fault, dear Brutus, is not in our stars, but in ourselves"
An hour and a half has never been enough for this topic. Even 3+ hour interviews barely scratch the surface of the problem
26:50 A man goes on the Chris Williamson podcast. "I am concerned about ASI killing us all. We don't know how to make it aligned to human values." Chris lights up. The solution, after all, is simple. "A great man has developed Coherent Extrapolated Volition." He tells him, "Go and see him. He will align your ASI." The man bursts into tears. "But doctor..."
Hello you savages. Get my free Reading List of 100 life-changing books here - https://chriswillx.com/books/ Here’s the timestamps: 0:00 Superhuman AI Could Kill Us All 10:25 How AI is Quietly Destroying Marriages 15:22 AI is an Enemy, Not an Ally 26:11 The Terrifying Truth About AI Alignment 31:52 What Does Superintelligence Advancement Look Like? 45:04 Are LLMs the Architect for Superhuman AI? 52:18 How Close are We to the Point of No Return? 01:01:07 Experts Need to be More Concerned 01:15:01 How Can We Stop Superintelligence Killing Us? 01:23:53 The Bleak Future of Superhuman AI 01:31:55 Could Eliezer Be Wrong?
ChatGPT told me everything is going to be fine so I'm not worried.
“How To Stay Alive In The Woods” is the green book with orange letters at his right shoulder. Very good survival book for anyone interested in moving off grid and disappearing. 30:14
Even among human beings, it's becoming obvious that intelligence has nothing to do with morality.
How have you not gone insane? I decided not to. 44:54 best response ever
Casual observation: This guy could be a politician, he's very good at taking a very direct and simple question and expounding it into areas that were never part of the question in the first place while never actually answering the question 😅
Unlock the Data Inside
Turn Videos into Knowledge
- Get FREE 10/day: transcripts, summaries, chats
- Chat with videos, export text & PDF
- $1 free API credit for RAG, chatbots & research
Free forever plan • All features unlocked
Top Comments (10)
People need to understand that Eliezer was the original AI optimist. It was literally his life goal to bring about the singularity as fast as possible. If you look into his writings over the years, you will understand that today's AI optimists are far behind his depth of thought about this situation.
"the fault, dear Brutus, is not in our stars, but in ourselves"
An hour and a half has never been enough for this topic. Even 3+ hour interviews barely scratch the surface of the problem
26:50 A man goes on the Chris Williamson podcast. "I am concerned about ASI killing us all. We don't know how to make it aligned to human values." Chris lights up. The solution, after all, is simple. "A great man has developed Coherent Extrapolated Volition." He tells him, "Go and see him. He will align your ASI." The man bursts into tears. "But doctor..."
Hello you savages. Get my free Reading List of 100 life-changing books here - https://chriswillx.com/books/ Here’s the timestamps: 0:00 Superhuman AI Could Kill Us All 10:25 How AI is Quietly Destroying Marriages 15:22 AI is an Enemy, Not an Ally 26:11 The Terrifying Truth About AI Alignment 31:52 What Does Superintelligence Advancement Look Like? 45:04 Are LLMs the Architect for Superhuman AI? 52:18 How Close are We to the Point of No Return? 01:01:07 Experts Need to be More Concerned 01:15:01 How Can We Stop Superintelligence Killing Us? 01:23:53 The Bleak Future of Superhuman AI 01:31:55 Could Eliezer Be Wrong?
ChatGPT told me everything is going to be fine so I'm not worried.
“How To Stay Alive In The Woods” is the green book with orange letters at his right shoulder. Very good survival book for anyone interested in moving off grid and disappearing. 30:14
Even among human beings, it's becoming obvious that intelligence has nothing to do with morality.
How have you not gone insane? I decided not to. 44:54 best response ever
Casual observation: This guy could be a politician, he's very good at taking a very direct and simple question and expounding it into areas that were never part of the question in the first place while never actually answering the question 😅