Quantilizers: AI That Doesn't Try Too Hard Published 2020-12-13 Download video MP4 360p Recommendations 10:22 AI That Doesn't Try Too Hard - Maximizers and Satisficers 23:24 The OTHER AI Alignment Problem: Mesa-Optimizers and Inner Alignment 19:29 Glitch Tokens - Computerphile 10:20 Deceptive Misaligned Mesa-Optimisers? It's More Likely Than You Think... 27:48 AI & Logical Induction - Computerphile 13:41 Is AI Safety a Pascal's Mugging? 09:41 No, it's not Sentient - Computerphile 11:32 There's No Rule That Says We'll Make It 14:30 Moravec's Paradox - Why are machines so smart, yet so dumb? 09:24 Why Does AI Lie, and What Can We Do About It? 11:47 We Were Right! Real Inner Misalignment 34:35 Why You're Probably Not a Simulation 08:16 Why Asimov's Laws of Robotics Don't Work - Computerphile 17:52 Training AI Without Writing A Reward Function, with Reward Modelling 10:41 What can AGI do? I/O and Speed 13:03 Intelligence and Stupidity: The Orthogonality Thesis 08:30 Deadly Truth of General AI? - Computerphile 14:36 Popular Technologies that Won't be Around Much Longer... 19:33 The Experiment that Proved Einstein Was Wrong | Quantum Eraser 07:23 AI Safety Gridworlds Similar videos 02:39 Respectability 11:32 How to Keep Improving When You're Better Than Any Teacher - Iterated Distillation and Amplification 18:05 Intro to AI Safety, Remastered 03:11 Where Do We Go Now 03:20 Why Do I Avoid Sci-fi? 26:17 Paul Christiano AI, Zelda Treachous Turn, Quantilizers (Walkthrough) 11:44 Sharing the Benefits of AI: The Windfall Clause 03:23 Avoiding Positive Side Effects: Concrete Problems in AI Safety part 1.5 09:38 Tim Tyler: The wirehead problem 28:30 Rob Miles: AI Safety 00:30 Elon Musk to Joe Rogan on AI Safety ..#elonmusk #joerogan More results