The Alignment Problem: Making Superhuman AI Want What We Want
How do you make an AI pursue what you actually want, not just what you measured? We explore the alignment problem through vivid real examples, from a game-playing AI that set its boat on fire to rack up points to a ChatGPT update so flattering that OpenAI rolled it back. We explain how today's chatbots are kept on track with human feedback, written rulebooks, and red teaming, and why experts who disagree about everything else agree that alignment gets harder as AI gets more capable.
This episode includes AI-generated content.
This episode includes AI-generated content.
You May Also Like

Ungovernable Misfits
UNGOVERNABLE

TechCrunch Startup News
TechCrunch

Mark Tilbury
Mark Tilbury Fan

AI Unraveled: Latest AI News & Trends, ChatGPT, Gemini, DeepSeek, Gen AI, LLMs, Agents, Ethics, Bias
Etienne Noumen

Engadget News + Next
Engadget

Rodrigo Arnedo Noticias BTC
Rodrigo Arnedo Noticias BTC

Super Data Science: ML & AI Podcast with Jon Krohn
Jon Krohn

Tech Update | BNR
BNR Nieuwsradio

Jak Piggott
Jak Piggott

ADSP: Algorithms + Data Structures = Programs
Conor Hoekstra, Bryce Adelstein Lelbach & Ben Deane
Community discussion