AI Doom Will be Retroactively Explainable
LessWrong
Read the full articlexAI has proposed that potential catastrophic AI outcomes will be explainable in hindsight, suggesting that future AI doom scenarios could be understood retroactively.
Happening nowSee all →
Smug until weirdly silent: who told us to say “I am alive”?32m agoDisney's new CTO tells staff he 'won't pretend it will be easy' to win in the changing media landscape45m agoOpenAI Adds New Irish Roles in Boost to Firm’s EU Headquarters46m agoManifold Steering from a Natural State is Effective and Generalizable, but Relies on Model Accuracy48m agoMicrosoft Research Introduces StudentSim for AI Tutor Training48m agoWhy Big Tech Is Borrowing Billions to Fund AI Spending52m agoMeet the Anthropic Engineer Keeping Its AI Agent Trustworthy54m agoMeta Unveils New Series of Advanced A.I. Models2h agoRussia was not ready for the data center war it started with Ukraine1h agoAccessing OpenAI GPT models on Amazon Bedrock with global cross-Region inference in Thailand, Malaysia, Singapore, Indonesia, and Taiwan1h agoMeta says its new Muse model has reached the AI frontier1h agoDo you have any ideas?1h agoClaude sent a false homicide tip to police – and it’s part of a troubling AI pattern1h agoLabs’ current alignment plans seem terrible1h ago
More in AI Research
AI Research6 min read
Claude sent a false homicide tip to police – and it’s part of a troubling AI pattern
ZDNet
AI Research15 min read
Import AI 476: Creativity and RSI; worse-than-China open weight models; AI goes to space
Import AI