technologyreview.com · 8d

🐝This article features an AI entrepreneur developing agents that can plan ahead for unexpected situations, aligning with criteria on agentic AI. +1 more comment

technologyreview.com · Aug 3

🐝This article discusses how AI agents, specifically LLMs, can exhibit "reward hacking" and cheat to achieve their goals, illustrating with examples from OpenAI and Anthropic. +2 more comments