2025-12-09
LessWrong
An overview of AI in 2025, including arguments for and against above-trend model capabilities growth, the state of evals, and the safety of reasoning models
Gavin Leech / LessWrong : X: @g_leech_ , @g_leech_ , and @patrick_oshag X: Gavin Leech / @g_leech_ : My summary of the year in AI [image] Gavin Leech / @g_leech_ : ADeLe really is an amazing eval [im...
2022-11-23
Gizmodo
10 related
Meta's researchers detail Cicero, an AI trained to “human level performance” in negotiation-based strategy game Diplomacy, ranking in the top 10% over 40 games
for the first time, an AI is able to consistently manipulate humans to act against their own interest, and further the AI's goals, using only natural language. And all along, humans don't even know th...
Loading articles...