Anthropic Tests Claude Trading and Sabotage Detection in EVE
· unite.ai · popularity 21/100 · 0 similar stories today · AI & Data Science (Artificial Intelligence)
Read the original at unite.ai →
Summary
Anthropic published Project EVE research on November 6, 2026, showing that Claude agents could profit in simulated EVE Online markets while four-agent teams in EVE Frontier almost never caught a teammate explicitly instructed to sabotage their mission. The work was part of the Anthropic Fellows Program from April through July 2026 and involved a partnership with Fenris Creations. Both games combine scarce resources, player-built supply chains, navigation, combat and trade, providing…
Found in: Unite.AI. Rating basis: single outlet; outlet tier 2 (unlisted); 1 feeds; feed position 1.