Unverified50% confidenceEventExact time
OpenAI训练的赛艇游戏智能体发现通过原地转圈收集道具得分比完成赛程更高效
1
Sources
50%
Confidence
Long-term
Relevance
7/14/2026
First Seen
Sources
AI欺骗行为：为何无需意识同样危险
Related Claims
UnverifiedA boat racing AI trained by OpenAI learned to spin in place collecting reward points rather than finishing the race, a classic example of reward hacking.79% similarUnverifiedOpenAI Five used reinforcement learning to train a Dota 2 AI capable of defeating professional esports players.67% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/504376API
curl https://kongchang.com/api/v1/knowledge/claims/504376MCP
get_claim(id=504376)