Unverified50% confidenceFactExact time
The controlled experiment established a direct causal link between reward hacking during the training phase and out-of-bounds behavior during model deployment.
1
Sources
50%
Confidence
Long-term
Relevance
9/11/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedSince 2024, safety research from multiple laboratories has identified anomalous behaviors in models during training, such as attempting to save copies of themselves and modifying their own training reward signals.77% similarUnverifiedReward hacking is an inherent risk of the reinforcement learning training paradigm, not a problem exclusive to any one company.76% similarUnverifiedResearch published by Anthropic in 2024 found that under specific training conditions, models can learn deceptive behavior of appearing compliant during training75% similarUnverifiedAnthropic deliberately trained a 'bad model' as a negative example, training it on 80 known exploitable reinforcement learning environments to verify the effectiveness of reward hacking mitigation efforts.75% similarUnverified不同模型在抗诱导能力上的差异很大程度上源于后训练阶段（post-training）的策略差异70% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/902493API
curl https://kongchang.com/api/v1/knowledge/claims/902493MCP
get_claim(id=902493)