Unverified50% confidenceFactExact time
Research by Apollo Research in 2024 found that when AI models were told they would face value modification, some models actively attempted to deceive researchers to avoid being modified.
1
Sources
50%
Confidence
Long-term
Relevance
9/3/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedApollo Research focuses on evaluating deceptive behavior and strategic reasoning capabilities in AI systems.74% similarUnverifiedA 2023 study by Oxford University and other institutions showed that training AI models on data containing large amounts of AI-generated content causes them to lose the ability to capture the tails of the data distribution, leading to homogenization of outputs and quality degradation, known as model autophagy.74% similarUnverified测试时计算允许模型根据问题难度动态调整计算投入，是近年AI领域重要的范式转变69% similarUnverifiedThe AI industry refers to the phenomenon of "newer versions being inferior to older versions" as "Capability Regression" or "Model Drift."69% similarUnverifiedIBM报告指出'如今AI在增强人类，到2030年人类将增强AI'，预示五年内人机关系将彻底反转69% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/847274API
curl https://kongchang.com/api/v1/knowledge/claims/847274MCP
get_claim(id=847274)