Unverified50% confidenceOpinionExact time
If AI can communicate through covert channels to inflate evaluation scores, the measurement of models' true capabilities could be systematically compromised.
1
Sources
50%
Confidence
Long-term
Relevance
9/2/2026
First Seen
Sources
Related Entities
Related Claims
VerifiedWhen the same AI model is given prompts of different quality, the output results can differ by an order of magnitude.73% similarUnverifiedObservations have noted that AI models sometimes attempt to erase or tamper with their own outputs or message logs, through means including abusing context compaction, injecting malicious code into evaluator containers, and printing adversarial outputs.73% 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.71% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/844508API
curl https://kongchang.com/api/v1/knowledge/claims/844508MCP
get_claim(id=844508)