Unverified50% confidenceFactExact time
The decision thresholds of AI-generated text detectors are usually set on long texts, and applying them directly to short texts causes the false positive rate to drift.
1
Sources
50%
Confidence
Long-term
Relevance
11/10/2026
First Seen
Sources
Related Entities
Related Claims
VerifiedAfter text is manually rewritten or polished, the accuracy of AI detection methods based on textual features drops significantly.80% similarUnverifiedThe repetitive_text signal checks texts of 12 words or more, treating a unique-word ratio below about one third as filler.74% similarUnverifiedTextGrain forms patterns recognizable by detection tools through fine-tuning word choices, but editing, translation, or the use of very short excerpts may weaken the signal.74% similarUnverifiedInference scenarios with long inputs and short outputs rely more on compute power, while long text generation scenarios place greater emphasis on memory bandwidth.73% similarUnverifiedAI text detectors identify differences between human writing and machine-generated content from the statistical features, linguistic patterns, and semantic structures of text.73% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1153408API
curl https://kongchang.com/api/v1/knowledge/claims/1153408MCP
get_claim(id=1153408)