Unverified60% confidenceFactExact time
The concept of AI text watermarking was first systematically proposed by John Kirchenbauer and others at the University of Maryland in their 2023 paper "A Watermark for Large Language Models".
2
Sources
60%
Confidence
Long-term
Relevance
8/16/2026
First Seen
Sources
Claude隐形水印争议：AI输出内容可被溯源追踪
Related Claims
Unverified基于词表分割的文本水印方法由马里兰大学John Kirchenbauer等团队于2023年提出79% similarUnverifiedThe paper "A Watermark for Large Language Models" published by the University of Maryland in 2023 proposes dividing the vocabulary into green words and red words, with green words appearing at a significantly higher frequency in watermarked text than the statistical baseline.69% similarUnverifiedWatermarks in large language models are not visible messages, but a subtle statistical pattern introduced by the model when selecting tokens.67% similarUnverifiedA recent arXiv paper proposes the Corporate Language Model (CLM) framework.62% similarUnverified研究发现语言模型更相信新发布的文章，且更相信由AI撰写的内容61% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/758685API
curl https://kongchang.com/api/v1/knowledge/claims/758685MCP
get_claim(id=758685)