Unverified50% confidenceCautionExact time
水印方案在开源模型上彻底失效的根本原因之一是哈希密钥无法保密，一旦绿/红名单划分规则被逆向推断，攻击者可主动规避绿名单词汇
1
Sources
50%
Confidence
Long-term
Relevance
7/16/2026
First Seen
Sources
Related Claims
Unverified水印方案的有效性依赖模型提供方对哈希密钥的保密，这导致水印在开源模型上彻底失效79% similarVerifiedOnce open-source model weights are made public, attackers can remove a model's safety restrictions through fine-tuning.60% similarUnverifiedThe fundamental ways to break symmetric encryption are brute-forcing all keys or discovering mathematical structural flaws in the algorithm, neither of which is a domain where LLMs excel.59% similarUnverifiedMatt Green推断OpenAI与Anthropic极可能都使用单一全局密钥加密所有客户端推理数据，属于密钥管理反模式58% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/533409API
curl https://kongchang.com/api/v1/knowledge/claims/533409MCP
get_claim(id=533409)