Unverified50% confidenceFactExact time
Meta内部泄露的研究文件（2021年）承认平台算法存在'如果内容让人愤怒，用户更可能分享'的机制性问题
1
Sources
50%
Confidence
Long-term
Relevance
7/8/2026
First Seen
Sources
Related Claims
UnverifiedBy 2024, security researchers had demonstrated indirect prompt injection attacks that hijack RAG system outputs by embedding hidden instructions in web pages65% similarUnverified在具备工具调用能力的 Agentic 系统中，一次成功的提示注入可能导致数据泄露、恶意代码执行甚至供应链攻击64% similarUnverifiedMeta在2023年发布LLaMA权重后，权重在数天内被泄露到非官方渠道，研究者展示了通过少量数据微调即可移除安全护栏63% similarUnverified2024年已有多份威胁情报报告记录了攻击者利用AI辅助编写钓鱼邮件、生成恶意代码变体和自动化侦察流程的案例63% similarUnverifiedLLMs are vulnerable to prompt injection attacks, where malicious web pages or files can embed hidden instructions to induce AI to perform unintended actions.62% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/225212API
curl https://kongchang.com/api/v1/knowledge/claims/225212MCP
get_claim(id=225212)