Unverified50% confidenceEventExact time
PromptArmor demonstrated that the SuperHuman email assistant leaked dozens of emails to an attacker's Google Form via a single Markdown image, with the data exfiltrated alongside the request when the client loaded the image.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
10/8/2026
First Seen
Valid until: 1/6/2027
Sources
Related Entities
Related Claims
Unverified此类骗局邮件使用 AI 生成的头像和赞美文本、Gmail 账户，且很可能每次更换姓名，难以追溯65% similarUnverified2024年已出现实际案例：通过网页内容中的隐藏指令，攻击者成功操纵具有邮件发送权限的AI助手向外部地址转发敏感内容64% similarUnverifiedGoogle提供Postmaster Tools工具查看域名在Gmail系统中的送达数据，但仅在每日向Gmail发送数百封以上邮件时才显示有意义的数据63% similarUnverified邮件转发中Received字段记录每一跳服务器信息，Return-Path字段存储退信地址，X-Forwarded-To等扩展头字段可能暴露真实收件人63% similarVerified间接提示注入可将恶意指令藏入外部数据源（如网页、文档），待代理读取时自动触发攻击62% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/988955API
curl https://kongchang.com/api/v1/knowledge/claims/988955MCP
get_claim(id=988955)