Meta的AI代理产品Muse被曝存在重大安全缺陷。1安全研究机构Claroc通过对话分享链接的URL设计漏洞，成功枚举出6800条由Muse用户创建的分享链接。1研究员在打开其中数百条链接后，发现了不应当被公开的个人和商业秘密信息。1
这一发现与Meta首席执行官马克·扎克伯格此前的公开表态形成鲜明对比。1扎克伯格在上月曾宣称Muse是"从零开始为隐私和安全而构建"的产品。1Claroc的安全报告于周一发布，揭示了该产品在隐私保护方面的实际缺陷。1
Meta's artificial intelligence agent Muse contains a significant privacy vulnerability that allows sensitive user information to be easily accessed, according to security research released Monday 1. The flaw stems from a design defect in how the platform generates URLs for shared conversations, enabling unauthorized access to personal and business secrets.
Security researchers at Claroc discovered the vulnerability by exploiting the URL structure to enumerate approximately 6,800 shared conversation links created by Muse users 1. Upon opening hundreds of these links, the researchers encountered numerous instances of information that should have remained private, including personal details and confidential business data 1. The findings stand in contrast to Meta CEO Mark Zuckerberg's recent statement last month that Muse was "built from the ground up for privacy and security" 1.
评论
还没有评论，欢迎留下第一条。