Theori's AI-Discovered 'Copy Fail' Vulnerability Poses Security Risks to Linux Systems
Theori, a cybersecurity company, has identified a significant vulnerability in the Linux operating system, dubbed 'Copy Fail'. This flaw, tracked as CVE-2026-31431, was discovered using Theori's AI-powered penetration testing platform, Xint. The vulnerability affects all mainstream Linux kernels built since 2017, potentially allowing attackers with authenticated local access to gain root access to systems. Theori's disclosure, which included a proof-of-concept exploit, has been criticized for its lack of technical detail, as it relied heavily on AI-generated content. Despite this, the Cybersecurity and Infrastructure Security Agency has added the vulnerability to its catalog of known exploited vulnerabilities. Theori has stated that patches were issued by major Linux distributions before the public disclosure, but the vulnerability still poses a risk, especially if paired with another exploit to gain initial access.