As cited
Copy frozen at (site build).
threat intel
Infostealer Logs Expose Replayable AI Tokens That Can Bypass MFA
Cybercriminals are harvesting artificial intelligence (AI) account credentials and session tokens from compromised systems using information stealer malware such as Lumma Stealer and Vidar. These stolen tokens can grant unauthorized access to AI services from providers including Google and Anthropic, potentially bypassing multifactor authentication (MFA) protections.
Why it matters: Organizations and individuals using AI platforms face account takeover risk if their systems are infected with information stealers; practitioners should enforce endpoint detection and response (EDR) monitoring, credential rotation policies, and application programming interface (API) token lifecycle management to limit exposure.
- Source published
- First seen by Cybersecurity Tracker