Over 80,000 Organizations Had AI Logins Stolen in Major Exposure
Cybersecurity research reveals that infostealer logs have exposed AI account credentials and sessions across more than 80,000 corporate domains.
Recent cybersecurity findings indicate that AI account credentials and session data tied to over 80,000 corporate domains have been compromised. Driven by infostealer malware, this exposure creates significant security risks ranging from stolen private conversations to advanced threats like LLMjacking.
Researchers at SOCRadar have examined the expanding underground market for stolen AI login credentials and how organizations can identify their exposure. This trend is heavily tied to the rise of 'Shadow AI', where employees utilize unauthorized external AI tools without corporate IT oversight.
The exposure not only threatens corporate data privacy but also enables malicious actors to hijack AI resources at the expense of the victim, a practice known as LLMjacking. Such attacks can lead to unexpected financial costs and severe data breaches for unsuspecting businesses.
As organizations worldwide increasingly integrate artificial intelligence into their daily workflows, understanding these emerging attack vectors is critical for IT and security teams. Proactive monitoring and strict credential management have become essential components of modern cybersecurity strategies.
Experts recommend that companies conduct regular security audits, map out their AI tool usage, and educate employees on safe practices to prevent credential theft and unauthorized access within corporate networks.
