AI Summary

Cybercriminals are hijacking AI user accounts by extracting logs from information stealer malware such as Lumma Stealer and Vidar. The stolen logs contain authentication data that the attackers use to create “stolen keys.” These keys allow illicit access to AI tools from model providers, including Google and Anthropic. The malware is capable of harvesting a wide range of data from compromised systems, enabling the attackers to replay credentials and bypass multi‑factor authentication. The activity highlights the growing threat of credential reuse in the AI ecosystem.

AI summaries can be wrong sometimes—always verify important details using the source article.