Skip to content
NewA calmer, faster ZeroHour
MalwarehighExploited in the wild

AI Accounts Are Becoming the New Target for Infostealers

Security Affairsby Pierluigi Paganini ()4 min read32 · 4 reads
AI summary · grok-4.7Summaries by Vani AI

SOCRadar found infostealer logs exposing AI sessions and API keys at 482 companies, mostly ChatGPT.

SOCRadar analyzed stealer logs from the last 90 days and found 482 companies with exposed AI accounts, including 295 in active logs and 5,434 records tied to 1,500 corporate emails. ChatGPT or OpenAI sessions appeared at 358 companies, about 90% of records, far ahead of Zapier, Notion, Hugging Face, and other tools. Stolen cookies and API keys can be replayed after password resets, exposing prompt history and enabling LLMjacking billed to victims. Anthropic previously invalidated Claude sessions after infostealer malware drained paid usage.

Key facts
  • 482 companies had exposed AI credentials; 295 appeared in active 90-day logs.
  • ChatGPT sessions were found at 358 companies, roughly 90% of records.
  • Password rotation does not invalidate stolen session cookies or API keys.
  • Underground forums sold Claude keys, ChatGPT cookies, and Cursor sessions.
  • Stolen Zapier sessions can automate exfiltration into CRM, email, and storage.
Analyze this story with Vani AIImpact, who is affected and what defenders should do, in a new chat

From the original article

4 min read at securityaffairs.com

Infostealers are exposing corporate AI accounts, sessions and API keys, giving attackers access to sensitive data, compute and connected systems. SOCRadar analyzed stealer log data from the last 90 days and found 482 companies with exposed AI accounts and credentials. Of those, 295 appeared in active logs during that period, suggesting the exposure is recent […]

Security Affairs

By Pierluigi Paganini · securityaffairs.com

Read the full article at Security Affairs

A short excerpt; the article and its images belong to the publisher. Summary, key facts and assessment above are AI-written from the article and can be wrong.