Scanner Impersonating 32 AI Bots Probed for Exposed Credentials

· Original article ↗

Summary

CQ Prime Research says a scanner impersonating 32 AI bots probed organizations for credential and configuration files from late September into late October. Nearly all requests were blocked or failed, and the researchers reported no successful credential theft.

Key points

  • CQ Prime Research identified a scanner impersonating 32 AI bots, including identities associated with OpenAI, Anthropic, Perplexity, and Meta.
  • The scanner requested configuration files, cloud credential files, and private SSH keys across organizations in several industries.
  • It sent bursts of 50–190 requests in about eight seconds, including random paths, and rotated among bot identities.
  • On one day in October, 48% of Cequence-protected organizations received the malicious bot traffic.
  • Nearly all requests were blocked or failed; the article reports no successful credential theft.
  • Researchers recommend verifying agent identities, monitoring for behavior inconsistent with those identities, and avoiding static credentials in exposed files.

Article Details

Event Type
Malicious credential-scanning activity impersonating AI bots
Impact
CQ Prime Research reported that a scanner impersonated 32 AI bots while probing organizations for exposed passwords, API keys, cloud credentials, and private SSH keys. Nearly all requests were blocked or failed, and the researchers reported no successful credential theft or breach. The activity affected organizations across retail, financial services, telecommunications, energy, and software.

People

Vendors

Industries