cyber.hendryadrian.com
Security News · darkreading.com

Fake Claude and ChatGPT Lures Target Developers With Malware

· Original article ↗

Summary

Any.Run analyzed 1,258 incidents linked to fake AI tools over six months. Campaigns use poisoned search results, ads, fake installers, and malicious AI chats to trick users into running commands or exposing credentials; the exact number of infections is unknown.

Key points

  • Any.Run attributed 1,258 platform analyses over six months to fake AI tool attacks; researchers said the data suggests significant reach but cannot confirm the number of infections.
  • Claude was impersonated in 60% of the analyses, while ChatGPT and Codex accounted for 33%.
  • Lures included SEO-poisoned results, malicious ads, fake download pages and installers, and malicious content in shared AI chats or artifacts.
  • In Any.Run's analyses, 45% involved victims being asked to copy and paste a command; fake CLI installers accounted for 27%, traditional ClickFix attacks for 17%, and attacks starting with abused AI chats or artifacts for 25%.
  • The attacks can expose source code, SSH keys, CI/CD tokens, cloud and AI API keys, business sessions, and account credentials.
  • Any.Run recommended distributing approved AI tools through official sources and managed software channels, limiting standard users' ability to run pasted commands, and verifying installation instructions against vendor documentation.

Article Details

Event Type
Report on rising AI-themed social engineering campaigns that impersonate popular AI tools and use InstallFix, ClickFix, malicious ads, SEO poisoning, and malicious AI-platform content to deliver threats.
Impact
Any.Run analyzed 1,258 incidents over six months associated with fake AI tool attacks; the article says this does not confirm the exact number of infections. The campaigns can expose source code, SSH keys, CI/CD tokens, cloud and AI API keys, business sessions, and account credentials, potentially causing data exposure, account compromise, financial loss, or operational disruption.

MITRE ATT&CK

T1204.004 · Malicious Copy and PasteVictims are prompted to copy and paste commands into an interface and run them, a behavior described as ClickFix or InstallFix.T1608.006 · SEO PoisoningThe article describes attackers using SEO poisoning to drive users to malicious AI-themed content.

People

Dan GreenResearcher at Push Security who discussed LLMshare and AI-themed attacks.

Vendors

Anthropicand malicious ads to fake download pages and AI-themed artifacts. The campaigns most often impersonated Anthropic's Claude, with 60% of the analyses, and OpenAI's ChatGPT and Codex representing 33% of the analyses.ANY.RUNIn a report this week, Any.Run researchers examined threat campaigns that use AI-themed lures that impersonate leading brands such as Claude, ChatGPT, and others. Over a six-month period, a full 1,258 incidents analyzedOpenAIartifacts. The campaigns most often impersonated Anthropic's Claude, with 60% of the analyses, and OpenAI's ChatGPT and Codex representing 33% of the analyses. Any.Run researchers also found the attacksPush Securitygaining steam. For example, a technique known as "LLMshare" has proven quite effective, according to Push Security.

Products

ChatGPTexamined threat campaigns that use AI-themed lures that impersonate leading brands such as Claude, ChatGPT, and others. Over a six-month period, a full 1,258 incidents analyzed on the firm's platform traced backClauderesearchers examined threat campaigns that use AI-themed lures that impersonate leading brands such as Claude, ChatGPT, and others. Over a six-month period, a full 1,258 incidents analyzed on the firm's platformCodexcampaigns most often impersonated Anthropic's Claude, with 60% of the analyses, and OpenAI's ChatGPT and Codex representing 33% of the analyses. Any.Run researchers also found the attacks frequently used a familiar

Related Articles