TECHNOLOGY

How Hackers Are Using AI to Steal Crypto

Sat Nov 08 2025

Hackers are getting smarter with AI. They're now using it to create and change their malicious code on the fly. This makes it harder for security systems to catch them.

Google's Discovery

Google's security team found at least five different types of malware that use AI. These malware strains can generate new code as needed, instead of having it pre-written. This is a big change from how malware used to work.

Notable Malware Families

  • PROMPTFLUX: Uses AI to rewrite its own code every hour.
  • PROMPTSTEAL: Uses AI to create Windows commands on demand.

North Korean Hacking Group

A group from North Korea, known as UNC1069, has also been using AI to steal cryptocurrency. They used AI to:

  • Find wallet data
  • Create scripts to access encrypted storage
  • Write phishing emails in different languages

Google's Response

Google has taken steps to stop these activities. They:

  • Disabled the accounts involved
  • Added new safeguards to prevent AI abuse
  • Implemented better filters and monitoring of AI access

The Future of Cyber Threats

This shows a new way hackers can use AI to steal digital assets. They can use it to:

  • Find wallets
  • Create custom scripts
  • Make convincing phishing emails

It's important to stay vigilant. As AI gets more advanced, so do the methods hackers use. Understanding these threats can help protect against them.

questions

    What role do AI models like Gemini and Qwen2.5-Coder play in the evolution of malware?
    Are the claims of AI-powered malware just a hoax to divert attention from more serious cyber threats?
    What are the potential legal and regulatory frameworks that could be established to address AI-enabled malware?

actions