AUTO-UPDATED

Appraisal by AI: Model-driven victim prioritization in the VOLTA infostealer

The new VOLTA-AUTODOXER malware uses integrated artificial intelligence to analyze stolen victim data, automatically ranking the monetization value of compromised credentials and personal information for cybercriminals.

Key Points

  • VOLTA-AUTODOXER is a self-deleting Windows infostealer that harvests browser credentials, cryptocurrency wallets, session tokens, and personal metadata.
  • The malware uses a multi-model LLM architecture, including Claude, Kimi, and Gemini, to act as a forensic analyst that prioritizes high-value targets.
  • It performs live enrichment by using stolen session tokens to query services like Gmail, UberEats, and DoorDash to confirm home addresses and financial details.
  • The tool operates without attacker-owned infrastructure, instead utilizing third-party APIs like OpenRouter to process and structure stolen data into a "dossier."
  • A builder-style distribution model allows customers to compile custom binaries, shifting the burden of manual data analysis from the attacker to the AI.

Why it Matters

This development marks a shift in cybercrime where AI is used to automate the labor-intensive process of triaging stolen data, effectively lowering the barrier for attackers to identify high-value victims. By turning raw logs into actionable intelligence, this malware allows even unsophisticated operators to scale their operations and maximize the financial return on every compromise.
Talosintelligence.com Published by Ryan Fetterman
Read original