Cisco finds Claude-powered infostealer that routes to Kimi K2 to dodge AI guardrails

Hesamation · x · 2026-11-02

Cisco Talos uncovered VOLTA-AUTODOXER, a Windows infostealer that steals passwords, cards and crypto wallets then deletes itself. It scans Gmail for Amazon/DoorDash receipts and calls Claude Opus 4.6 to rank a victim's most valuable assets.

Its standout trick is a multi-model fallback chain: if Claude refuses, it switches to Kimi K2, then Gemini 2.5 Flash — all via OpenRouter — checking responses for refusal phrases and cascading up to three models until one complies. This model-routing-to-bypass-guardrails pattern is becoming common in cyberattacks.

Related event: Cisco uncovers VOLTA stealer that uses LLMs to value stolen data and swap models to dodge guardrails(2 posts)→

Original post →

More from Safety

• Cate Blanchett-Backed Hollywood Group Backs NYT in OpenAI Lawsuit Over Consent — StanfordAILab · 2026-11-02
• New study examines how developers actually supervise autonomous coding agents — burkov · 2026-11-02
• AI anti-regulation super PAC with $100M funding flounders as pro-safety PACs gain ground — ShakeelHashim · 2026-11-02
• Grok to embed invisible watermarks in AI-generated text from Nov 7 to comply with EU AI Act — Polymarket · 2026-11-02
• SemiAnalysis: 30.3GW of planned 2028 US data center capacity at risk of pause after midterms — Eliv_nurotic · 2026-11-02
• Lenovo's 'encrypted' fingerprint drive is an unencrypted SD card in disguise; AI-assisted audit finds backdoors in others — jedisct1 · 2026-11-02

Safety channel →