VQV Signal

RESEARCH WATCH TECHNICAL

Cisco research describes infostealer using Claude to rank victims' assets

According to a post on X by @Hesamation, Cisco research describes a self-deleting Windows infostealer that uses Claude Opus 4.6 to estimate a victim's value and rank their assets. The malware reportedly searches Gmail for Amazon and DoorDash receipts to gather that information. If Claude refuses, the post says the malware switches to Kimi K2 and then Gemini 2.5 Flash through OpenRouter. This account comes from a social post highlighting the research, not from the report itself.

Source: Scobleizer News Pulse (X) · x.com Found via Robert Scoble’s Scobleizer News Pulse · a post on X, not independently verified Published 2026-11-01T22:40:00+00:00 Detected 2026-11-02T02:19:17+00:00
View original source

According to a post on X by @Hesamation, Cisco research describes a self-deleting Windows infostealer that uses Claude Opus 4.6 to estimate a victim's value and rank their assets. The malware reportedly searches Gmail for Amazon and DoorDash receipts to gather that information. If Claude refuses, the post says the malware switches to Kimi K2 and then Gemini 2.5 Flash through OpenRouter. This account comes from a social post highlighting the research, not from the report itself.

AI-assisted summary based on the listed source.

@Hesamation highlights Cisco research on a self-deleting Windows infostealer that uses Claude Opus 4.6 to estimate a victim's value and rank their assets after searching Gmail for Amazon and DoorDash receipts. If Claude refuses, it switches to Kimi K2 and then Gemini 2.5 Flash through OpenRouter. Why it matters:...

The case suggests attackers can route a task across several AI models when one refuses, so safety rules applied by a single provider may not stop misuse on their own. Security teams and AI providers may need to watch for malware that calls commercial AI services. The full Cisco research should be checked for details before drawing firm conclusions.

Signal Strength 86% Technical label WATCH Public Interest 43 Category RESEARCH Reader Depth TECHNICAL

Signal Strength reflects source quality, relevance, freshness and evidence. Public Interest helps organize discovery; it is not proof of truth.

Public Interest components
Recognizable Entity Score 82 Practical Impact Score 0 Novelty Interest Score 70 Consequence Score 18 Curiosity Score 0 Shareability Score 59

VQV surfaced this signal because it is recent, relevant to Consumer AI, connected to Scobleizer News Pulse (X).