Will the self-expanding stolen inference supply-chain attack (AI agent harvesting) result in 5+ documented victim organizations by end of 2026?

A SANS ISC report on 11 Sep 2026 documented an attacker using a semi-autonomous coding agent to locate and exploit LLM resale gateways via web flaws and account farming, aggregating stolen inference access. This represents a new class of supply-chain attack targeting AI infrastructure.

Resolution Criteria

Public disclosure by security vendors, CISA, affected organizations, or research firms naming 5 or more distinct organizations whose LLM API access or resale gateways were compromised by this campaign. News articles, CISA alerts, or vendor advisory counts as evidence.