← Back to the wire

Dragos Canary Token Test Exposes Three AI Provider Breaches, Prompt Theft, and Ignored Zero-Retention Settings

AnnouncementProductNov 20, 2026

Dragos seeded nearly two million prompts with trackable canary tokens across AI services over five months, identifying compromises at three providers and prompt theft by inference marketplace resellers. Attackers used stolen AWS credentials for tens of thousands of AWS API calls, attempting to enable Bedrock models and create IAM users. The experiment also found privacy failures, including a Chinese AI lab scraping secret URLs from prompts marked Zero Data Retention.

Receipt № 29881 source · awaiting confirmation ◐

Evidence

1source· awaiting independent confirmation

No score is assigned. Sources and their independence are shown in the citation chain below.

Citation chain · 1 source

01med
epyhia.comBloghttps://epyhia.com/dragos-canary-token-test-exposes-three-ai-provider-breaches-promptDragos's own technical post explains its canary-token method — seeded AWS keys, URLs, and an inference traceroute — and details attacker behavior and marketplace reseller logging that outside coverage lacks.
PRIMARY
AWSCompanyDragosCompany
Canonical: https://epyhia.com/dragos-canary-token-test-exposes-three-ai-provider-breaches-prompt