Skip to content
NewA calmer, faster ZeroHour

Dragos canary test finds AI providers leaking prompts and AWS keys

Original title: A leaky AI supply chain is exposing secret user data at an alarming rate

Ars Technica · Securityby Dan Goodin ()5 min read1
Part of a story covered by 2 sourcesDragos canaries expose AI hosting breaches and leaked promptsMerged summary & timeline
AI summary · x-ai/grok-4.7Summaries by Vani AI

Attackers reused bait AWS keys tens of thousands of times after copying prompts at three hosting providers.

HD Moore of Dragos seeded nearly 2 million prompts with unique AWS keys, passwords, and secret URLs, one secret per provider. At least three hosting providers were compromised; attackers copied requests at the front door and reused bait keys tens of thousands of times. Some resellers logged prompts, and a Chinese lab fetched secret URLs even from requests marked Zero Data Retention.

Unconfirmed: The three hosting providers and the Chinese lab are not named; same-adversary linkage is the researcher's inference.

Key facts
  • Each secret was placed in exactly one prompt sent to a single provider.
  • Stolen keys from two providers were used together from the same rented server.
  • Attackers searched S3, secrets managers, and parameter stores, and added IAM users.
  • The lab fetched URLs from roughly two-thirds of no-collection or ZDR requests.
  • Prompts sent to some US resellers still reached the Chinese lab's scraper.
Security assessment · AI, from the article

Impact: Prompts and planted cloud credentials stolen and reused

Analyze this story with Vani AIImpact, who is affected and what defenders should do, in a new chat

From the original article

5 min read at arstechnica.com

Test using prompts with unique, trackable secrets proves that data doesn't stay private.

Ars Technica · Security

By Dan Goodin · arstechnica.com

Read the full article at Ars Technica · Security

A short excerpt; the article and its images belong to the publisher. Summary, key facts and assessment above are AI-written from the article and can be wrong.