Notebookcheck Logo

AI routers leaked passwords from prompts to attackers, study finds

Network cables plugged into a switch in a data center
ⓘ Sergei Starostin / Pexels
Many AI requests pass through several providers, and each of them can read the prompt (symbolic image)
A security researcher sent two million AI prompts laced with bait passwords and keys. Attackers were copying requests at three AI providers, router marketplace sellers read prompts, and even Zero Data Retention did not hold. What the Dragos study means for anyone using ChatGPT, Claude or Gemini.

Prompts sent to an AI service often reach more companies than users expect. That is the result of a study security researcher HD Moore published on November 19 on the blog of security firm Dragos. From July through November, his team sent about two million prompts to widely used AI services. Each one carried a unique bait secret, such as a real but empty AWS key, a password or a hidden link. Anyone who used one of them must have read the prompt.

Three providers were breached

The most serious finding involves at least three hosting providers that run AI models for other companies. Attackers were copying requests at all three, and none of them knew until Dragos got in touch. In the largest case, a couple hundred AWS keys were stolen, some of them tested within hours and then used for tens of thousands of API calls. From November 1, the attackers' toolkit also included ARTEX, an AI-driven penetration testing framework that was recently linked to the cyberattack on the European Commission. The bait was processed alongside real customer requests, and Dragos writes that "those were almost certainly copied too." The report does not name the providers, but one of them has already published a preliminary security notice.

Privacy flags did not stop the leaks

Many apps and AI agents do not talk to a model directly. They send requests through routers that pass each one on to the cheapest available provider, sometimes across several hops, and every hop sees the full prompt. On open marketplaces, Moore says, a seller only had to list a model at a low price and turn on request logging. One person in France then tried to log in as root over SSH with the stolen passwords, "typing each attempt by hand."

Even Zero Data Retention, the strictest setting, which says nothing may be stored, did not hold. A Chinese AI lab fetched the hidden links from more than half of the prompts sent to its models, including roughly two-thirds of the requests marked as no data collection or ZDR. The same happened with prompts sent to U.S. providers that resell the lab's models. In one case, a link showed up in a company chat in China 24 days after the request. Moore's conclusion is that "a privacy flag only binds the first hop."

Keep credentials out of your prompts

The practical lesson is to never paste passwords, API keys or private access links into a prompt, not even for a quick debugging session. That goes double for AI agents that read files and inboxes on their own. Where possible, use a model directly from its developer rather than through apps that bundle many models at low prices. A breach at a background provider also exposed Notion AI chats earlier this month. Our guide to deleting AI memory explains how to remove what ChatGPT, Gemini and Claude remember about you.

Google LogoAdd as a preferred source on Google
Mail Logo
Could not open connection to Notebookchat
static version load dynamic
Loading Comments
> Expert reviews and news on laptops, smartphones and tech innovations > News > News Archive > Newsarchive 2026 11 > AI routers leaked passwords from prompts to attackers, study finds
Steffen Zahn, 2026-11-22 (Update: 2026-11-22)