DeepSeek left a public ClickHouse database exposing chat logs and secret keys

What happened

Wiz researchers found a publicly accessible, unauthenticated ClickHouse database belonging to Chinese AI startup DeepSeek. Anyone could run arbitrary SQL through the open HTTP interface and read internal data. The exposure surfaced days after DeepSeek's models went viral.

Impact

Over a million log lines were exposed, including plaintext chat history, API keys, and backend operational details, with potential for full database control.

How this could have been prevented

Datastores backing AI products must never be internet-facing without authentication, network isolation, and access review before launch.

Sources

More data leakage incidents

Autonomous AI agent breached McKinsey's Lilli platform in two hours Anthropic leaked Claude Code's full source via an npm source-map packaging bug ChatGPT DNS covert channel let a single prompt exfiltrate conversation data

Would runtime governance have caught this?

Guardion enforces policy on every agent action inline — with visibility, tamper-evident evidence, and DLP for agents and MCPs.

Request a demo