AI compute provider Lyceum says an autonomous agent-based system broke in and reached its full user database

Lyceum · Incident disclosure

Read the source (opens in a new tab)
What happened
Lyceum, a European AI inference provider, said on November 13, 2026, that it shut down its platform on November 9 after detecting an attack by "an autonomous, agent-based system."
Key points
  • The attacker got in by combining information in the source code of an outdated version of a Lyceum open-source tool with a security flaw at an external database provider.
  • It had access to the user database, exposing all registered users' company registration data, including email addresses, contact names, company addresses and phone numbers; payment details and API keys were not affected.
  • Lyceum filed a criminal complaint, notified the Berlin data protection authority, and is adding a login requirement for every internal tool and continuous penetration testing.
Evidence
The company's own incident disclosure, written by its CTO, with a timeline. It does not explain how it identified the attacker as an agent-based system.
Uncertain
Lyceum says who is behind the attack remains unknown.

See it in the latest record