AI Agent Security and Red Teaming for GenAI
Check your Description
Lakera, now part of Check Point, provides runtime protection and red teaming for generative AI applications, AI agents and workforce AI use.
Sold as Check Point AI Agent Security, its model agnostic runtime layer screens prompts, documents and agent interactions for prompt attacks and data leakage, while Check Point AI Red Teaming simulates direct and indirect attacks before release.
Detection draws on more than 80 million adversarial patterns from Lakera's Gandalf hacking game and runs with under 50 milliseconds of latency.
The products are built for enterprise teams putting conversational, document and connected agents into production.
Founded in 2021 by David Haber, Mateo Rojas-Carulla and Matthias Kraft, Lakera was acquired by Check Point in November 2025.