Lakera
Real-time GenAI runtime security — prompt-injection and jailbreak defense for LLM apps.
Acquired — still operating
acquired by Check Point Oct 2025 (~$201.8M)
About
Lakera built Lakera Guard, a real-time security layer that screens LLM inputs and outputs against prompt injection, jailbreaks, data leakage, and model manipulation across frameworks. The Swiss startup became famous partly through Gandalf, its viral prompt-injection education game with over 1 million players generating 50M+ attack data points — used even by Microsoft for security training — and raised a $20 million Series A led by Atomico in July 2024, with customers including Dropbox.
Important status change: Check Point Software acquired Lakera in a deal announced September 2025 and closed October 22, 2025, for approximately $201.8 million in total consideration per SEC filings (press reported ~$300M). The lakera.ai brand remains live, but it now operates as Check Point's AI-security engine and the foundation of its Global Center of Excellence for AI Security — evaluate it as a Check Point product line, not an independent vendor.
What it can do
- Real-time prompt-injection and jailbreak screening
- Output and data-leakage protection
- Continuous AI red teaming
- Gandalf AI-security training platform
Setup
Contact via lakera.ai (now under Check Point) for Guard API access or a red-teaming engagement.
Prerequisites
A GenAI application to protect; enterprise procurement via Check Point.
Use cases
Works with
Trust score — 7/10
Scored 7/10 (metrics tier, metrics-v1). Evidence: Verified via vendor site/docs/GitHub and reputable coverage, wave 60; Trending score 70/100 from researcher assessment.
Sources
Every fact above was checked against these sources. Where a claim comes from the provider and couldn't be independently verified, the text says so.