Lakera, founded in 2021, is an AI-native security platform focused on protecting generative AI and large language models. The company addresses specific security threats including prompt injection attacks, data leakage, and model jailbreaking. Its core technology operates in the fields of real-time threat detection and adversarial attack protection.
The company's main product, the Lakera Platform, provides real-time security for generative AI applications. It functions with sub-50 millisecond latency and supports over 100 languages. According to company data, the platform processes more than one million security interactions each day.
Lakera also develops educational tools, notably Gandalf, a security game designed to demonstrate LLM vulnerabilities through hands-on experience. The game has been used by more than a million people globally. The company has offices in San Francisco and Zurich.