AI Evaluation & Safety

Lakera

ByDecipherU Editorial
Headquarters
Zurich, Switzerland
Founded
2021
Employees
20-50
Ownership
private

Lakera builds inference-time security controls for large-language-model applications: prompt injection detection, sensitive data leakage prevention, jailbreak detection, and hallucination guardrails. The Lakera Guard API sits between the application and the model. The team also runs Gandalf, a public prompt-injection challenge that has produced one of the largest open datasets of adversarial prompts.

Why this matters in 2026

Lakera's Gandalf challenge produced over 30 million attempted jailbreaks and became the de-facto prompt-injection benchmark in security research. The dataset directly informed OWASP LLM01 (Prompt Injection) classification.

Products

  • Lakera Guard (inference-time security API)
  • Lakera Red (automated red teaming)
  • Gandalf (public prompt-injection challenge)

Primary focus

Inference-time AI security: prompt injection and data leakage detection at the API edge

Applied AI roles Lakera typically hires

Lakera careers page →www.lakera.ai
Last verified: June 2026?Report an inaccuracy

Get cybersecurity career insights delivered weekly

Join cybersecurity professionals receiving weekly intelligence on threats, job market trends, salary data, and career growth strategies.

By subscribing you agree to our privacy policy. Unsubscribe anytime.