Do you sell AI governance or compliance software?Founding Vendor Placement — $499 for 12 months →
EU AI Act · Article 4 active · Article 50 from 2 Aug 2026 · high-risk rules transition through 2027–2028
AI Act Navigator
Directory/Lakera
Source-checked profile

SOFTWARE INTELLIGENCE PROFILE

Lakera

AI Security & Guardrails

Swiss AI security company providing real-time inference firewalls against prompt injection, jailbreaks, and PII leaks.

Visit official website ↗ Compare tools
HeadquartersZurich, Switzerland
Starting price$800/mo
Last verified9 Oct 2026Source-checked profile
Source evidence2 recordsAutomated monitoring
PROFILE SNAPSHOT

What this platform provides

2 source records available

Swiss AI security company providing real-time inference firewalls against prompt injection, jailbreaks, and PII leaks.

PROVIDERDEPLOYER$800/mo

Visit official website ↗

CUSTOMER FIT

Deployment & technical profile

Hosting / residency signalSwiss Entity
DeploymentGlobal Edge API / VPC
APIAvailable
Python SDKAvailable
On-premiseNot recorded
Customer fitApplications Requiring Active Runtime Defense

Recorded export formats: Live Webhook, SIEM Syslog, JSON

PROFILE QUALITY

Profile completeness & evidence signals

99/100 completeness
Evidence completeness100%
Technical readiness signalStrong
Requirement mappings2
Editorial stateNo review flag recorded

These are transparent catalogue signals used for discovery and prioritisation. They do not establish legal compliance, procurement suitability or endorsement.

CAPABILITIES

Operational capabilities

ai-securitytesting-monitoring
REGULATORY MAPPING

Relevant compliance capabilities

These are Navigator's recorded capability mappings, not a statement that the vendor itself makes an AI Act compliance guarantee.

  • Article 15 (Continuous AI Cybersecurity)
  • Supporting capability — runtime injection defense

Recorded active / current scope

  • Supporting capability — transparency/security workflow
  • PII Redaction Guardrails

Recorded prepare / transition scope

  • Article 15 Continuous AI Cybersecurity Defenses
LIMITATIONS

Recorded limitations

  • Runtime firewall focus; does not generate static Annex IV technical files
ASSURANCE SIGNALS

Evidence & standards

Labels below are separated by evidence type. A standard, framework, product capability or hosting note is not presented as a certification.

Certifications & attestations

SOC 2 Type II

Standards & frameworks

No standards/framework evidence recorded.

Other recorded evidence

Gandalf Threat Intel
Verification methodTechnical Documentation Review
ConfidenceHigh
Source records2
PROVENANCE

Evidence trail

Source-checked profile

Automated monitoring records source availability and change signals. Non-material changes may be automatically assured; material regulatory, certification, commercial or technical changes remain subject to editorial confirmation.

  • ok — https://www.lakera.ai/ (checked 2026-10-09)
View recorded source URLs
Editorial & legal scope

This profile is informational and based on source information recorded in the Navigator dataset. “Source-checked” describes the Navigator's editorial/data process; it is not a legal certification, conformity assessment, endorsement, or guarantee of regulatory compliance.

Framework reference: Regulation (EU) 2024/1689 · Profile reviewed: 9 Oct 2026

Read the methodology →