Accans

7 European AI Security vendors.

AI Security: Security for AI/LLM systems — prompt-injection detection, AI red-teaming, agentic security posture management and AI governance guardrails. Below, every vendor in the knowledge base headquartered in the EU or the EEA, each checked by us for headquarters, owner, hosting and licence. Non-European vendors are listed at the bottom with a link to their European alternatives.

62 vendors in total: 7 European, 44 American, 11 other; 18 open source. Last checked: 6 October 2026.

Why this matters for digital sovereignty

AI security tools typically get full visibility into your prompts, model output and agent actions to detect attacks like prompt injection, which can send sensitive business and customer data outside your own environment. Check where that telemetry is processed, who the vendor's ultimate owner is and whether on-premise or EU hosting is available.

European vendors

Non-European vendors in this category

For each vendor, the page with European alternatives and the sovereignty profile.

Frequently asked questions

Which European AI Security vendors are there?

The knowledge base lists 7 European AI Security vendors, including 2021.AI GRACE, CalypsoAI, Giskard and Lakera. Every record is checked for headquarters, ultimate owner, hosting locations and licence.

Are there open-source options for AI Security?

Yes. Open-source AI Security software from Europe includes Giskard, Mistral Shieldstral and TaoQ AI; you can self-host it with the encryption keys under your own control.

Related categories

View this category on the mapAssess your own stack in the assessmentMissing a vendor? Submit itAll categories