GRACE AI Platform for AI Governance and Security

2021.AI is a Copenhagen-based enterprise AI company founded in late 2016 by CEO Mikael Munck to make advanced AI accessible beyond the large tech companies. It supplies models, platform and governance so organizations can implement and operate AI responsibly, with a focus on data sovereignty and compliance with the EU AI Act and GDPR.

The GRACE AI Platform combines modules for governed GenAI chat and assistants, AI security, governance of risks, controls and assessments, MLOps for developers and agentic AI. The AI Auto Registry automatically detects, registers and monitors AI use cases, systems, models and datasets. The GRACE AI Security module, launched in June 2026, links the registry with an AI Gateway controlling which models staff can use, Chat Guardian that flags GDPR-protected data in prompts and Gatekeeper Agent that reveals external AI tools in use. It runs in cloud, private cloud or on-premise, with local or hybrid LLMs. Users include Maersk, AXA, DNV and Rigshospitalet.

• AI Auto Registry: automatically detects and registers AI use cases, systems, models and datasets.
• AI Gateway: single access point controlling which models employees may use.
• Chat Guardian: flags GDPR-protected and sensitive data before it reaches a model.
• Gatekeeper Agent: visibility into external AI tools in use.
• Governance module: risk, controls and assessments aligned to the EU AI Act.
• Sovereign deployment: cloud, private cloud or on-premise with local or hybrid LLMs.

Vietnamese provincial agencies and hospitals can run governed AI chat on local infrastructure with prompts scanned for personal data. Banks and insurers can keep a register of all AI models and use cases for audits. European-invested manufacturers and logistics firms in Vietnam can apply group AI policies locally.

Organizations get governed GenAI chat and an AI registry in one platform from under 2 EUR per user per month on published plans. On-premise and local LLM options keep sensitive data inside the organization. Automatic registration of AI systems and prompt scanning reduce manual compliance work. The security module can go into production within weeks, according to the company.

Data sovereignty has become a key concern as organizations depend on foreign AI providers, and regulators in Europe and Asia expect registers of AI systems. Vietnam's data localization and personal data protection rules give public bodies and regulated firms similar reasons to prefer on-premise, governed AI platforms.

Founded in late 2016 by CEO Mikael Munck; headquartered at Ryesgade 3F, Copenhagen, Denmark, with English, Danish and Swedish sites. Public-sector deployments include North Denmark Region on Dell on-premise infrastructure and the Danish Climate Data Agency. Partner integrations include Safe Online, Velatir and Lex Futura. Published pricing runs from 2,675 EUR per month for 1,000 users.