DAMM and the EU AI Act
DAMM is the decision governance layer that helps AI agent deployers meet key requirements of the AI Act — human oversight, transparency, risk management and audit trail.
The problem
Autonomous AI agents make decisions with real-world impact — on customers, money, contracts, people.
The EU AI Act has been in force since August 2024, with high-risk and transparency obligations progressively applying from 2026. It places growing obligations on deployers, not just providers: human oversight, decision traceability and transparency toward users are required. Whoever ships an agent to production needs concrete tools, not just good intentions.
How DAMM helps
An honest mapping between DAMM’s capabilities and some specific AI Act requirements. Each card shows the relevant article.
Human Oversight
High-risk systems must enable effective and proportionate human oversight. DAMM places a structured assessment (4 pillars) with a verdict and an oversight recommendation between the agent and the action: a concrete way to demonstrate human oversight over decisions.
Risk Management
DAMM assesses the risk of each individual decision — Asymmetry (risk/reward ratio) and Margin (capacity to absorb an error) — feeding a continuous operational risk management process, decision by decision.
Transparency
Every report generated via the DAMM API includes a transparency block declaring that the verdict is AI-generated and recommending human oversight — contributing to the disclosure obligation toward users.
Audit Trail & Accountability
DAMM produces a readable record of why each decision was taken or blocked — who, when, on which dimensions, with which alternative. Useful for traceability and internal documentation.
The transparency layer
Every report generated via the API automatically includes a machine-readable transparency block. This is the field actually returned by the engine:
{
"transparency": {
"aiGenerated": true,
"regulation": "EU AI Act",
"article": "Article 50",
"system": "DAMM Decision Engine",
"humanOversightRecommended": true,
"disclosure": "This verdict was generated by an AI system. Human oversight is recommended before acting on this recommendation."
}
}Machine-readable and included automatically in every API report: it declares that the verdict is AI-generated and recommends human oversight, contributing to the Art. 50 disclosure obligation.
Two sides of the same requirement
For technical teams
- Integration via a single REST API
- Programmatic oversight before every agent action
- Machine-readable transparency block in every response
- Readable audit log for traceability and debugging
For compliance & management
- Mapping to specific AI Act requirements
- Reduction of single-decision risk
- Documentation supporting the compliance strategy
- Readable reports, not technical logs to interpret
Frequently asked questions
Concise, fact-checked answers on the EU AI Act and DAMM’s role as a decision governance layer.
What is the EU AI Act?
The EU AI Act is Regulation (EU) 2024/1689, the world’s first comprehensive legal framework on artificial intelligence. It entered into force on 1 August 2024 and applies on a staggered schedule through 2027, classifying AI systems by the risk they pose to health, safety and fundamental rights.
Who is the deployer under the AI Act?
The deployer is whoever uses an AI system under their own authority in the course of their activity — the organization that puts an AI agent to work in its own processes. Deployers have their own obligations, which grow especially when the system is high-risk, and are distinct from providers, who develop and place the system on the market.
What does Article 50 of the AI Act require?
Article 50 sets transparency obligations: when an AI system interacts with people or generates or manipulates content (text, audio, images, video), users must be clearly and distinguishably informed that they are dealing with an AI or with artificially generated content, at the latest at the moment of first interaction or exposure.
What is human oversight in Article 14?
Article 14 requires high-risk systems to be designed to allow effective human oversight, proportionate to the risk and context. The aim is to prevent or minimize risks to health, safety and fundamental rights, keeping a point where a person can understand, check and — when needed — stop or correct the system.
Does DAMM make you compliant with the EU AI Act?
No. DAMM helps meet specific requirements — especially human oversight, transparency and a decision audit trail — as one component of a broader compliance strategy. It is not a complete compliance solution, does not constitute legal advice, and does not replace a legal advisor or a DPO.
From when does the AI Act apply?
Application is staggered: prohibitions on banned practices apply from February 2025, obligations for GPAI models from August 2025, most obligations for high-risk and transparency systems from 2026, and obligations for high-risk systems linked to regulated products from 2027.
Important note
DAMM helps meet specific EU AI Act requirements (human oversight, transparency, decision risk management, audit trail) as one component of a broader compliance strategy. It does not constitute legal advice, is not a complete compliance solution, and does not replace a legal advisor or a DPO.
See DAMM at work on your decisions
Try decision analysis for free, or talk to us about your API integration.