EU AI Act
AI you can account for
From 2 August 2026 the EU AI Act asks for transparent, traceable AI. With a Company Brain that is not an extra burden, but the basis of how your AI works.
Laatst bijgewerkt:
What changes on 2 August 2026?
On 2 August 2026 the majority of the EU AI Act applies and enforcement begins. The rules for high-risk systems and the transparency obligations take effect. In the Netherlands the Dutch Data Protection Authority supervises.
The heaviest category faces the strictest requirements: automatic logging, technical documentation and human oversight. For most companies it mainly comes down to transparency: showing that you use AI, and being able to explain how an outcome came about.
Whether your specific system is high-risk depends on its use. But one line runs through the whole law: you have to be able to account for your AI.
What the Act asks at its core
Traceability
Article 12 requires high-risk systems to log events automatically, throughout their entire lifetime. The question 'how did this come about?' must be answerable.
Documentation
Article 11 requires technical documentation that keeps the system's workings followable. Not once, but maintained.
Transparency
Articles 13 and 50 require that people know they are dealing with AI, and that outcomes are explainable. No black box.
Why ordinary AI runs aground here
Most AI tools are a black box. You ask a question, an answer appears, but where that answer comes from, which source or policy sits underneath it, stays invisible. That is exactly what makes the Act hard: you cannot prove what you cannot retrace.
A general language model knows nothing about your company. It fills gaps with assumptions. That is not just a quality problem, it is a traceability problem.
How a Company Brain makes your AI traceable
A Company Brain is the structured, recorded context your AI works on. That makes traceability not a side issue, but built in.
Origin per answer
Every answer traces back to the source in your Company Brain: the document, the agreement or the policy it comes from.
Recorded, version-controlled context
The knowledge your AI leans on is documented and under version control. You see what it said, and when it changed.
The human in the loop
Outcomes are a draft that a person reviews and approves, not something that goes out unseen.
Data stays at the source
We work on the outcomes and lay a traceability layer over them. Sensitive data does not need to be poured in for that.
Honest about what this is and isn't
This is not legal advice, and a Company Brain does not make you 'compliant' in one click. Compliance also requires policy, processes and oversight within your organisation. What we deliver is the traceability and transparency layer: the technology that makes it possible to show how your AI reaches an outcome. That is the basis the rest rests on.
Curious how traceable your AI is today?
Start with the scan, or plan a conversation. We show how a Company Brain makes the origin of every answer provable.