- Modell
- qwen2.5:7b (lokal via Ollama, aihub.iio.space Fallback)
- Erzeugt
- 2026-08-05
- Felder
- 77
- Norm
- Einfache Sprache ~B1 — nicht Leichte Sprache (DIN SPEC 33429)
- Geprüft von
- Operator-Chat-Freigabe (gate.iio-space-einfache-sprache-content-review-v2, Session AGT-1DED3AED) · 2026-08-05
The operating system for
AI organizations.
IIO makes AI in the enterprise controllable: 259 IIO layers — governance, operation and cooperation — 606 automated flows, HITL gates at every critical point. No LLM in the compliance path. EU-operated.
What IIO is.
IIO isn't an AI app — it's the governance infrastructure beneath your AI tools. IIO controls which data which model may see, when a human must review, and generates the audit trail regulators demand.
You keep using AI models — but controlled, documented and GDPR-compliant. Local models on EU hardware, no US cloud lock-in.
Governance. Operation.
Cooperation.
IIO started as a governance tool. Today, the vast majority of layers are operational — finance, HR, fleet, tenant operations. Governance remains the control layer, not the only layer.
Control: HITL gates, policy engine, EU AI Act and ISO 42001 evidence. No LLM in the compliance path — decisions are deterministic and auditable.
Operations: finance, HR, fleet, tenant management, payroll, billing — the vast majority of the 259 IIO layers run real business processes, not just control.
Collaboration: MCP and A2A connect IIO directly with Claude, Cursor, LangChain, AutoGen. Node-to-node cooperation over open protocols — documented at Open Cognition Commons.
Cooperation protocols (node↔node, agent swarm) are draft/experimental — documented at Open Cognition Commons, not submitted to any external standards body.
Intelego GmbH.
IIO is operated and supported by Intelego GmbH. Company history, background and contact: intelego.solutions/about.
When uncertain, the system stops and escalates. Safe over fast.
Every significant action produces evidence. No silent intervention.
The framework is open.
IIO's research and open framework live at the Open Cognition Commons — for developers and researchers. iio.space is the commercial, operated path; OCC is the open one.
opencognitioncommons.org →How we think.
When uncertain, IIO stops and escalates. No silent failures, no self-healing attempts without the operator's knowledge.
Every significant action produces machine-readable evidence. The audit trail isn't an afterthought — it's the product.
IIO runs on standard Linux, open-source models, EU hardware. No proprietary LLM API lock-in.
Humans stay in the decision loop. AI makes proposals, humans approve. Always — not just for high risk.
All production data on Hetzner servers in Germany. No US subprocessors, no transatlantic data transfers.
What IIO does is explainable. No black-box decisions. Every flow is documented, every rule is readable.
Governance by design — not by compliance.
IIO wasn't built to tick checkboxes. The HITL gates, the audit trails and the policy engine grew out of operational necessity — compliance is a side effect, not the goal.
Art. 9 (Risk Management), Art. 14 (Human Oversight), Art. 52 (Transparency)
Human-in-the-loop prevents fully automated individual decisions
AI management system analogous to ISMS structure (clauses 5–10)
GOVERN, MAP, MEASURE, MANAGE — all four functions implemented