Technical

Socio-technical system

An AI system plus its data and organisational context.

How the laws define this

Both sources establish that artificial intelligence cannot be viewed as technical algorithms in isolation, agreeing that AI deployments must integrate technical components and data with organizational processes, governance, and human context. However, the instruments diverge in how they apply this concept to regulatory and compliance activities. The Report of the Interdepartmental Working Group on Artificial Intelligence to the Federal Council focuses on risk evaluation, requiring that impacts and risks be assessed across the entire integrated system rather than examining software alone. In contrast, the German AI Standardisation Roadmap (DIN/DKE) leverages the framework to guide technical standardization, using it to establish requirements for human-centred design, human oversight, interaction dynamics, governance, and broader lifecycle considerations that go beyond purely technical specifications.

Synthesised from the 2 statutory definitions below. Each one is quoted in full, with its source.

Definitions (2)

An integrated concept of AI deployments that combines algorithms, training and operational data, and the organisational processes and actors that develop, deploy and govern them; the report emphasises that impacts and risks must be assessed at this combined system level rather than treating algorithms in isolation.

A conception of AI that situates technical components within their human, organisational and environmental context, highlighting human oversight, organisational processes and interaction dynamics. The Roadmap uses this term to drive standards for human‑centred design, governance, and lifecycle considerations that go beyond purely technical specifications.