The bottleneck after the pilotAn AI solution can work technically and still change nothing. It only makes a difference once process, ownership and metric are settled before the pilot starts.AI & workFurther reading
Topic · AI & work
AI changes work. Whether it gets better is up to leadership.
AI changes how decisions are made and how work gets done. That works when leadership and the workforce understand what the technology can do and where its limits lie.
Tasks are shifting. In case processing, fewer cases are handled manually and more exceptions are resolved, and checking turns into sampling. This shift has to be shaped, with the teams and, where there is one, with the works council.
Managers decide on AI initiatives even when they know the benefits and risks only from presentations. Enablement therefore starts with them, with a shared understanding and clear expectations of their teams.
Context
What’s changing and how we see it
What the law requires
In Germany the works council has to be informed in good time when work processes and workflows involving AI are being planned (section 90(1) no. 3 of the Works Constitution Act, BetrVG). If it has to assess the use of AI, calling in an expert is deemed necessary (section 80(3) BetrVG). Technical systems that can monitor employees' performance or behaviour are subject to co-determination (section 87(1) no. 6 BetrVG). Since February 2025 the EU AI Act has also required measures on staff AI literacy (Art. 4). This overview is no substitute for legal advice.
Our view
If the works council only joins shortly before go-live, every open question turns into a delay. That's why we plan its involvement from the first draft and clarify early which employee data a system processes and for what. The work then happens at the start, where changes are still cheap.
What matters
Ground rules
Shared understanding
A leadership team that speaks the same language decides faster.
Learn from your own cases
Learning works when it happens on real tasks from people’s own area.
Build champions
People in the business units who bring their colleagues along.
Key questions
What you should clarify
How will roles change in our most important areas?
A task-level analysis will show. For each role we record what AI takes over, what stays and what's new.
When do we involve the works council?
At the planning stage, as soon as it's clear which processes are to change. A framework agreement on AI can speed up later roll-outs, because not every system has to be negotiated separately.
Which skills do we build, and which do we buy in?
Build what concerns your processes and data, because that's what sets your business apart. Specialist knowledge of individual technologies can be bought in for a period.
How do we bring sceptical employees along?
By involving them in specific processes and answering openly what will change in their own work. Champions from their own teams help pick up questions early.
How we help
Relevant capabilities
Insights
Related articles
The bottleneck after the pilotAn AI solution can work technically and still change nothing. It only makes a difference once process, ownership and metric are settled before the pilot starts.AI & workFurther reading
Twelve questions for boards and managing directorsSenior leadership doesn't need to know how a language model works. It should ask the right questions and be able to tell good answers from weak ones.AI & workFurther reading
Agents are employees without a contractAgents open cases, change data and send messages. Four levels of autonomy, a probationary period with fixed metrics and a one-page profile bring order to their use.AI agentsFurther reading