For more than a century, organisations have been structured around people. We group employees into departments, assign them roles, draw boxes around them and connect those boxes with lines to show who reports to whom. It is neat, orderly and, depending on where you work, occasionally bears some resemblance to how things actually get done.
But what happens when an increasing amount of work is completed by AI?
As AI assistants, agents and automated workflows become deeply embedded into organisations, the traditional org chart may tell us who works where, but it will tell us increasingly little about how the work actually gets done.
A job is no longer entirely human
Most roles are made up of multiple tasks. The same employee might research information, analyse data, prepare documents and respond to stakeholders as well as make decisions and coordinate with colleagues.
Increasingly, some tasks may be completed with AI, some entirely by humans and others a mix of the two.
So, if AI eventually performs 30 per cent of the tasks associated with a particular role, does the role remain the same?
Perhaps on paper it does. The employee still has the same title, reports to the same manager and sits in the same department. But underneath that box, the composition of the work may change significantly.
Multiply that across hundreds or thousands of employees and the organisation itself begins to look very different, even if the org chart looks the same.
Map the work, not just people
Traditional organisational design focuses heavily on positions and reporting lines. In an AI-enabled workplace, organisations may also need to understand how work moves between humans and technology by creating a blueprint that rethinks:
- Who initiates the task?
- What part does AI complete?
- Where is human judgement required?
- Who checks the output?
- What happens when something goes wrong?
- And who remains accountable?
A future map of the organisation may therefore look less like a neat pyramid of job titles and more like a network of people, AI agents, systems and workflows.
Admittedly, it will be much harder to fit on the staffroom wall.
Departments may change too
AI is also likely to challenge the boundaries between traditional departments. An AI agent may gather information from finance, analyse data from operations, draft a report for management and trigger a communication to customers. The work crosses several organisational boundaries without necessarily respecting once carefully-drawn boxes.
If an AI-enabled process operates across four departments it raises new questions about ownership and accountability by posing:
- Who owns it?
- Who monitors it?
- Who has authority to change it?
The answer cannot always be “IT”. Technology teams may manage the systems, but the work itself still belongs to the organisation.
Redesign the work before redrawing the chart
Organisations need not immediately delete their org charts but it does mean they should start redesigning what sits beneath the boxes. As AI takes on more tasks, organisations need to examine how roles are changing, where responsibilities are shifting and whether existing structures still make sense.
Future organisation may not be defined solely by who works for whom. Increasingly, it may be defined by who or what does the work, where human judgement sits and who remains responsible for the outcome. Your org chart may still look the same for a while. The work underneath it probably won’t.
At GAPS Professional Writing, our workplace AI training helps organisations and teams examine how AI is changing not simply individual tasks, but the way work itself is designed and delivered.