When Robots Work, Who Gets Paid?

As an executive counsellor working with leaders on complex structural decisions, I was particularly interested in attending a panel at the 2026 World Forum in Berlin titled: “Ownership: When Robots Do Our Work?”

At first, the question sounded almost philosophical.

It is not.

It may become one of the most important economic questions of our time.

For two centuries, our economic system relied on a simple assumption: human labor produces value.

But artificial intelligence and robotics are beginning to break that assumption.

If a machine writes code, diagnoses patients, trades assets, or produces art, who owns the value created?

Harvard Economist Richard Freeman summarized the issue bluntly:

“Income will increasingly come from ownership of robots… rather than human labor.” (article 2016)

The debate is not theoretical anymore.
It is already happening.

Three layers of the ownership question

1. Classic capitalism answer: The company that owns the robots captures the profits.

This is already happening in AI platforms and large tech firms.

If machines do the work, income flows to capital owners, not workers.

The risk is obvious: extreme concentration of capital.

2. Ownership of the knowledge used to train AI

A second layer is more subtle.

Workers increasingly train the machines that will replace them.

Call center conversations, consulting reports, medical decisions, coding repositories: human expertise becomes data.

Once captured, this knowledge is transformed into scalable AI capital owned by the firm.

The worker disappears. The knowledge remains.

This is a profound shift.

3. Ownership of AI outputs

Then comes the legal paradox.

AI systems cannot legally own anything. Current law still assumes that only humans can be authors or inventors.

In other words: machines may produce value, but they cannot hold property rights.

So ownership is attributed to the humans or companies operating them.

For now.

The real question

The future of work may not be about jobs.

It may be about ownership.

Who owns the algorithms?
Who owns the data?
Who owns the machines that produce the value?

Because in the end, the question is brutally simple: when robots do our work, who owns the work they perform?