AI can thin out businesses without making them disappear.
Sounds strange, but just separate the business into 2 layers:
Coordination layer: find partners, negotiate, delegate tasks, monitor, troubleshoot
Organizational layer: hold assets, be the name holder, grant permissions, record commitments, take responsibility
The part that AI is eroding first is the coordination layer.
If AI makes searching, negotiating, and executing cheaper, part of the work can go outside the business.
But cheaper coordination does not automatically create a new entity that can hold assets, act within clear rights, and leave a history reliable enough for others to rely on.
That’s where blockchain starts to make sense, if it makes sense.
Not like "intelligence".
But like a thin infrastructure layer for the organization: holding digital assets, enforcing rules, delegating actions, recording commitments.
Therefore, the question worth considering is no longer:
Can AI replace businesses?
The more relevant question is:
Which functions of the business are becoming cheaper at the coordination layer, which functions can be encoded at the organizational layer, and which functions still need to stay with the old-style governance?
Sounds strange, but just separate the business into 2 layers:
Coordination layer: find partners, negotiate, delegate tasks, monitor, troubleshoot
Organizational layer: hold assets, be the name holder, grant permissions, record commitments, take responsibility
The part that AI is eroding first is the coordination layer.
If AI makes searching, negotiating, and executing cheaper, part of the work can go outside the business.
But cheaper coordination does not automatically create a new entity that can hold assets, act within clear rights, and leave a history reliable enough for others to rely on.
That’s where blockchain starts to make sense, if it makes sense.
Not like "intelligence".
But like a thin infrastructure layer for the organization: holding digital assets, enforcing rules, delegating actions, recording commitments.
Therefore, the question worth considering is no longer:
Can AI replace businesses?
The more relevant question is:
Which functions of the business are becoming cheaper at the coordination layer, which functions can be encoded at the organizational layer, and which functions still need to stay with the old-style governance?