Insights · The Growth Coach HK
The Businesses That Will Struggle With AI Aren't the Slow Adopters — They're the Ones With Unclear Decision Rights
22 May 2026
Introduction
The conventional wisdom on AI adoption frames the risk as a speed problem: organizations that move slowly will fall behind those that move fast.
This framing is incomplete. And the incompleteness is costing organizations that have focused on speed at the expense of the organizational clarity that AI adoption requires.
The businesses struggling most with AI right now are not the slow adopters. They are the ones that adopted AI into unclear decision rights — and are now producing more output faster without anyone being clearly in charge of where it points.
Main Insight
AI accelerates the implementation of decisions. It does not make the decisions. Someone still has to.
And when organizations scale AI adoption without having clarified who decides what, the result is AI executing efficiently in the absence of clear direction — which means either nothing meaningful happens, because AI waits for human direction that never arrives clearly, or the wrong things happen, because the direction comes from whoever fills the vacuum.
Decision rights — the explicit allocation of who decides what, at what level, with what scope — are the organizational infrastructure that AI adoption multiplies. Where decision rights are clear, AI adoption creates compounding efficiency. Where they are unclear, AI adoption creates compounding confusion.
Note: This article addresses the organizational design problem of decision rights as it applies to AI adoption specifically. For the broader framework on how to build clear decision rights and ownership in any business context, see The Founder Bottleneck: Why Growth Stalls When Everything Runs Through You.
Common Mistakes
One mistake is assuming AI adoption will clarify decision rights by forcing the organization to articulate how decisions get made. Sometimes it does. More often, it reveals the ambiguity without resolving it — producing faster execution of confused authority.
Another mistake is treating decision rights as a one-time design problem. Decision rights need to be maintained as the organization evolves. What was clear at twenty people is often unclear at sixty.
A third mistake is equating AI tool adoption with organizational AI readiness. A team equipped with AI tools but operating under unclear authority is not AI-ready. It is a faster version of the same confusion it had before.
Framework: Decision Rights Before AI Scale
For each function where you are scaling AI adoption, work through four questions.
Who decides — explicitly, by name and role — for each significant decision category in this function? Not a team. One person.
What is the scope of that authority — what can they decide without escalating, and what requires escalation and to whom?
How do AI-supported decisions get reviewed — is there a quality check built in, or does AI output go directly to action?
What happens when AI and human judgment diverge — is there a clear protocol for situations where the AI output and the decision-maker's judgment point in different directions?
Working through these questions before scaling AI adoption in each function identifies the decision rights gaps that would otherwise become visible as operational problems after the fact.
Practical Lessons
The organizations getting the most from AI adoption have done the organizational clarity work either before or simultaneously with adoption.
They have mapped their significant decision categories and named an owner for each. They have defined the scope within which each role can decide without escalating. They have established the escalation criteria clearly enough that people can apply them without asking up.
This work is not glamorous. It often feels like organizational housekeeping rather than strategic leadership. But in an AI-accelerated context, it is one of the highest-leverage investments a leader can make — because it determines whether AI creates capacity or creates confusion.
Conclusion
AI adoption is not just a technology decision. It is an organizational design decision.
The leaders who understand this treat AI adoption as a prompt for decision rights clarity work, not just a productivity intervention. They ask: what would our organization need to look like for AI to create compounding advantage here? And they address the organizational design gaps alongside the technology adoption.
The ones who do not will discover the design gaps when AI has made them more expensive.
FAQs
Why are some businesses struggling with AI adoption despite moving fast?
Because they adopted AI into unclear decision rights — and are now producing more output faster without clear direction. AI accelerates the execution of decisions. It does not make the decisions. Where decision rights are unclear, AI adoption creates compounding confusion rather than compounding efficiency.
How do I know if my decision rights are clear enough for AI adoption?
For each significant decision category in the function you are scaling AI into, ask: can the person responsible for this decision make it without asking up? If the answer is no — if escalation is the default because authority is unclear — the decision rights need to be addressed before or alongside AI adoption.
What should I do first — clarify decision rights or adopt AI?
Where decision rights are already clear, adopt AI aggressively. Where they are unclear, address the clarity work simultaneously. Waiting until all decision rights are perfectly designed before adopting AI is too slow. Running AI adoption into significant authority ambiguity is too costly.
This site requires JavaScript for the full experience. Email us or enable JavaScript to continue.