The Problem of Trust

Capability, on its own, is not always sufficient.

For it to have an effect, it needs to be trusted.

This is not always straightforward.

In many contexts, decisions are made under conditions of uncertainty. Time may be limited, information may be incomplete, and the consequences of being wrong may carry weight.

In these situations, what matters is not only whether someone is capable, but whether that capability is recognised as reliable.

Trust becomes a practical requirement.

It shapes who is listened to, whose judgment is relied upon, and whose decisions carry weight.

But trust does not always align neatly with capability.

It is often built through signals that are easier to interpret — past experience, credentials, institutional affiliation, or other indicators that suggest reliability.

These signals can be useful. They provide a way to make decisions when direct observation of capability is not possible.

But they can also create a gap.

Capability may be present, but not yet trusted. Or trust may be placed in signals that do not fully reflect how someone engages with a situation in practice. This is not necessarily a flaw.

It reflects the difficulty of interpreting capability in environments where it is not always directly visible, and where decisions need to be made despite that uncertainty.

As AI becomes more present in how work is carried out, this gap can become more pronounced.

Outputs may appear well-structured and coherent, but it may not always be clear how they were formed, or what assumptions sit behind them.

Trust, in these cases, may shift toward the system, the interface, or the apparent completeness of the response — rather than toward the underlying reasoning.

This introduces a subtle tension.

If capability is not fully visible, and trust is placed elsewhere, decisions may be made on the basis of what appears reliable, rather than what is.

Over time, this can shape how capability is recognised — and whose judgment is relied upon.

Seen in this way, the question is not only whether capability exists.

It is how trust is formed around it — and how that trust influences the decisions that follow.