Attention Is Not the Scarce Resource. Trust Is.

The economics of information have inverted twice in thirty years. First, distribution stopped being scarce: anyone could publish. Now production has stopped being scarce: anyone can generate credible-sounding material on any subject in seconds. The bottleneck that remains is neither getting heard nor making things. It is knowing which of the things is true.

Abundance without verification

A search that returns forty articles on a question is not forty times more useful than one that returns one. It is less useful, if the reader cannot tell which of the forty is grounded in anything. The cost of evaluation now exceeds the cost of production by a wide margin, and evaluation is the part that cannot be automated away, because automating it requires trusting the automation.

What people will actually rely on

• Named individuals with a track record. Reputation accrues to people more readily than to publications now, and a person who was right in public before is a usable signal.
• Sources with something to lose. Accountability is a form of verification: a writer whose reputation is damaged by being wrong has a stake the anonymous aggregator does not.
• Work that shows its evidence. Not citations as decoration, but data, methodology, and the ability to check.
• Direct experience. First-hand accounts of things the writer actually did are the hardest category to fake convincingly and the most valued as a result.

Consequences for anyone publishing

The strategic implication is uncomfortable for volume-based approaches. Producing more competent, unremarkable content adds to the pile that readers are trying to filter, and being in that pile is worse than not being seen at all. The alternative is fewer pieces that carry evidence of a real person having done something and formed a view about it.

In an abundance of information, the scarce good is a name you can trust when you do not have time to check.

Consequences for institutions

Organisations built on being a source of information face a harder question than distribution: why should anyone believe them specifically. The answers that hold up are the traditional ones, which they have often let erode: transparent methods, visible corrections, and consequences for being wrong. These are expensive and they are the entire product now.

The optimistic reading

This is not obviously a decline. Cheap production means more people can contribute, and trust systems adapt: readers get better at signals, communities develop verification norms, and the value of genuine expertise rises rather than falls. The transition is disorienting, but the resulting equilibrium may reward substance more than the previous one did.