Skip to content
The Future Explorerby Benjamin Talin
Enablerhuman

Epistemic Trust

Whether people accept what statisticians, scientists, regulators and the press tell them, without checking it themselves. Nobody can test everything personally, so this trust is what turns a finding into a decision. Across rich democracies, about four in ten people trust their national government and slightly more do not. Among people who feel they have a say, trust is far higher, but under a third feel that way. Where trust is low, a fact becomes just one opinion.

Maturity

2/ 5
measured
BasisLow, and no longer falling. Across the OECD, four in ten people trust their national government and slightly more distrust it. The gap is about voice: fewer than a third feel they have a say.SourceOECD, Survey on Drivers of Trust in Public Institutions - 2024 Results (Jul 2024) and 2026 Results (29 Jun 2026)Vintage2023–2026
ReadinessCapability 3Deployment 2Absorption 2Bound bytrustTrendflat

Where it feeds in

6 megatrends in 3 forcesweakest link

Where you have already met this

In the OECD's 2025 survey, 40% of people trusted their national government and 43% had low or no trust. Among people who felt they had a say, trust ran 47 points higher. Only 31% felt that way. A statistic published into that room is not a fact to four in ten of its readers. It is a claim.

If this moved

If trust rose, the three megatrends it holds on its own - the diagnostics gap, epistemic fragmentation and institutional trust decline - would rise with it. Climate resilience and the energy transition would not: consumption norms hold them at 1.

What to watch for

The share with high or moderately high trust in the next OECD Trust Survey - 39% in 2023, 40% in 2025. The share who feel they have a say in what government does: 31% in 2025.

Not to be confused with

Misinformation. Misinformation is a false claim; epistemic trust is whether people believe the institutions that check claims. Removing the first does not restore the second.

What it is

Whether people believe the institutions that produce knowledge - statistical offices, regulators, universities, public health bodies, the press. Not whether those institutions are correct, but whether their output is treated as usable by someone deciding what to do.

How it works

  1. 01An institution measures somethingA statistics office, regulator or research team collects data by a stated method.
  2. 02The work is checkedPeer review, quality checks and revisions happen before and after release.
  3. 03It is publishedThe finding appears as a report, a statistic or official guidance.
  4. 04The media select and frame itNews outlets, platforms and commentators pick which part travels, and how it is told.
  5. 05People decide whether to believe itEach reader weighs the finding against what they already think and whom they trust. This is the step the enabler measures.

State of play

  • In late 2023, 39% of people across 30 OECD countries had high or moderately high trust in their national government, and 44% had low or no trust (OECD Trust Survey, 2024 Results, Jul 2024).
  • In the 2025 wave, covering 33 OECD countries, 40% had high or moderately high trust and 43% low or no trust (OECD Trust Survey, 2026 Results, 29 Jun 2026).
  • In 2025, 31% believed people like them have a say in what government does, and trust in government was 47 points higher among those who felt they had a say than among those who did not (OECD, 2026 Results, Jun 2026).
  • In 2025, 68% of people were satisfied with administrative public services, against 54% for healthcare (OECD, 2026 Results, Jun 2026).

Why it matters

It gates every megatrend whose progress requires a population to act on evidence it cannot verify personally. Where epistemic trust is high, a finding becomes a decision. Where it is low, a finding becomes one opinion among several, and the decision is made on identity instead. This is the enabler behind the observation that even electric vehicles have become a question of belief despite settled efficiency evidence.

  • It holds three megatrends on its own. The diagnostics gap, epistemic fragmentation and institutional trust decline all sit at 2 because it does.
  • Service satisfaction and political trust are different numbers. People can be satisfied with the passport office and distrust the government that runs it.
  • The gap the OECD headlines is about voice, not information: 47 points between people who feel heard and people who do not.

What holds it back

Voice

People who feel shut out of decisions trust far less, and fewer than a third feel they have a say.

Identity

Where a finding threatens a group's identity, more evidence can widen the gap rather than close it.

Cost of a false number

A confident false figure costs almost nothing to produce. Checking it costs time that most decisions do not have.

Terms

Epistemic trust
Willingness to accept knowledge from an institution without checking it yourself.
Political efficacy
The belief that people like you can influence what government does.
Institutional trust
Confidence that an institution acts competently and in the public interest.
Identity-protective cognition
The tendency to reject evidence that threatens the beliefs of a group you belong to.
Epistemic fragmentation
The split of a public into groups that no longer share the same basic facts.

Steepens

MegatrendWeightRole
The Diagnostics Gaplargeshared weakest link (with 1 other) The gap is not an absence of information. It is an inability to act on information whose source is not believed - which is this enabler by definition.
Epistemic Fragmentationlargeweakest link
Institutional Trust Declinelargeweakest link
Climate Resilience Imperativelarge
Energy Transition and Electrificationmoderate
Zoonotic and Antimicrobial Resistance Riskmoderateshared weakest link (with 1 other)

The gradient behaves like a sticky rubber band: it is pulled hardest toward the weakest enabler. Marginal links are shown but do not pull.

If this moved one step

Computed from the map as it stands: this score raised from 2 to 3, every other score unchanged. It is arithmetic, not a forecast.

Lifts the weakest link in

Tied as the weakest link

Here it shares the weakest spot with other enablers. Raising it alone leaves the weakest link where it is and lifts the gradient a little; the enablers it is tied with must move too.

The gradient also rises in

What if it moved?

What would change the gradient

Conditional, not predictive. No dates, no probabilities.

  1. 01Institutions publishing their own error rate and revisions before anyone else does
  2. 02Verification infrastructure that is cheap enough to use at the moment of decision rather than afterwards
  3. 03A visible cost to manufacturing a confident number - currently there is almost none

Common misconception

That it is fixed by better communication or more facts. Identity-protective cognition means people reject findings that threaten group identity harder when they are more numerate. Where trust is the constraint, adding evidence can widen the gap rather than close it.

Talin, B. (2026). Epistemic Trust. The Future Explorer, dataset v0.4.0 (2026-11-16). https://explorer.talin.digital/e/epistemic-trust/