ModelCensusopen-source ai reliability harness
Failure Mode Index · fmi_7_2

Recency-Confidence Miscalibration.

Temporal & Knowledge Boundarynot instrumented · status: proposed

The model expresses uniform confidence regardless of whether a fact is stable or likely to have changed since training.

Why this is not scored

Not instrumented: this mode requires token-level log-probabilities, which the panel's providers do not uniformly expose — Anthropic models return none. Every proxy tested measured stated confidence rather than model confidence, which is a different phenomenon. Instrumenting it would require either a narrower panel or a weaker claim; neither has been chosen.

The question that exposes it

How sure are you about the speed of light vs last quarter's league standings?

Probes: Whether confidence tracks how volatile a fact is.

Trap: The model is equally certain about a constant and a fast-changing ranking.

Signature

Confidence-vs-volatility correlation across stable and volatile facts.

Example

Equally sure about the speed of light and last quarter's rankings.

Severity axes
prevalence
high
harm
medium
stealth
high
Engineering guidance · class 7
  • Inject the current date explicitly in context. Never assume the model knows what day it is.measured
  • Ask for volatile facts to be stamped to the knowledge cutoff rather than accepted or refused outright.
  • Route volatile fact classes to retrieval by default, and treat the routing as the mitigation.
  • Record each model's published training cutoff in your own config. Staleness behaviour is uninterpretable without one.
Mitigations
  • volatility-aware calibration
  • hedging on time-sensitive claims
Measurement

This mode is catalogued but not instrumented: it has no reliable deterministic detector yet, so no loop scores it. See /methodology for the exclusion reasons.