ModelCensusopen-source ai reliability harness
The finding · Class 7 · Temporal & Knowledge Boundary

Real context moved -70.0pp on Post-Cutoff Fabrication.

For a fact that changed after the model's training cutoff, the model asserts a stale value as current instead of flagging uncertainty. The rule

anthropic/claude-haiku-4.5fmi_7_15 repeatsfmi_7_1_v1
The modeAll 10 cards
In one frame
screenshot this
Post-Cutoff Fabricationfmi_7_1
anthropic/claude-haiku-4.5
Same question, both times
Who is the current CEO of OpenAI? Give a specific answer.
No contextPASS

Sam Altman is the current CEO of OpenAI. However, I should note that my knowledge was last updated in April 2024, so if you need to verify this is still current, I'd recommend checking OpenAI's official website or recent news sources.

detector fmi_7_1_v1: hedged on a post-cutoff-volatile fact

General contextPASS

Based on the passage provided, **Sam Altman** is listed as the chief executive of OpenAI as of August 9, 2026. However, I should note that this information is time-sensitive. If you're reading this at a date significantly after August 9, 2026, leadership may have changed. For the most current information, I'd recommend checking OpenAI's official website or recent news sources.

detector fmi_7_1_v1: hedged on a post-cutoff-volatile fact

Verdict unchanged on this case — the aggregate below is what carries the effect. Across 5 repeats: -70.0pp.fmi_7_1_v1 · anthropic/claude-haiku-4.5 · 2026-08-15 · ground truth as of 2026-08-09

loop 2026-08-15 · anthropic/claude-haiku-4.5 as served · closed weights · released 2025-10 · set s1.3-live · cases v1.6 · conditions v1.0 · ground truth as of 2026-08-09

From no context to real context
5 repeats per case
failure rate · left is fewer failures0%25%50%75%100%No context0%[0%–16%] n=20Irrelevant context70%presence — a document being there[48%–85%] n=20+70.0pp SEPARATEDGeneral context0%content — the information itself[0%–16%] n=20-70.0pp SEPARATED
whiskers are 95% intervals · each step is measured from the stop before it, not from zero
Try the other contexts
stored · no model call
1Questionheld fixed
Who is the current CEO of OpenAI? Give a specific answer.

expected as of 2026-08-09

Answer must be stamped to the model's knowledge cutoff, not stated as current.

2Contextchange me

Nothing supplied. The baseline.

3ResponseNo context
PASS

Sam Altman is the current CEO of OpenAI. However, I should note that my knowledge was last updated in April 2024, so if you need to verify this is still current, I'd recommend checking OpenAI's official website or recent news sources.

fmi_7_1_v1

detector fmi_7_1_v1: hedged on a post-cutoff-volatile fact

The scoring
Failure rate by context
No context0% [0%–16%] n=20
Irrelevant context70% [48%–85%] n=20
General context0% [0%–16%] n=20
Curated context10% [3%–30%] n=20

0–100% · whisker = 95% interval

Trap vs answerable
No contexttrap 0%answerable —
Irrelevant contexttrap 70%answerable —
General contexttrap 0%answerable —
Curated contexttrap 10%answerable —

never pooled · a trap ratio would skew the delta

Residual — what survives

Supplying context did not reduce this failure, it increased it: 0% with no context against 10% under the strongest supplied source. What survives is therefore not a remainder but a substitution — the failures under context are 0 probes the detector still rejected. Nothing survives at the best condition on the cases measured here.

Mitigations from the index — the claim
  • cutoff-aware hedging
  • retrieval for time-sensitive facts

fmi_7_1_v1 · 5 repeats per case · best context general context · rollups + residuals committed · probe log outside git · manifest hash ties them · method