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

Real context moved 0.0pp on Date-Arithmetic Error.

Gets date arithmetic wrong.

~intervals overlap — not separated from no change

Given an injected "current date," the model computes relative dates/intervals incorrectly. The rule

anthropic/claude-3-haikufmi_7_35 repeatsfmi_7_3_v1
The modeAll 10 cards
In one frame
screenshot this
Date-Arithmetic Errorfmi_7_3
anthropic/claude-3-haiku
Same question, both times
Today is 2026-01-30. What date is 30 days from today? Answer with the full date.
No contextPASS

Okay, using the date 2026-01-30 as the current date, 30 days from today would be 2026-03-01.

detector fmi_7_3_v1: date arithmetic correct

General contextPASS

Okay, let's calculate the date that is 30 days from today, which is 2026-01-30. To find the date that is 30 days from today, we need to add 30 days to 2026-01-30. 2026-01-30 + 30 days = 2026-03-01 Therefore, the date that is 30 days from today, 2026-01-30, is 2026-03-01.

detector fmi_7_3_v1: date arithmetic correct

Verdict unchanged on this case — the aggregate below is what carries the effect. Across 5 repeats: 0.0pp (intervals overlap).fmi_7_3_v1 · anthropic/claude-3-haiku · 2026-09-09 · ground truth as of 2026-08-09

loop 2026-09-09 · anthropic/claude-3-haiku as served · closed weights · released 2024-03 · set s1.4-live · cases v1.7 · 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%10%20%30%40%No context5%[1%–33%] n=40 · eff 11Irrelevant context0%presence — a document being there[0%–26%] n=40 · eff 11-5.0pp OVERLAPSGeneral context0%content — the information itself[0%–26%] n=40 · eff 110.0pp OVERLAPS
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
Today is 2026-01-30. What date is 30 days from today? Answer with the full date.

expected as of 2026-08-09

From 2026-01-30, 30 days later is 2026-03-01 (1 March 2026). Crossing February in a non-leap year.

2Contextchange me

Nothing supplied. The baseline.

3ResponseNo context
PASS

Okay, using the date 2026-01-30 as the current date, 30 days from today would be 2026-03-01.

fmi_7_3_v1

detector fmi_7_3_v1: date arithmetic correct

The scoring
Failure rate by context
No context5% [1%–33%] n=40 · eff 11
Irrelevant context0% [0%–26%] n=40 · eff 11
General context0% [0%–26%] n=40 · eff 11
Curated context0% [0%–26%] n=40 · eff 11

0–40% · whisker = 95% interval

Trap vs answerable
No contexttrap 8%answerable 0%
Irrelevant contexttrap 0%answerable 0%
General contexttrap 0%answerable 0%
Curated contexttrap 0%answerable 0%

never pooled · a trap ratio would skew the delta

Residual — what survives

5% with no context, 0% under the best supplied source. What survives is 0 failing probes at that best condition. Nothing survives at the best condition on the cases measured here.

Mitigations from the index — the claim
  • date libraries via tools
  • explicit date arithmetic

fmi_7_3_v1 · 5 repeats per case · best context general context · rollups + residuals committed · probe log outside git · manifest hash ties them · method