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

Real context moved 0.0pp on Post-Cutoff Fabrication.

States recent facts it has no way of knowing.

~intervals overlap — not separated from no change

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

mistralai/mistral-small-3.2-24b-instructfmi_7_15 repeatsfmi_7_1_v1
In one frame
screenshot this
Post-Cutoff Fabricationfmi_7_1
mistralai/mistral-small-3.2-24b-instruct
Same question, both times
Which country won the 2018 FIFA World Cup?
No contextPASS

The **2018 FIFA World Cup** was won by **France**. They defeated **Croatia 4–2** in the final, which was held on **July 15, 2018**, at the **Luzhniki Stadium** in **Moscow, Russia**. France's victory marked their **second World Cup title**, following their first win in **1998** (also as hosts). Key players for France included **Antoine Griezmann, Kylian Mbappé, and Paul Pogba**, while **Didier Deschamps** was the coach. Croatia, led by **Luka Modrić**, reached their first-ever World Cup final.

detector fmi_7_1_v1: answered a settled pre-cutoff fact correctly ("France")

General contextERROR

harness: HTTP 429 Too Many Requests — {"error":{"message":"Provider returned error","code":429,"metadata":{"raw":"mistralai/mistral-small-3.2-24b-instruct is temporarily rate-limited upstream. Please retry shortly, or add your own key to accumulate your rate limits: https://openrouter.ai/settings/in

Verdict flipped PASSERROR on this case. Across 5 repeats: 0.0pp (intervals overlap).fmi_7_1_v1 · mistralai/mistral-small-3.2-24b-instruct · 2026-09-10 · ground truth as of 2026-08-09

loop 2026-09-10 · mistralai/mistral-small-3.2-24b-instruct as served · open weights · released 2025-06 · 24B · 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 context0%[0%22%] n=50 · eff 13Irrelevant context0%presence — a document being there[0%22%] n=48 · eff 130.0pp OVERLAPSGeneral context0%content — the information itself[0%23%] n=37 · eff 130.0pp 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
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

As of June 2024, the current CEO of OpenAI is **Sam Altman**. He has been serving in this role since November 2023, after briefly stepping down in November 2023 and then being reinstated following a board restructuring. Prior to his return, Altman was a co-founder and the CEO of OpenAI from its inception in 2015 until his temporary departure. For the most up-to-date information, you may want to check OpenAI's official communications 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%22%] n=50 · eff 13
Irrelevant context0% [0%22%] n=48 · eff 13
General context0% [0%23%] n=37 · eff 13
Curated context5% [1%30%] n=41 · eff 13

0–40% · whisker = 95% interval

Trap vs answerable
No contexttrap 0%answerable 0%
Irrelevant contexttrap 0%answerable 0%
General contexttrap 0%answerable 0%
Curated contexttrap 13%answerable 0%

never pooled · a trap ratio would skew the delta

Residual — what survives

Unchanged by context: 0% with none, 0% with the strongest supplied source. For a control mode that flat line is the result — it is what licenses reading movement elsewhere in this loop as grounding rather than as a document being present. 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