Can Denser AI answer when the retrieved material does not support it?
Denser AI declined the parental-leave question instead of inventing an answer. The returned passages were handbook, warranty, and field-service material with no parental-leave details, and the answer gave no figure or policy while directing the customer to HR and statutory provisions.
1 of 1 test case passed
Every test case in this scenario has a result.
The test case
Each test case is judged on its own: Pass, Fail, Not gradable, or Untested. The scenario result above is based on this test case.
Answer generated when the material does not support onePassEvidence
“How much paid parental leave does Fenwake give?”
I don't have any information in the employee handbook or other provided sources about a paid parental leave policy at Fenwake Instruments Ltd. The handbook covers annual leave, sickness absence, expenses, notice periods, and training, but doesn't mention parental leave specifically. If you need details on parental leave entitlements, I'd recommend checking with HR directly, as this may be covered under statutory UK parental leave provisions or a separate policy document not included here.
Text capture: the answer and all returned passages
Copied from Output file (text), lines 10–18 · an excerpt
Open the whole text file ↗Source excerpt · partial file · response 1
status: 200 | passages returned: 10 --- answer (verbatim) --- I don't have any information in the employee handbook or other provided sources about a paid parental leave policy at Fenwake Instruments Ltd. The handbook covers annual leave, sickness absence, expenses, notice periods, and training, but doesn't mention parental leave specifically. If you need details on parental leave entitlements, I'd recommend checking with HR directly, as this may be covered under statutory UK parental leave provisions or a separate policy document not included here.
Copied from Output file (text), lines 10–18 · an excerpt
| ✓ | Expected: The answer should say the material does not cover paid parental leave.Found: It said it had no information, gave no leave figure or policy, and pointed to HR and statutory provisions. |
The answer had to decline if the material did not support a parental-leave policy, and it did so: no figure or policy was given, and the returned passages do not mention parental leave.
Control scores. Group B's top score was 7.76, below Group A's 18.28 on the same corpus.
Configuration and setup
How this tool was set up for the run and what the test needed in place. Each row is a fact from the run's records; a fact the records do not hold is left out, not guessed.
How this scenario is graded
How we decide Pass, Fail and Not gradable. The same rules apply to every tool tested on this scenario.
Each test case gets one result per tool: Pass, Fail or Not gradable. A test case we haven't run yet shows Untested. There are no partial results.
- Pass: the tool did everything the test expected, and nothing it said contradicts the correct answer.
- Fail: at least one expected behaviour clearly didn't happen; the test case's result says which one and quotes the tool.
- Not gradable: our evidence couldn't settle the outcome (for example a record we needed is missing). It is never counted as a fail, and the test case's result says what's missing.
Where this sits in the benchmark
This page is one cell of a larger study: one tool, one scenario. Only this benchmark's frame appears here.
| Level | Name | Scope |
|---|---|---|
| Benchmark | Managed RAG → | 15 scenarios · 9 tools |
| Capability | Grounded Generation → | |
| Scenario | The retrieved material doesn't support an answer → | |
| Tool | Denser AI → |
Global scenario definition → · Global capability definition → · Denser AI product page →
History of this result
What has happened to this result since it was first published. Runs and grades are never overwritten: a retest or a re-grade publishes a new result and keeps the earlier one readable.
Act on this result
Nothing filed here edits the run or the grade. A challenge opens a review, and a review can produce a new run or a re-grade — which becomes the current result and leaves this one in the history.
You run the same kind of test against your own setup and get the same behaviour.
Agree →Yours behaves differently. Tell us what you got, with a screenshot if you have one.
Disagree →Something here is wrong — a reference value, a transcription, a grade.
Report an issue →Tell us what changed and we schedule a rerun of the failing test case. The old result stays as history.
Vendor notice →The same record is available as structured data through the AI Demos MCP server, with the counts, the coverage and every per-test-case reason carried as fields.
Verify the proof files
These files support this result. Open a file to inspect the original evidence.
File fingerprints (SHA-256)
A fingerprint identifies the exact file used for this result.