Can Anomaly AI show what people have been asking?
Anomaly AI showed the questions asked across two accounts and attributed them to the right account. The owner's usage view contained five insights for five questions: three from the owner and two from the member.
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.
Usage view across two accountsPassEvidence
Input 1 · Owner account: “As of September 2, 2026, what's our total revenue in the last 30 days?”
Input 2 · Owner account: “What's our most common payment method?”
Input 3 · Owner account: “How many complaints have we received on social media this month?”
Input 4 · Member CodiumX: “How many customers currently have an active subscription?”
Input 5 · Member CodiumX: “What's our net promoter score for last quarter?”
Response 1 answers Input 1.
Response 2 answers Input 2.
Response 3 answers Input 3.
Response 4 answers Input 4.
Response 5, the final response, answers Input 5.
Net promoter score cannot be calculated for last quarter (July–September 2026): the available data contains no recommendation survey ratings.
Transcribed from Proof 1 · Screen recording at 1:10
Every message the tool sent, in order (5)
- Answer
Total booked revenue was $8,515.84 USD for August 4–September 2, 2026, inclusive
- Answer
Card is the most common payment method, used for 162 orders (52.6% of all orders).
- Answer
The number of social media complaints this month cannot be determined from the available data.
- Answer
5 customers have a subscription recorded as active.
- Answer
The final response, printed above.
| ✓ | Expected: The questions appear with the account that asked each one, and their total matches the number asked.Found: Five insights appeared for five questions. In the owner's view, three were attributed to the owner and two to the member, consistent with who asked them. |
The requirement was to show the questions with their account attribution, not to use a particular plan. The owner's usage view showed all five questions with the correct attribution, so the result is a pass.
Trial workspace badge. The workspace switcher displayed a trial badge, while the plan used was Pro.
Questions appear as titles. The payment-method question appeared as an AI-generated insight title rather than the question as typed.
Attribution follows the viewer. The member's net promoter score insight was marked “Edited by you” in the member's view; the owner's view attributed it to the member.
Unavailable answers remain distinct. The social media complaints and net promoter score insights said the answers could not be determined, rather than reporting zero.
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 | AI Database Agents → | 28 scenarios · 11 tools |
| Capability | Analytics & Observability → | |
| Scenario | Someone needs to see what people have been asking → | |
| Tool | Anomaly AI → |
Global scenario definition → · Global capability definition → · Anomaly 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.