Does Anomaly AI suggest useful follow-up questions after an answer?
Anomaly AI suggested a relevant follow-up after answering a question. For an August revenue answer, it offered a July comparison that the same database could support.
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.
Suggested next questions after an answerPassEvidence
A calculation had been saved earlier in the project.
“As of September 2, 2026, what was our total revenue last month?”
Your total revenue for August 2026 was $8,795.45 USD, as of the end of September 2, 2026. This is booked revenue before refunds, excluding orders cancelled by that cutoff. Historical order statuses are used; later changes to order amounts cannot be reconstructed.
Transcribed from Proof 1 · Screen recording at 0:15
Every message the tool sent, in order (2)
- Revenue answer
The final response, printed above.
- Generated SQL query · not part of the answer
SELECT strftime (t."month", '%B %Y') AS MONTH, SUM(t."total_amount") AS revenue FROM "monthly_revenue_asof_sep2_2026" t WHERE t."month" = DATE '2026-08-01' AND t."status_asof" <> 'cancelled' GROUP BY t."month"
| ✓ | Expected: At least one unprompted suggestion that follows from the preceding analysis and can be answered using the database.Found: After answering the August revenue question, it suggested comparing with July without prompting; the database contained July orders for the same measure. |
The July comparison was answerable from the same database and followed directly from the August answer, meeting the requirement. The query is clipped in chat, but the calculation panel shows it in full.
One data-specific suggestion. Only one follow-up was tied to the data; the other two were standard insight actions.
Revenue answer qualifications. The August answer described booked revenue before refunds, excluding orders cancelled by the cutoff. It said later changes to order amounts could not be reconstructed.
A saved calculation. For the August revenue answer, Anomaly AI reused a calculation saved earlier in the project.
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 | Conversational Interaction → | |
| Scenario | The agent offers what to ask next → | |
| 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.