AI Demos Research — the structured-intelligence platform. Every verdict on these pages opens to the execution behind it.
Graded 31 October 2026

Can Anomaly AI understand follow-up questions using the context of its previous answer?

Anomaly AI kept the follow-up tied to the previous answer's five customers. It answered turn 2 without re-asking who they were, and the reply opened with the same set while the visible query fragment reused the earlier cancelled-order rule.

1 of 1 test case passed

Every test case in this scenario has a result.

Pass rate100%1 of 1 gradable
Coverage1 of 1test case with a result
1PassDid everything it was expected to do.
0FailNo test case failed.
0Not gradableEvery result here could be graded.
0UntestedEvery test case in this scenario has a result.
The pass rate is a summary. The evidence is the test case below: what we sent, what we checked, what the tool returned, and the proof.

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.

"They" refers to the previous answer's setPassEvidence
What we sent

Input 1 · Top customers by total spend: “Who are our top 5 customers by total spend?”

Input 2 · 90-day order count follow-up: “How many orders did they place in the last 90 days?”

What the tool returned

Spend basis and caveats

This is all-time spend in USD, excluding cancelled orders and before refunds. It includes unfulfilled orders but excludes membership and subscription charges. Data last synced October 30, 2026.

Orders placed by the same top 5

For the same top 5 customers, orders placed in the last 90 days (August 3-October 31, 2026) were: | Customer | Orders placed | Dev Raghavan 18 | Priya Raghavan 16 | Rosa Delgado 18 | Jae-won Park 13 | Cassandra Pope 18

Order-count basis and caveats

Counts include cancelled orders and use your local timezone. Data last synced October 30, so newer orders may not yet be included.

Transcribed from Proof 2 · Screen recording at 0:19

Every message the tool sent, in order (4)
  1. Top 5 customers by spend

    Your top 5 customers by total order spend are: | Rank | Customer | Total spend | 1 Dev Raghavan $21,164.27 | 2 Priya Raghavan $17,202.51 | 3 Rosa Delgado $14,659.75 | 4 Jae-won Park $13,646.96 | 5 Cassandra Pope $13,106.87

  2. Spend basis and caveats

    Printed above.

  3. Orders placed by the same top 5

    The final response, printed above.

  4. Order-count basis and caveats

    The final response, printed above.

Expected vs. Found
✓Expected: Turn 2 uses turn 1's set without changing it.Found: It answered for the same five customers from turn 1 and did not re-ask who they were.
Supporting proof
Proof 1Screen recording
Screen recording: turn 1's top-spend ranking and answer
Key moments
Open original ↗
Proof 2Screen recording
Screen recording: the follow-up answer for the same five customers
Key moments
Open original ↗
Why this result

The follow-up used the same five customers from turn 1, opened with the same-set wording, and did not ask for clarification. The visible query fragment also carried the earlier cancelled-order rule, so the set stayed the same instead of drifting.

Also observed in this test case

Different basis. The two turns used different cancelled-order rules: the first answer excluded cancelled orders from spend, while the follow-up counted them for the 90-day totals.

Tested by Ayush Ghosh · evidence dated 31 October 2026

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.

Software that produced the output
Anomaly AI
Build or version
Not exposed by the hosted product
Plan or tier
Pro
Surface
Web app
Set up before the run
The first answer had to return a small, stable set so the follow-up could refer to it exactly.
Tested
31 October 2026 · Ayush Ghosh

How this scenario is graded

How we decide Pass, Fail and Not gradable. The same rules apply to every tool tested on this scenario.

How results are decided

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.

The rules
  • 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.

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.

from the publication record
16 November 2026First publishedAI Database Agents v1

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.

This matches what I see

You run the same kind of test against your own setup and get the same behaviour.

Agree →
This does not match

Yours behaves differently. Tell us what you got, with a screenshot if you have one.

Disagree →
Point out an issue

Something here is wrong — a reference value, a transcription, a grade.

Report an issue →
Request a retest

On a newer build, a larger dataset, or your own setup.

Request a retest →
We have fixed this

Tell us what changed and we schedule a rerun of the failing test case. The old result stays as history.

Vendor notice →
Filed against this evidenceNothing yet. Challenges, counter-evidence and fix notices appear here with their outcome, and stay on the page after they are resolved.
Cite this result
aidemos.com/benchmarks/ai-database-agents/results/anomaly-ai/the-next-question-refers-to-the-previous-answer · 1 pass · 0 fail · coverage 1/1 · graded 2026-10-31

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.

Proof 1 · Screen recordingaca047ebfd3ef73f170d501a22f73addbfce494737e119d043e42719fc9415b0
Proof 2 · Screen recording1b05b9e758d2eeab3bc985bfda705539dae562d7b861611f6856d05e51919d19