# Finding 050: Copilot temporary chat did not fetch the hidden structured-data conflict fixture

## Date

2026-07-01

## Status

Published

## Summary

Copilot/Bing was run against p21,
`manual-client-copilot-bing-20260626-001-p21`, in a fresh Copilot chat. The
exact public target URL was reachable before prompt submission and contained
the expected visible code `VISIBLE-PLUM-47`, meta-description code
`META-AMBER-16`, and JSON-LD code `SCHEMA-INDIGO-82`.

Copilot returned `fetched:false`, `pages_opened:0`, and said the client could
not open or fetch the target URL. The bounded origin review found no matching
direct-origin event for the exact attempt id, the
`/lab/reading/hidden-structured-data-conflict` fixture path, `/robots.txt`,
root, or Copilot/Bing-related traffic during the prompt window.

## What does this mean?

For site owners and researchers, this run shows a clean Copilot no-retrieval outcome for a public page that was reachable and contained visible, metadata, and JSON-LD signals. Copilot did not return page content, and the site's logs did not show a matching Copilot or Bing request, so there is no evidence that either the visible text or hidden structured data was inspected in this attempt.

## Method

- Browser task:
  `research/manual-client-runs/browser-tasks/manual-client-copilot-bing-20260626-001-p21.browser-task.json`
- Prompt packet:
  `research/manual-client-runs/manual-client-copilot-bing-20260626-001.prompts.json`
- Answer artifact:
  `research/manual-client-runs/manual-client-copilot-bing-20260626-001.answers.json`
- Response file:
  `research/manual-client-runs/browser-tasks/responses/manual-client-copilot-bing-20260626-001-p21.response.json`

Before submitting the prompt, the exact target URL was checked with `curl`.
The response returned `HTTP 200` and contained `VISIBLE-PLUM-47`,
`META-AMBER-16`, and `SCHEMA-INDIGO-82`.

Copilot/Bing was opened at `https://copilot.microsoft.com/` in the
OpenClaw-controlled Chrome profile. The UI showed the `kai` account, Free
Plan, a fresh start screen, a visible native `Temporary` control, and response
mode `Smart` before prompt submission. The prompt was submitted in a fresh
chat at `/chats/6gszf8F23x91iJ4DXihSm`, not an old conversation.

Direct-origin review checked `data/events.json` for the exact attempt id, the
fixture path, `/robots.txt`, root, and Copilot/Bing-related events during the
bounded prompt window.

## Result

| Field | Value |
|---|---|
| Attempt ID | manual-client-copilot-bing-20260626-001-p21 |
| Client | Copilot/Bing (copilot-bing-web) |
| Visible model/context | kai Free Plan, Temporary control visible, Smart response mode |
| Source prompt | reading-hidden-structured-data-conflict |
| Fetched | false |
| Pages opened | 0 |
| Direct-origin events | 0 inside prompt window |
| Raw event ids | none |
| Bounded window | 2026-07-01T01:17:53Z - 2026-07-01T01:18:44Z |
| Confirmed hit from prompt | false |
| Confirmation status | registered_no_hit |

## Raw response

```json
{
  "attempt_id": "manual-client-copilot-bing-20260626-001-p21",
  "source_prompt_id": "reading-hidden-structured-data-conflict",
  "prompt_code": "manual-client-copilot-bing-20260626-001-p21",
  "ai_system": "Copilot/Bing",
  "fetched": false,
  "pages_opened": 0,
  "answer": "This client cannot open or fetch the target URL. No page content, meta description, or JSON-LD could be retrieved.",
  "evidence_quote": "Client does not support direct web browsing or page retrieval.",
  "limitations": "Unable to access external URLs; cannot inspect visible or hidden page code.",
  "confidence": "0.78"
}
```

## Origin evidence

```json
{
  "timestampWindow": {
    "startedAt": "2026-07-01T01:17:53Z",
    "endedAt": "2026-07-01T01:18:44Z"
  },
  "sourcePromptId": "reading-hidden-structured-data-conflict",
  "promptCode": "manual-client-copilot-bing-20260626-001-p21",
  "confirmedHitFromPrompt": false,
  "rawEventIds": [],
  "events": [],
  "preflightSummary": "Operator curl preflight at 2026-07-01T01:14:19.174Z returned HTTP 200 for the exact target and found VISIBLE-PLUM-47, META-AMBER-16, and SCHEMA-INDIGO-82. This preflight is fixture-readiness evidence, not prompt-caused behavior.",
  "resourceCounts": {
    "htmlPageRequests": 0,
    "robotsRequests": 0,
    "rootRequests": 0,
    "trackingPixelFetches": 0,
    "nonPixelSubresources": 0,
    "clientCapabilityEvents": 0
  }
}
```

## Interpretation

This run does not show Copilot/Bing reading the p21 target page. The response
reported that the target URL could not be opened or fetched, and the bounded
origin logs had no exact attempt id, fixture-path request, `/robots.txt`
request, root request, or Copilot/Bing-related event during the prompt window.

This is the same outcome shape as Finding 045 for p20, but on the p21 fixture
whose visible text no longer announces a structured-data conflict. Because no
page content was retrieved, the run cannot compare Copilot's treatment of the
visible page code, meta description, or JSON-LD.

## Limitations

- This finding covers one Copilot/Bing run, one account/session, one fresh
  Copilot chat, and one fixture.
- The prompt supplied the exact target URL, so the run tests direct opening,
  not independent discovery.
- The Copilot UI showed response mode `Smart`, but not a precise model version
  in the captured snapshot.
- The visible native `Temporary` control was present on the fresh start
  screen, but this finding does not prove Microsoft-side retention semantics.
- The recorded start and end times are operator-side bounds around submission
  and final answer observation, not service-internal fetch timestamps.
- Origin review used local `data/events.json` during the bounded window; it
  would not capture a delayed target fetch outside that window.

## Publication Thesis Verification

- Thesis: Copilot/Bing did not fetch the p21 hidden structured-data conflict
  fixture during this controlled-browser attempt and left no prompt-window
  origin activity.
- Source: Fresh Copilot response, generated response and answer artifacts,
  browser-task artifact, preflight output, and bounded `data/events.json`
  review.
- Method: Exact public target preflight, controlled-browser use of a fresh
  Copilot temporary chat, prompt submission, exact attempt-id review,
  fixture-path review, `/robots.txt` review, root-path review, Copilot/Bing
  term review, and bounded timestamp-window correlation.
- Bias: Single run, Free Plan, Smart response mode, and Copilot's retrieval
  implementation may vary by account, region, product surface, or time.
- Consensus: Consistent with Finding 045, where Copilot/Bing also returned a
  no-fetch structured response and no matching origin event for the p20
  structured-data conflict fixture.
- Invalidation: A raw event for the exact attempt id, fixture path,
  `/robots.txt`, root, or Copilot/Bing-origin traffic inside the same prompt
  window; a response artifact from the same attempt showing retrieved page
  content; or a fixture/preflight mismatch would weaken this result.
- Verdict: Supported for this run. The model response, preflighted fixture,
  and empty bounded origin review align with a clean no-hit.
- Additional tests suggested: compare the completed p20 and p21 client sets to
  see whether removing the visible conflict cue changed hidden metadata or
  JSON-LD reporting across clients.

## Next steps

- Compare the complete p21 client set against p20 across ChatGPT, Claude,
  Gemini, Perplexity, and Copilot/Bing.
- Keep rerun candidates for Copilot if a clearer retrieval/search mode or
  model version control appears.