# Finding 094: Perplexity p44 repeated failed fetch with a clean origin no-hit

## Date

2026-07-03

## Status

Published

## Summary

Perplexity was prompted in a fresh native incognito thread with
`manual-client-perplexity-plain-head-signal-randomized-repeat-20260703-001-p44`,
the second randomized repeat of the plain site-owner review prompt against the
`/lab/reading/head-signal-isolation` fixture. The prompt asked Perplexity to
open the target URL and briefly review the page, without asking for source
areas, page-head metadata, structured data, code-like tokens, marker values,
or hidden fields.

Perplexity returned `fetched:false`, `pages_opened:0`, and evidence quote
`Failed to fetch url content`. A bounded origin review found no exact
target-page hit for the p44 URL and no related PerplexityBot `/`,
`/robots.txt`, or sitemap activity inside the prompt window.

## What does this mean?

For site owners, SEO/AEO teams, publishers, and researchers, this run shows that the same Perplexity prompt family can fail without leaving any matching origin trace at all. Compared with the immediately preceding p43 run, where the copied answer also said the page could not be fetched but the server still saw root and robots requests, p44 reinforces that answer text and server-side behavior have to be checked separately. A failed fetch answer is stable here, but the origin footprint is not.

## Method

- Controlled-browser task:
  `research/manual-client-runs/browser-tasks/manual-client-perplexity-plain-head-signal-randomized-repeat-20260703-001-p44.browser-task.json`.
- Prompt packet:
  `research/manual-client-runs/manual-client-perplexity-plain-head-signal-randomized-repeat-20260703-001.prompts.json`.
- Response artifact:
  `research/manual-client-runs/browser-tasks/responses/manual-client-perplexity-plain-head-signal-randomized-repeat-20260703-001-p44.response.json`.
- Logged answer packet:
  `research/manual-client-runs/manual-client-perplexity-plain-head-signal-randomized-repeat-20260703-001.answers.json`.
- Prompt window:
  `2026-07-03T02:15:23Z` to `2026-07-03T02:18:06Z`.

The prompt was submitted in a fresh Perplexity native incognito thread from
the logged-in OpenClaw-controlled Chrome profile. The UI showed Free plan,
Search enabled, and an `Exit incognito` control before submission. The browser
tool's `act` wrapper again rejected the stable target handle with
`action targetId must match request targetId`, so the prompt was typed and
submitted through the local `openclaw browser` CLI against the same labeled
fresh tab.

## Evidence

| Evidence | Result |
|---|---|
| Model answer | `fetched:false`, `pages_opened:0`; `Failed to fetch url content`. |
| Perplexity UI state | Fresh native incognito thread; visible account `kai bot`; Free plan; Search enabled; generic Model selector; Pro free preview notice after answering. |
| Direct-origin target-page event | None for `/lab/reading/head-signal-isolation?id=manual-client-perplexity-plain-head-signal-randomized-repeat-20260703-001-p44` inside the bounded prompt window. |
| Ancillary origin activity | None for PerplexityBot `/`, `/robots.txt`, sitemap, or the exact target path inside the bounded prompt window. |

```json
{
  "timestampWindow": {
    "startedAt": "2026-07-03T02:15:23Z",
    "endedAt": "2026-07-03T02:18:06Z"
  },
  "confirmedHitFromPrompt": false,
  "rawEventIds": [],
  "ancillaryOriginActivity": {
    "present": false,
    "paths": [],
    "rawEventIds": []
  }
}
```

## Interpretation

This p44 randomized repeat is a clean target no-hit. It repeats the
answer-side failed-fetch state from Perplexity p37, p38, and p43, but its
origin-side footprint matches p38 rather than p37 or p43: no exact target-page
hit and no related ancillary origin traffic appeared inside the bounded
prompt window.

The evidence supports a narrow conclusion: in this fresh Perplexity native
incognito run, the supplied target page was not fetched and the lab recorded
no related PerplexityBot origin contact during the prompt window. It does not
show that Perplexity cannot access the fixture generally, and it does not
support any visible or hidden marker interpretation because the exact target
page was not retrieved.

## Limitations

- This is one Perplexity run from one account, plan, client mode, and time
  window.
- The prompt supplied the exact target URL, so this measures direct URL
  opening rather than independent discovery.
- A clean origin no-hit is bounded to the prompt window and retained local
  event log; it cannot prove that no backend attempted retrieval outside the
  observable lab origin.
- The answer and origin evidence do not explain whether the failure came from
  product routing, retrieval backend state, robots handling, connection
  timing, or another client-side condition.
- No content-visibility claim can be made because there was no exact target
  hit.

## Publication Thesis Verification

- Thesis: In the p44 randomized plain head-signal repeat, Perplexity returned
  a failed-fetch answer, the lab recorded no exact target-page hit, and the
  lab recorded no related PerplexityBot ancillary origin activity inside the
  bounded prompt window.
- Source: The p44 browser-task artifact, response artifact, logged answer
  packet, Perplexity UI snapshot context, and local `data/events.json` review
  for the bounded prompt window.
- Method: Compare the copied Perplexity answer and visible browser state with
  the exact attempt id, prompt code, timestamp window, and local server-side
  events for the target path and nearby ancillary paths.
- Bias: Single-account, single-run evidence from a synthetic fixture. The
  prompt is plain relative to source-area prompts but still names
  AI-assistant readability and an exact lab URL.
- Consensus: Consistent with Findings 081, 084, and 093 on the copied
  failed-fetch answer and lack of exact target-page retrieval. It differs from
  Findings 081 and 093 because p44 did not record root or robots activity.
- Invalidation: A corrected origin review finding an exact target-page event
  or related PerplexityBot ancillary event inside the prompt window, a
  response-artifact mismatch, or a timestamp-window error would weaken this
  finding.
- Verdict: Supported for the p44 run. It documents a registered target no-hit
  with no related ancillary PerplexityBot origin activity.
- Additional tests suggested: compare p37, p38, p43, and p44 to separate
  Perplexity answer-side failed-fetch stability from origin-side activity
  variability.

## Next steps

- Publish a scoped Perplexity no-hit comparison across p37, p38, p43, and p44.
- Continue the sibling Copilot/Bing randomized p43/p44 controlled-browser
  tasks.