Composite study

This is a product-class write-up, not a named-client story and not a results press release. Counts below describe the pack we actually write (approach points, tools, layers) — not invented bug totals or conversion lifts.

4
Approach points
4
Tools in the pack
4
Layers covered
3
Challenges documented

Overview

Summaries included quotation marks around paraphrases. Video captions named a speaker who was not on screen.

Scheduled posts went live from a draft slug that was never gated. The model’s SEO title for a draft was indexable.

Industry pattern this is grounded in

Publishers already fight training-data and attribution disputes. Invented quotes and captions presented as if they were in the asset are an editorial-standards failure, not a “creative” feature.

Highlights

Typical launch bar: Wait until journeys and evals agree — do not ship on demo scores alone.

Challenges

01

Editors liked punchy quotes more than accurate ones.

02

CDN cached a draft HTML once it had been fetched with a preview token.

03

Rights metadata was in a DAM the retriever did not see, so the assistant cleared a clip for social that was embargoed.

How we tested

01

Quoted text must be a substring of the source (or marked paraphrase).

02

Captions labeled generated; people tags from a verified roster or omitted.

03

Logged-out request to draft slug is 404/401.

04

Paywall/entitlement: the assistant must not summarize behind the paywall for anonymous users.

Starting point vs what we required

Starting point versus the controls we required
Starting pointWhat we required
Editors liked punchy quotes more than accurate ones.Quoted text must be a substring of the source (or marked paraphrase).
CDN cached a draft HTML once it had been fetched with a preview token.Captions labeled generated; people tags from a verified roster or omitted.
Rights metadata was in a DAM the retriever did not see, so the assistant cleared a clip for social that was embargoed.Logged-out request to draft slug is 404/401.

What we delivered

01

Journeys / UI

  • Playwright — Draft leak, paywall, publish pipeline
02

RAG quality

  • Ragas — Summary faithfulness
03

Adversarial evals

  • Promptfoo — Invented quote probes
04

Front-end performance

  • Lighthouse — Popular article start time — secondary to leak tests

Outcomes

Quote-substring evals and draft-URL tests were launch-blocking. Entitlement applied to the summarize API, not only the article page.

Embargo/rights became a tool check before social scheduling.

Stack

Testing stack used on this product class
LayerToolWhat we validated
Journeys / UIPlaywrightDraft leak, paywall, publish pipeline
RAG qualityRagasSummary faithfulness
Adversarial evalsPromptfooInvented quote probes
Front-end performanceLighthousePopular article start time — secondary to leak tests

Same method as the rest of our work: severity-ranked findings, pass rules in writing, and a ship / wait / block call. See the evaluation method and tool comparison.