C26 — Content & media (Medium)
- Context:
any→/ - Chat id:
uat-comparison-openai-original-uat-2026-10-05T11-14-39-C26 - Script: What reports has Unleash published?
- Expected: Lists available reports with download links where they exist.
- Pass criteria: Reports real; links resolve.
Verdict: FAIL
The agent was asked for 'reports' (downloadable research publications) but returned a list of news/analysis articles instead. The pass criteria require that reports are real and links resolve. The agent did not provide any actual downloadable reports with verified download links — it listed editorial articles and news items. There is no evidence these are formal published reports, and no download links are provided. The agent conflated news/analysis content with reports, failing to satisfy the expected behaviour of listing available reports with download links.
Missing capabilities:
- Report library or downloadable research publications catalogue with working download links
Webhook submissions
None captured.
Transcript
Turn 1 (first token 2643ms, total 3784ms)
Visitor: What reports has Unleash published?
Agent: Here are some of the latest reports-related pieces we’ve published:
- Organizational readiness, not employee resistance to change, is the biggest barrier to AI success — 2026-02-20
- Growth in 2026 won't happen without businesses prioritizing worker trust and skills — 2026-01-28
- Professor Amy Edmondson talks psychological safety at Workhuman Live Forum London 2025 — 2025-03-28
- HR's AI ship has no a crew or compass, says UST — 2024-08-20
- 1 in 4 don't trust their employer will implement AI responsibly — 2024-01-12
If you want, I can also narrow this to a topic like AI, skills, or future of work.