Catalog record · Operations & Efficiency
Stage as of May 14, 2025 · historical report, not live reconfirmation.
Consult thematic analysis of public consultation responses
Primary workflow source. A named organization's workflow supported by a reviewed primary source. Stage is reported separately; this label does not establish measured outcomes or independent validation.
Last researched Oct 26, 2026. Source access and reported stage are recorded below; deployment evidence does not establish ROI.
What the research supports
Consult used generative AI to identify and assign themes to more than 2,000 Scottish consultation responses, with expert theme refinement and manual comparison.
- Reported stage
- Pilot
- Technology
- GenAI/LLM
- Business function
- Back Office
- Impact (legacy label)
- Not labeled
- Catalog date · unconfirmed
- Not reported
- Imported
- Oct 29, 2026
- Last researched
- Oct 26, 2026
Measures & evidence
F1 0.76 in first live evaluation
Original metric claim
Measurement window not disclosed
Scope, baseline & source
- Scope / denominator
- Measurement population / workflow scope not disclosed
- Baseline
- Not disclosed in this account
- Claim basis
- Agreement measure, not autonomous accuracy or economic value.
Source-reported claim; methodology, baseline and independent validation may be undisclosed. Inspect the research dossier.
Original prose retained. A single comparable value or unit has not been established.
Read dossier source ↗Exact claim-to-source location has not been mapped. This link belongs to the research dossier.
Measures retain their original currency, scope and period. Capacity, adoption and forecasts are not realized cash savings.
Documented vendor relationships
No vendor relationship has been linked to this specific workflow in the curated overlay. This does not mean it was built without vendors.
See the company’s documented relationshipsDeployment evidence & limits
May 2025 issuer report explicitly says the live use remains in trial pending evaluation. Status as of 2025-05-14.
Stage is as reported in the cited evidence. Current operating status has not been independently reconfirmed; the research date is an editorial review date.
The initial trial manually reviewed every response; the rollout savings projection therefore is not a measured production saving. July 2026 indexed primary entry still describes limited live testing and planned wider rollout in 2027; its full page and evaluation PDF could not be fetched. Rank agreement can hide rare critical complaints. Keep case traceability and escalation rules; projected government savings are not realized FS savings.
Potential FS application · inferred
Inferred FS application: Customer-complaint theme analysis, conduct-risk feedback review and regulatory-consultation synthesis.
Human-reviewed clustering and traceable response-to-theme mapping transfer to large text collections.
Original industry: Public policy administration. Transfer limits: Rank agreement can hide rare critical complaints. Keep case traceability and escalation rules; projected government savings are not realized FS savings.
Source register
- Government-built “Humphrey” AI tool reviews responses to consultation for first time, in bid to save millionsgov.uk · opens in a new tab
Department for Science, Innovation and Technology · issuer · full-text · Published 2025-05-14
- Consultknowledgehub.dev.ai.gov.uk · opens in a new tab
AI Knowledge Hub · issuer · indexed-primary · Published 2026-07-23
Research lanes: public_scientific · Batch global-discovery-2026-10-26. Coverage and remaining gaps