United Kingdom, Swansea
Swansea University
DRAGON-Spotter (AI linguistic detection of online grooming in seized-device chat logs)
In developmentOfficial publicationPAI-4251808F
- Body
- Swansea University
- Country
- United Kingdom
- Area
- Swansea
- Technology
- DRAGON-Spotter (AI linguistic detection of online grooming in seized-device chat logs)
- Supplier
- Not stated in the source
- Status
- In development
Being researched, built, tested or marketed for police use. No police body has confirmed using it. - Stage
- On the horizon
- Since
- Not known
- Legal basis
- Not stated in the source
- Oversight
- Not stated in the source
- Evidence
- Official publication
Police force, government department or public body publication - Source
- Police STAR Fund project outcome report, DRAGON-Spotter (Police Science website)
- Last checked
- Record code
PAI-4251808F
Notes
DRAGON-Spotter, from the DRAGON-S project (Developing Resistance Against Grooming Online), is a tool that uses AI and linguistics to detect grooming language in chat logs from seized devices, to speed up digital forensic work in child sexual exploitation investigations. A Police STAR Fund project outcome report on the Police Science website says the tool was developed by Swansea University's DRAGON team with West Mercia Police, was tested on operational datasets (processing more than 28,000 messages in 10 hours) and integrated into digital forensic workflows, and used a 7-billion-parameter Mistral model to test feasibility. A follow-on bid to the 2024 to 2025 Police STAR Fund, which would have set up a virtual lab for real-world use, was shortlisted but unsuccessful, and the report says further adaptations are needed. Police Science magazine (issue 2, summer 2025) lists it among West Mercia Police projects, developed in partnership with Swansea University through Police STAR funding. No source read shows operational use by any force.
DRAGON-Spotter (AI linguistic detection of online grooming in seized-device chat logs) elsewhere in United Kingdom
See them in the tracker| Deployment | Country | Supplier | Since | Status | Checked |
|---|---|---|---|---|---|
| Serious Fraud OfficeAI-assisted document review and technology assisted review | United Kingdom | OpenText (Axcelerate, per the SFO's 2018 announcement) | 2018 | Operational |