Kunshan police AI flags gambling ring and freezes fraud transfers

AI system

System: Kunshan Public Security Bureau 'AI police' (AI警察): more than 220 'main-attack, main-defense' intelligent models built by the Kunpeng Squad
Model: Not assessed
Release: Not assessed
Application developer: Kunshan Public Security Bureau, Kunpeng Squad (鲲鹏战队), with Suzhou Public Security Bureau support
Role: Not assessed
Configuration: Not assessed
Inputs and outputs: Not assessed
Autonomy: Not assessed
Workflow: Bulk police data on gambling and payment behavior that models can score for anomalies. A dedicated squad of officers who build and run models inside the bureau. Bank cooperation that lets a traced account be frozen within minutes.
Human oversight: Not assessed
Relation to AI: Both accounts are the police's. They credit the models with flagging the outdoor casino from behavioral features in bulk data and mapping the ring's structure “with one click”. They also traced card-tier fund flows in parallel, so a freeze came within 10 minutes. Officers made the arrests and freeze requests. Deputy criminal-investigation chief He Yongliang describes the gain as 24-hour, parallel rather than serial casework. Neither source separates the models' share from ordinary investigation.
AI contribution: Both accounts are the police's. They credit the models with flagging the outdoor casino from behavioral features in bulk data and mapping the ring's structure “with one click”. They also traced card-tier fund flows in parallel, so a freeze came within 10 minutes. Officers made the arrests and freeze requests. Deputy criminal-investigation chief He Yongliang describes the gain as 24-hour, parallel rather than serial casework. Neither source separates the models' share from ordinary investigation.

Benefits

Outcome identifiers: Not assessed
Benefit: traced three tiers of card transfers and froze 500,000 yuan within 10 minutes
Category: Reduced Risk
Status: Not assessed
Beneficiary groups: People. Organizations. Society.
Beneficiaries: Not assessed
Reach: Not assessed
Distribution and access: Not assessed
Duration: Not assessed
Expected?: Yes
Costs and harms: Not assessed

Measurement

Studies: Not assessed
Study design: Not assessed
Measurement: Not assessed
Setting and sample: Not assessed
Measures: traced three tiers of card transfers and froze 500,000 yuan within 10 minutes
Comparison: Not assessed
Results: traced three tiers of card transfers and froze 500,000 yuan within 10 minutes
Additional value from AI: Not assessed
Limitations: Not assessed
Denominators: Not assessed
Materials for replication: Not assessed

Money

Kind of value: Not assessed
Amount: Not assessed
Currency and prices: Not assessed
Period and unit: Not assessed
Realization: Not assessed
Recipients and payers: Not assessed
Gross or net: Not assessed
Included costs: Not assessed
Comparison: Not assessed
Method: Not assessed
Return and payback: Not assessed

People and places

Participants: Developer: Kunshan Public Security Bureau, Kunpeng Squad (鲲鹏战队), with Suzhou Public Security Bureau support. Operator: Kunshan Public Security Bureau.
Jurisdictions: Benefit: Kunshan, China. Developer: China. Operator: China.
Ownership and control: Not assessed
Dependencies: Not assessed
Cross-border connections: Not assessed
Geopolitical implications: Not assessed

Follow-up

Conditions for replication: Officers verify each model flag and make the arrests and freeze requests themselves. Figures are the deployer's own; an outside audit would be needed to confirm them.
Replication attempts: Not assessed
Later findings: Not assessed
Availability: Not assessed
Institutional decision: Not assessed
Grounds: Not assessed
Action and result: Not assessed
Success criterion: Not assessed

Evidence

Source numbers: AIBD-0380-source-0001–0003
Citations: Numbered references below.
Language: Not assessed
Source relationships: Independent, Operator
Original evidence: Not assessed
Access and versions: Not assessed
Supported findings: Not assessed
Evidence links: Not assessed
Evidence strength: Not assessed
Missing evidence: Not assessed

Grades

Assessed benefits: People. Organizations. Society.
Depth: People: 2. Organizations: 2. Society: Not assessed.
Reasons: People: The fraud victim who lost 980,000 yuan had 500,000 yuan frozen within 10 minutes and 200,000 yuan recovered after the arrests, per the police. Whether the frozen sum was returned is not reported. Organizations: The Kunshan bureau reports a 20-plus-suspect bust, nine fraud arrests in three days and 609 fraud cases solved in six months, recovery rates up 62%. A measurable gain for the unit, on its own figures. Society: The six-month aggregate of 32.47 million yuan recovered is the deployer's own claim with no method or independent check, so no population effect is established.
Observed or intended: Not assessed
Rubric: Not assessed
Assessor and date: Not assessed
Alternative assessments: Not assessed
Fill: Independently reported

Entry details

Kind: Not assessed
Related entries: Not assessed
Field: Law enforcement / criminal investigation. Gambling rings and telecom fraud.
Purpose: Models built by the Kunshan police to detect crime patterns and trace funds for investigators. The bust and the freeze were that purpose.
Dates: Date basis unassessed: 2024 (precision unassessed). Updated: 21 October 2026.
Stage: Not assessed
Version: SHA-256: 8c7fdbc00abd…

References

Other identifiers: oecd_aim: 2024-10-03-edf0.
Category mappings: Not assessed
Brief: Not assessed
Revisions: 3 revisions. Latest: 21 October 2026.

Sources

News report Independent author

Ji, Y. (2024). 江苏昆山公安：“AI警察”上岗 “算力”引爆“战力”. Modern Express (现代快报) via NetEase, October 6, 2024.
News report Deployer

Suzhou Daily (2024). 昆山警方释放出“专业+机制+大数据”新型警务运行模式强大新质战力. Suzhou Municipal People's Government website, October 7, 2024.
Registry label Independent author

OECD.AI (2024). AI Police Deployed in Kunshan, China, Lead to Major Crime Busts and Fraud Prevention. OECD AI Incidents and Hazards Monitor, 2024-10-03.