AI Risk Wire

Cases › produced harmful content › LR-2026-0774

Police officer arrested in Seoul over AI-generated sexual deepfakes

Confirmed Why: 2 independent outlets · 2 independent reports

Status: action reported, no fix

Severity moderate
Reported 4 Oct 2026

No source states: loss or records, developer and model, agent tool, could do, deployer, industry.

Severity moderate: Tier moderate: the test is a bounded scope stated, here more than 400 deepfakes depicting acquaintances of the officer, with data of others touched.

A serving Seoul police officer was arrested on Sept. 16 on charges including creating and distributing sexually explicit deepfakes. Police allege he used AI applications to make such images of people he knew and shared them on Telegram. The case came to light through an undercover investigation by Busan police into a Telegram chat room where such images were circulating. Investigators found about 5,600 images and videos in his possession, including roughly 2,600 deepfakes. Police allege more than 400 deepfakes were created between March 2022 and July 2026, and that the conduct began about two years before he joined the police in April 2024. The officer was relieved of duties, and reportedly confessed after the investigation intensified.

What happened

1 · What it was doingA serving officer allegedly used AI applications to make sexual deepfakes of people he knew and shared them on Telegram.
2 · What went wrongPolice allege more than 400 sexually explicit deepfakes were created using AI applications.
3 · What it reached or didInvestigators found about 5,600 images and videos in his possession.
4 · How it was caughtThe case came to light through an undercover Busan police investigation of a Telegram chat room.

What was done

The officer was placed on standby and has since been relieved of duties.

Risks

Privacy harm to a person
Primary41 incidents

Read from the sources; not yet reviewed by a person.

Controls

Not yet read for controls. Controls that mitigate the primary risk:

Application

Data minimisation · Sensitive-data checks on outbound data · Human approval for high-impact actions · Output filtering and redaction · People told they are dealing with AI · Risk assessment per deployment

Infrastructure / target

Bot and scraping defences

Similar incidents

Students caught wearing or possessing AI smart glasses during Seoul university essay examinations2026-10-30
Yoido Full Gospel Church (Seoul) customer and member database breach, 850,000 members potentially affected, AI sub-agent traces found2026-10-07

All cases of this kind ›

Evidence

No source states: when it happened, when it was found, how many records were exposed, the direct loss, a count of the other parties affected, any court case, the cause, what set it off.

Show the full account: 4 items from the primary sources

What happened, in order

  1. 2024-04 Officer joins the Seoul police force.
  2. 2025-12 Court approval obtained for undercover investigation of the Telegram chat room.

Not dated in the sources:

  • Officer placed on standby duty after the station is notified of the Busan investigation.
  • Officer reportedly confesses to the allegations.
Show every field, with its source and grade

What happened

Earliest source we hold
2026-10-04
D inferred by us from the described behaviour, reason shown
Who caught it
an outside party
C one secondary source states it
What went wrong
produced harmful content
C one secondary source states it
primary risk
RK15
D inferred by us from the described behaviour, reason shown

Impact

Severity
moderate
D inferred by us from the described behaviour, reason shown
Kind of harm
privacy
C one secondary source states it
Why this severity
Tier moderate: the test is a bounded scope stated, here more than 400 deepfakes depicting acquaintances of the officer, with data of others touched.
D inferred by us from the described behaviour, reason shown

Notes on gaps

The sources report allegations, not proven facts. Standby timing is not dated and is placed approximately in the timeline. No AI-system control or provider remediation is reported.

LR-2026-0774 · version 1 · published 2026-11-01 after automatic checks; not yet checked field by field by a person · JSON