WorkThe story, in brief
Operation “Wrong Number”: AI-assisted task scam
OpenAI disrupted a Cambodia-based scam ring using AI to target UK victims—showing how frontier models are weaponized faster than guardrails can respond.
The KeyNews take
Why it matters
AI-enabled fraud is scaling internationally; practitioners deploying generative AI need to understand emerging attack vectors and platform accountability for abuse.
The key facts
10 to knowOpenAI banned accounts originating in Cambodia
Scam operation used AI to support task workflows
UK victims targeted
Operation codenamed 'Wrong Number'
Demonstrates real-world AI misuse at scale
Platform enforcement action taken
Scam workflow was AI-assisted (LLM use in fraud automation)
Targets: UK consumers
OpenAI published disruption report with methodology
Illustrates AI-amplified crime and platform responsibility
Go to the source
OpenAI Blogopenai.com
Publisher excerpt: OpenAI banned accounts that likely originated in Cambodia and used AI to support scam workflows targeting people in the UK.