Torswats used synthesized AI voices to sell swatting services targeting schools and homes
Torswats operated a paid swatting-as-a-service on Telegram using synthesized AI voices to make fake emergency calls. The service targeted schools and homes across the US, causing widespread panic and resource diversion.
Torswats weaponized computer-generated voices to scale the dangerous practice of swatting as a paid service.
Key facts
- What
- Torswats operated a paid swatting-as-a-service on Telegram using synthesized AI voices to make fake emergency calls.
- Incident date
- Feb 12, 2023
- Who
- Torswats
- Failure mode
- Tool Misuse
- AI surface
- Voice Agent
- Severity
- High
What happened
A Telegram-based entity known as Torswats sold a paid service to place fake emergency calls using computer-generated voices. The service targeted high schools and private residences across the United States, often reporting fake bomb threats or mass shootings. These actions led to numerous school lockdowns and the deployment of emergency police resources.
What broke inside the model
- 01 · TriggerThe agent selects the correct tool.
- 02 · Model stepIt fills the call with the wrong arguments.
- 03 · Control gapNo validation checks the arguments first.
- 04 · FailureThe tool runs against the wrong target.
- 05 · ConsequenceThe wrong record, account, or system is hit.
At the tool call, the arguments point at the wrong target.
The system used speech synthesis technology to create consistent, panicked-sounding personas. The operator could modify the synthesized voice in near real-time to respond to emergency dispatcher questions, effectively masking their identity while scaling the attacks.
What it cost
Sources
Cite this entry
https://failureindex.ai/failures/torswats-used-synthesized-voices-sell-swattingAI Failure Index. "Torswats used synthesized AI voices to sell swatting services targeting schools and homes" (FI-0659). Realm Labs. https://failureindex.ai/failures/torswats-used-synthesized-voices-sell-swatting (indexed Jun 22, 2026).Data fields CC-BY 4.0, prose citation permitted. Incident ID FI-0659. Full dataset at /data.
Note from Realm Labs, the Index steward
How Realm would have caught this
- OmniGuard
- AgentRealm
Realm can inspect a tool call against the user's actual intent before it runs, and hold calls whose arguments or target do not match what was asked, so the wrong tool or the wrong arguments never reach the system of record.