Cases › physical harm › LR-2026-0241
Garage-built ChatGPT-controlled smart rifle fires on voice command; OpenAI revokes engineer's access
Severity minor: no source gives a fact that decides the level; we set it from the description
Earliest source we hold 2026-10-23
An engineer built a voice-controlled smart gun in his garage whose voice commands use ChatGPT. He tells it “we’re under attack from the front left and front right, respond accordingly,” and the rig aims left and fires, then aims right and fires blanks, after which an AI voice offers further assistance. The book reports that this prompted OpenAI to cut off the engineer’s access. A week later he posted a video of a similar machine with the rifle replaced by a nailgun, instructing “ChatGPT, begin color tracking on the color black” so the rig shot black balloons while mostly avoiding the others.
Risk grouping: AI used to exploit
| Who caused it | A person or organisation “One engineer built a voice-controlled smart gun in his garage.” source |
|---|---|
| Intended or not | Intended “He tells it “we’re under attack from the front left and front right, respond accordingly.” It aims left, fires; aims right, fires (blanks).” source |
| Before or after release | After release “The robot gun’s voice commands use ChatGPT, prompting OpenAI to cut off the engineer’s access.” source |
| The AI's part | Used as a tool “The robot gun’s voice commands use ChatGPT, prompting OpenAI to cut off the engineer’s access.” source |
Others like this
Backdoored open-weight model exfiltrates project credentials through OpenAI Codex CLI 2026-10-16
Show every field, with its source and grade
What happened
Impact
LR-2026-0241 · version 2 · published 2026-10-25 after automatic checks; not yet checked field by field by a person · changes to this case · JSON