System and method are disclosed to protect a voice-activated assistant against accidental or malicious activation. The system includes a device interacting with the cloud having an assistive device, which in turn is protected by another, protective assistant. The protective assistant detects incoming malicious (or spurious) actions triggered by a third party which tries to connect to the device through the assistant. The protective assistant then emits a signal that would perturb the third party signal, or prepares and issues “cancellation” commands to automatically cancel the action. The system may also notify the user via SMS / phone call / email / screen notification to further manually check the legitimacy of the detected malicious action. User experience is improved as accidental or unintended orders will not be executed by their assistants.
Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 License.
Feuz, Sandro; Deselaers, Thomas; and Carbune, Victor, "Active Protection Of Voice-Activated Assistant Against Accidental And Malicious Activations", Technical Disclosure Commons, (April 24, 2017)