> Audio used for wake-word detection is processed locally on the TV and, if no wake word is detected, audio is not converted to text, stored, or transmitted. Voice-recognition results and related technical logs may be generated as part of processing a voice command. These records are associated with specific voice interactions and do not indicate continuous recording of conversations occurring outside an active voice recognition session.

From their statement. This entire thing is contradictory. You cannot both say "not converted to text, stored, or transmitted", while at the same time saying the results may be generated and stored in the form of logs.

We need to hold companies responsible for this shit.