
While most hearables focus on biometric smarts or language translation, a new earpiece is aiming to notify users which voices are humans ones and which ones aren't.
The prototype device, known as the Anti-AI AI, uses Google's Tensorflow machine learning software to distinguish speech and alert the wearer if it isn't, in fact, a real human.
Read next: The best hearables
It was created as a proof of concept in just five days by a team at creative tech agency DT, who trained up the AI through a database of synthetic voices and teaching the offline network to recognise artificial speech patterns.
So, how exactly does this dystopian-level wearable work?
Well, when in operation, it captures audio spoken in the vicinity of the user and uploads it to the neural network in the cloud. If you're surrounded by humans, all is well and you'll receive no feedback. If, though, one of the robots is walking and talking among us, the device can pick up on the synthetic speech and alerts the user.
However, this isn't done through a simple haptic vibration or noise. Instead, the device uses a thermoelectric cooling element to send a light chill to the neck. That's right, a literal chill down the spine.
Now, of course, it's worth bearing in mind that the Anti-AI AI is just a concept for now, but something similar will no doubt become useful as robots and fake voices become more prevalent. For now, the researchers at DT will be refining the device and improving its ability to decipher different speech patterns.
Hey, maybe they can also develop a hearable that detects fake news?
Source: R&D
How we test