Imagine an AI so adept at mimicking voices that it can clone anyone’s speech with just a tiny audio sample, making it virtually.