AI Voice Cloning Lands Baltimore Teacher in Hot Water: The Risks of Deepfake Technology

And so it begins. A Baltimore gym teacher has just been arrested for using generative AI to clone the voice of the principal, using the AI to make it sound like the principal is going on a racist hate filled rant. This gym teacher was arrested yesterday and experts have gone through the recording and realized that it's a fake because it has a flat tone, unusually clean background sounds lack consistent breathing or pauses. Here's the thing though. These tools are getting better every single day. And what can be caught now because it doesn't quite sound right in a month or two, maybe six months is going to be perfect. In part, this might be why OpenAI have withheld their text to voice generation platform because they know that this is a really powerful tool that can be abused extremely easily. And in America right now, there is the no Fakes act and the no AI Fraud act, which are bills which are going to be seeking to prevent this kind of technology. I imagine we're going to be seeing more of that this year because it's an election year both here in the UK and obviously in the us. So I think this kind of deep fake voice and video technology is going to become very prevalent and it's going to become a large talking point as we move forward. Make sure you drop me a follow and then I can give you the latest AI news as and when it comes in.