I’ve been interested in deepfake technology recently.
We’ve all been exposed to it: fraudulent audio or video content usually created by AI. Our aunts and uncles might believe every meme they see on Facebook or other social media channels, but it can be a lot more nefarious.
Deepfakes can trick us into believing fake news, they can mimic our voices to trick banks or insurance companies, or they can create untrue documentation designed to destroy the reputation of politicians, business leaders, or celebrities.
But then there are more traditional use cases. This week, I spoke to Michael Matias, co-founder and CEO of Clarity. The AI cybersecurity startup protects against deepfakes, as well as new social engineering and phishing attack vectors in the context of recruitment: it makes sure that companies hire people who really are who they claim to be.
“You have nation states that are targeting enterprises in other states where they want to learn information, they want to steal money, they want to disrupt critical infrastructure,” Michael explained. “One of the reasons hiring fraud has become top of mind for many companies is that a lot of Fortune 500s over the last few months have discovered that they accidentally hired North Korean IT workers… Those specific talents did not disclose they were from North Korea, and they posed as US citizen employees earning critical access to sensitive information.”
This week, we discuss the evolution of deepfake technology and AI voice duplication, highlighting the advancements made in the last three years, basically when ChatGPT was launched.
This isn’t the first time that Michael and I have connected with a podcast. For a few years, I helped publish his podcast, 20 Minute Leaders on CTech - over the years we produced more than 100 episodes together.
It was a pleasure to speak with him today on my new platform. I hope you enjoy it.
Learn more about Clarity here: https://www.getclarity.ai/
