Tech
Not that it matters, but AI is now capable of detecting deepfakes
We’re playing a very dangerous game here.
As has already been stated by every legitimate writer at KnowTechie, deepfake technology has already gotten way out of hand.
Whether it’s a horrifying mashup of celebrities or a dubbed over scene from Game of Thrones, this AI-birthed phenomenon is a dog that needs to be taken out to the woodshed, so to speak, as it will only be used to further spread disinformation and misanthropy throughout the cesspool that is the Internet.
So how do you fight this menace?
The answer is not to simply destroy every last piece of tech capable or producing deepfakes, as any writer here would tell you. No, the best solution – one that our brightest minds came up with – was MORE AI.
In a report picked up by Vice today, computer scientists at the University of Southern California’s Information Sciences Institute claimed to have trained AI to detect deepfake videos by tracking “facial inconsistencies.”
Meet Therabody's All-New TheraFace Mask
TheraFace Mask is a breakthrough FDA-cleared LED skincare mask with added tension-relieving vibration therapy. Clinically proven results include firmer, smoother, healthier-looking skin with a visible reduction in fine lines, wrinkles, and uneven tone in as little as 8 weeks
To automate the process, the researchers first fed a neural network—the type of AI program at the root of deepfakes—tons of videos of a person so it could “learn” important features about how a human’s face moves while speaking. Then, the researchers fed stacked frames from faked videos to an AI model using these parameters to detect inconsistencies over time. According to the paper, this approach identified deepfakes with more than 90 percent accuracy.
Impressive? Sure.
But it’s worth noting that this latest step in the deepfake wars does not matter. Fighting AI with AI is a losing battle for two main reasons:
- It will never change your dumb uncle’s opinion that the video of “Mike Facebook” was legitimate because “I saw it with my own eyes.”
- Using AI to police AI will only make future deepfakes all the more difficult to spot, as they will be the superintelligent, radicalized bastard of AI-monitored AI
As Vice points out, we could soon be living in a world where AI is determining what reality is (you know, in addition to telling us how to feel and express love). But yeah, just keep laughing at these hi-larious mashups I guess.
https://twitter.com/jjvincent/status/1141676965281370113
What do you think? Worried about deepfakes and AI or will it prove to a positive thing? Let us know down below in the comments or carry the discussion over to our Twitter or Facebook.
Editors’ Recommendations:
- Make your space cooler with this smart coffee table
- Samsung just added a bunch of new SmartThings to its lineup
- Cloudflare glitched and took half of the internet with it
- Oh great, now a California police station is allowing robotic cops to patrol its parks
Follow us on Flipboard, Google News, or Apple News
Buying Guide
The best Steam Deck accessories (2023)
Buying Guide
The best wireless earbuds (2023)
