Just Type and Make People Say Whatever with Increasingly Advanced AI “Deepfakes”

Share: 

AI and its impact on photography and videography is becoming a bigger issue of concern and if you thought it was all overblown, wait until you see the kinds of things that even the simple stuff is capable of producing when it comes to fake videos and speech.

Photo by Joseph Redfield from Pexels.

The Verge highlighted research led by Adobe Research and joined by Stanford, Princeton, and the Max Planck Institute for Informatics which looked at how software can be used to edit the literal words coming out of someone’s mouth to make it look as if they said something else. Naturally, the group created a pretty cool video to illustrate their point and its results are pretty surprising. For fans of the film Apocalypse Now, you’ll notice how easy and seamless the change of one of that film’s famous quotes was using this software.

None of the efforts demoed by this consortium are going to be available for consumer use but that might not be the permanent state of things. After all, as The Verge points out, Adobe, for its part, has shown off tech in the past called VoCo that purported to make editing video clips and the speech in them as easy as editing a picture.

How it works is that the software analyzes the subject’s speech and isolates common sound patterns from that. The software can then link up these component sounds with the facial expressions the speaker makes when saying those sounds. They then combine these two together into a 3D model and impose that on the lower portion of the speaker’s face to make the appropriate facial movements.

Probably most frightening of all was a survey of a group of individuals found that the majority were fooled by the “deepfake” videos.

In a blog post discussing the research the team wrote, “Although methods for image and video manipulation are as old as the media themselves, the risks of abuse are heightened when applied to a mode of communication that is sometimes considered to be authoritative evidence of thoughts and intents. We acknowledge that bad actors might use such technologies to falsify personal statements and slander prominent individuals.”

As always, we would love to know your thoughts on this technology and how it could impact society in the future. Let us know in the comments section below. 

[The Verge]

What We Recommend to Improve Your Photography Fast

It's possible to get some pretty large improvements in your photography skills very fast be learning some fundamentals. Consider this the 80:20 rule of photography where 80% of the improvements will come from 20% of the learnable skills. Those fundamentals include camera craft, composition, understanding light and mastering post-production. Here are the premium guides we recommend.

  1. html cleaner  Easy DSLR –  Friend of Light Stalking, Ken Schultz has developed this course over several years and it still remains the single best source for mastering your camera by identifying the main things that are holding you back.
  2. Word to html  Understanding Composition – As one of the core elements of a good photograph, getting your head around composition is essential. Photzy's guide to the subject is an excellent introduction. Their follow-up on Advanced Composition is also well worth a read.
  3. Word to html  Understanding Light – Also by Photzy, the other essential part of photography is covered in this epic guide and followed up in Understanding Light, Part 2. This is fundamental stuff that every photographer should aim to master.
  4. Word to html  5 Minute Magic Lightroom Workflow – Understanding post production is one of the keys to photographs that you will be proud of. This short course by one of the best in the business will show you how an award-winning photographer does it.

About Author

Avatar

Kehl is our staff photography news writer and has over a decade of experience in online media and publishing and you can get to know him better here

Leave a Reply

Your email address will not be published. Required fields are marked *