What's new
Frozen in Carbonite

Welcome to FiC! Register a free account today to become a member! Once signed in, you'll be able to participate on this site by adding your own topics and posts, as well as connect with other members through your own private inbox!

Forbes: The Best (And Scariest) Examples Of AI-Enabled Deepfakes

Aaron Fox

SB's Minor Junker Descendant and Hunter of Nazis
Author
Source
There are positive uses for deepfake technology like making digital voices for people who lost theirs or updating film footage instead of reshooting it if actors trip over their lines. However, the potential for malicious use is of grave concern, especially as the technology gets more refined. There has been tremendous progress in the quality of deepfakes since only a few years ago when the first products of the technology circulated. Since that time, many of the scariest examples of artificial intelligence (AI)-enabled deepfakes have technology leaders, governments, and media talking about the perils it could create for communities.
...
Deepfake technology is improving faster than many believed it would. In fact, researchers have created a new software tool that allows users to edit the transcript of a video to alter the words—add, change, or delete—coming out of someone's mouth. This technology isn't available to consumers—yet—but examples of what has been done illustrate the ease with which the tool can be used to alter videos.
...
However, we can't merely rely on software to do the job for us. As deepfake technology is here and getting better every day, it would be prudent for us all to remember to critically assess the authenticity of videos we consume to understand their real intent. This means not just relying on the quality of the video as an indicator of authenticity but also assessing the social context in which it was discovered—who shared it (people and institutions) and what they said about it.

Basically, it has become increasingly impossible for anyone that isn't an expert in the field and doesn't have the equipment available to tell 'Deepfakes' from the real thing. Makes democracy increasingly impossible because you can easily make propaganda that is all but impossible to differentiate from reality.
 
Well, that is, in societies that are educated to distrust experts and whose leadership uses propaganda as their main political tool rather than actual expertise. One would think that serious media work to clarify fakes from real stuff, and yes they do in places where 'information media' hasn't merged with 'clickbait TV'. Also helps to have decent limitations on free speech to criminalize the use of such deep fakes as part of political and media campaigning.
 
I can't imagine with the current way we do machine learning that it'll ever be a perfect match for us from what I've read of things.

There will likely be ways in which we should theoretically be able to distinguish fakes from the real deal as certain types of stochastic phenomena mean it's unlikely to be ever 100% accurate.

Likely an entire field of "Deep Fake Forensics" will arrive and that will use its own ML algorithms to spot patterns in fakes and doing so show that show they're artificial in nature.

That isn't to say this won't be bad, but it's more the case it becomes a lot harder to distinguish reality from fact giving birth to a new wave of anti-establishment vulnerabilities in the plebiscite.
 
I can't imagine with the current way we do machine learning that it'll ever be a perfect match for us from what I've read of things.

There will likely be ways in which we should theoretically be able to distinguish fakes from the real deal as certain types of stochastic phenomena mean it's unlikely to be ever 100% accurate.

Likely an entire field of "Deep Fake Forensics" will arrive and that will use its own ML algorithms to spot patterns in fakes and doing so show that show they're artificial in nature.

That isn't to say this won't be bad, but it's more the case it becomes a lot harder to distinguish reality from fact giving birth to a new wave of anti-establishment vulnerabilities in the plebiscite.
Then again, this is a quickly-evolving field, evolving faster than thought possible... and that is problematic.
 
Then again, this is a quickly-evolving field, evolving faster than thought possible... and that is problematic.
A lot of predictions that have been in the lines of, "Its changing so fast" have historically been false.
 
A lot of predictions that have been in the lines of, "Its changing so fast" have historically been false.
Eh, with this field it has outpaced what was thought possible. Maybe it'll slow in the coming days, find a metastable pace, or go even faster... but it doesn't change the fact that deepfake has evolved faster than previously thought (from what I understand, the advances in deepfake was supposed to be more pedestrian)...
 
Back
Top Bottom