Here Come the Fake Videos, Too
nytimes.com
Here Come the Fake Videos, Too
1–10 of 107 posts
Re: Here Come the Fake Videos, Too
#2Since reddit banned deepfakes, including SFW content, does anyone know where the community is now congregating? The appeal of the technology from online avatars to cheaper cgi is undeniable. Is this a case of throwing out the baby with the bathwater?
Re: Here Come the Fake Videos, Too
#310+ years ago we started getting the technologies for facial land mark detection, and now we have facial swapping. Currently we have the beginnings of good full-body pose derection, and I imagine soon we’ll eventually have full body swapping (maybe along with clothes).
That raises interesting questions. As legitimate looking sources become harder to trust, what other ways can we verify them? One idea that was floating around is key signing each datafile. That raises the question though of how to manage keys. [ Maybe have each key tied to a digital id, the id similar to Estonia e-residency? ]
At low levels of risk, like a recorded automobile accident, is such scrutiny useful?
Thoughts everyone?
Re: Here Come the Fake Videos, Too
#4Since reddit banned deepfakes, including SFW content, does anyone know where the community is now congregating? The appeal of the technology from online avatars to cheaper cgi is undeniable. Is this a case of throwing out the baby with the bathwater?
There is one mention here:
“Following the ban, many of the users moved to the /deepfakes/ board on 8chan[18] and the /v/DeepFakes community on Voat.[17]” [0]
[0] - http://knowyourmeme.com/memes/cultures/deepfakes?full=1
Re: Here Come the Fake Videos, Too
#5The era of verifiability-- when photographic and video evidence could plausibly be trusted-- is over.
Photos have been dubious since photoshop, but still required expertise and artistry to use. Fake moving pictures required considerably more extensive expertise, lots of time, and usually specialized equipment.
None of that is true anymore. We are back to a pre-industrial age of discourse, where rumor and hearsay and anecdote dominate again. Buckle up.
Re: Here Come the Fake Videos, Too
#6Since reddit banned deepfakes, including SFW content, does anyone know where the community is now congregating? The appeal of the technology from online avatars to cheaper cgi is undeniable. Is this a case of throwing out the baby with the bathwater?
I believe dpfak.com has become as close as you can get to a central hub. The landing page is pornography, but you can click through to the forums and there's discussion to be found.
Re: Here Come the Fake Videos, Too
#7The era of verifiability-- when photographic and video evidence could plausibly be trusted-- is over. Photos have been dubious since photoshop, but still required expertise and artistry to use. Fake moving pictures required considerably more extensive expertise, lots of time, and usually specialized equipment. None of that is true anymore. We are back to a pre-industrial age of discourse, where rumor and hearsay and…
Is this true though?
Sure, for a casual observer, these new methods for generating videos appear convincing, but is that the right bar to judge "ability to fake evidence"?
As far as I know, there have always been more sophisticated techniques and forensics to determine if an image is doctored, and likewise for video. I've not seen any research tackling fooling those methods yet, and I would bet that naive implementations of neural networks for generating videos would leave very obvious "neural network" artifacts. Of course, this is still new technology, so it will obviously get better at fooling our other tools over time too, but as of right now, I don't think the clamoring for "all evidence can now be faked" is all that justified.
Re: Here Come the Fake Videos, Too
#810+ years ago we started getting the technologies for facial land mark detection, and now we have facial swapping. Currently we have the beginnings of good full-body pose derection, and I imagine soon we’ll eventually have full body swapping (maybe along with clothes). That raises interesting questions. As legitimate looking sources become harder to trust, what other ways can we verify them? One idea that was floatin…
I don't believe a signing solution is feasible. It's currently too complicated for even a lot of IT experts, and the amount of actors involved is too large to even move to a single unified solution.
Given that as of now authenticity can still be verified by experts, it's back to trusting journalistic outlets to properly verify news and sources.
Re: Here Come the Fake Videos, Too
#9The era of verifiability-- when photographic and video evidence could plausibly be trusted-- is over. Photos have been dubious since photoshop, but still required expertise and artistry to use. Fake moving pictures required considerably more extensive expertise, lots of time, and usually specialized equipment. None of that is true anymore. We are back to a pre-industrial age of discourse, where rumor and hearsay and…
No, this is an opportunity for new technology that adds additional data stream(s) to audio/video to ensure authenticity. Perhaps something like a one way hash of physical properties related to time/position and the rasterized data, in a way that cannot be faked.. or maybe a move away from rasterized data. Idk. Just brainstorming.
I’m confident there will be an innovation in this space since there is clearly a very urgent need. In fact, this could be a start-up opportunity. Maybe there is relevant academic research already on which a startup idea could be built.
Re: Here Come the Fake Videos, Too
#1010+ years ago we started getting the technologies for facial land mark detection, and now we have facial swapping. Currently we have the beginnings of good full-body pose derection, and I imagine soon we’ll eventually have full body swapping (maybe along with clothes). That raises interesting questions. As legitimate looking sources become harder to trust, what other ways can we verify them? One idea that was floatin…
Posted in a sibling thread already, but what about supplemental data streams related to environmental data (to verify time and position) that cannot be faked, somehow combined with the rasterized/sampled a/v in an irreversible way, or a new technology that doesn’t use traditional pixelized video or sampled audio that can’t be faked?
Surely there will be an innovation since there clearly is a need for this.