3 ms·
That’ll work for maybe a week, unless the US government has a significant lead in deepfake technology, which seems unlikely. Once deepfakes become that widely
by philwelch 4y ago
That’ll work for maybe a week, unless the US government has a significant lead in deepfake technology, which seems unlikely.
Once deepfakes become that widely feasible, people will post them on the internet for general amusement or other frivolous motivations all the time, and no one will take video at face value anymore.
- brutusurp 4y agoOne can ask ChatGPT to make them. There's tons of public and private photo, video, and audio data for fine-tuning/customization. It is really that simple now. Yay...
- xgbi 4y agoWe already have a semi famous daily show here in France where they are using deepfakes of our president and other powerful people as an humorous take on the current political situation. https://youtu.be/ampFHEXawUY https://youtu.be/ampFHEXawUY It’s quite well done and I think it proves your point: general amusement, and now face value doesn’t work anymore.
- pyuser583 4y agoThat’s ready happened: https://youtu.be/NTWsUMDSqcg https://youtu.be/NTWsUMDSqcg (Not safe for work. Not pornographic, just not appropriate for workplace.
- krapp 4y agoPeople already take Facebook memes and Youtube videos of godawful CGI spaceships and "reptoids" and other such nonsense seriously all the time. Deepfakes don't need to become widely feasible or even that good to be effective. Hell, deepfakes are all over Youtube now. US Presidents as shittalking gamers is a whole genre. You're assuming most people are rational, skeptical, and won't trust anything that doesn't have a valid chain of evidence or proof of validity behind it. That's incorrect - most people believe that about themselves, but still tend to take anything they're shown at face value if it aligns with their preconceptions and there is no emotional or social risk in doing so.