“I am always surprised that in the physical world, when we release products, there are really strict guidelines,” says Farid. “You can’t launch a product and hope it doesn’t kill your customers. But with software, we thought, ‘This doesn’t really work, but let’s see what happens when we release it to billions of users.’”
If we start to see a significant number of deepfakes go viral during the election, it’s easy to imagine someone like Donald Trump sharing this kind of content on social media and claiming it’s true. A deepfake of President Biden says that something ineligible could appear right before the election and many people may never find out it was AI-generated. Research has been consistent showAfter all, that fake news spreads further than the real news.
Even if deepfakes don’t become popular before the 2024 election, which is still 18 months away, the fact that this type of content is created could affect the election. Knowing that fraudulent images, audio, and videos can be created with relative ease can cause people to distrust the legitimate material they encounter.
“In some respects, deepfakes and generated AI don’t even need to participate in elections for them to still cause disruption, because now the well is tainted with the idea that every single one of the people in the world can’t get it,” Ajder said. anything can be fake,” Ajder said. “That provides a really useful excuse if something inconvenient happens to you. You can remove it as fake.
So what can be done about this? A solution is something called C2PA. This technology cryptographically signs any content generated by a device, such as a phone or video camera, and documents who took the photo, where, and when. The cryptographic signature is then kept on a centralized immutable ledger. This will allow legitimate video producers to prove that they are, in fact, legit.
Some other options related to what is called Fingerprinting and bookmark images and videos. Fingerprinting involves taking what’s called a “hash” from the content, which is basically just strings of its data, so it can be verified as legitimate later. Watermarking, as you might expect, involves inserting digital watermarks on images and videos.
It is often suggested that AI tools could be developed to detect worms, but Ajder does not sell that solution. The technology is not reliable enough, he says, and it won’t be able to keep up with the constantly changing general AI tools being developed.
One final possibility to solve this problem is to develop some kind of instant information checking tool for social network users. Aviv Ovadya, a researcher at the Berkman Klein Center for Internet & Society at Harvard, says you can bookmark a piece of content in an app and send that content to a contextualization engine to inform you regarding the authenticity of that content.
“Communication literacy evolves with the rate of advancement of this technology is not easy. You need it to happen almost instantaneously—when you look at something that you see online and you can understand the context of that thing,” Ovadya said. “What are you looking at? You can cross-reference it with sources you can trust.”
If you see something that could be fake news, the tool can quickly notify you of its authenticity. If you see an image or video that appears to be fake, it can check the sources to see if it’s been verified. Ovadya says it could be available in apps like WhatsApp and Twitter, or it could simply be its own app. The problem, he says, is that many of the founders he has spoken to simply don’t see a lot of money developing such a tool.
It remains to be seen whether any of these possible solutions will be passed before the 2024 election, but the threat is growing and there is a lot of money pouring into the development of generalized and very powerful AI. little to find ways to prevent the spread of this kind of misinformation.
“I think we’re going to see loads of tools, as we’ve seen, but I think [AI-generated political content] will continue,” Ajder said. “We’re fundamentally not in a position to deal with these powerful, incredibly fast-moving technologies.”
