Experts know that generative AI has the potential to dramatically change the information landscape, and that problems that have long plagued tech platforms – such as misinformation and misinformation, fraud and hateful content – are likely to worsen despite the guardrails that companies say they put in place.
There are a few ways to tell if something has been created or manipulated using AI: people or campaigns may have confirmed its apply; fact-checkers could analyze and debunk something that was circulating around the world; or maybe the AI content is clearly being used for purposes such as satire. Sometimes, if we’re lucky, it’s a watermark, which means there’s something indicating it was generated or changed by artificial intelligence. But the reality is that this probably only applies to part of what already exists. Even our own data set is almost certainly an underestimate.
And this brings us to the next point: as British journalist Peter Pomerantsev said: “When nothing is true, anything is possible.” In an information ecosystem where everything can be generative artificial intelligence, it is effortless for politicians and public figures to say that something real is false – this is the so-called “liar’s dividend”. This means that people may be less likely to believe information, even if it is true. As for fact-checkers and journalists, many do not have readily available tools to assess whether AI has created or manipulated something. Whatever this year brings, it will likely be just the tip of the iceberg.
But just because something is bogus doesn’t mean it’s bad. Deepfakes have become popular in satire, chatbots can (sometimes) provide good information, and personalized campaign reach can make people feel seen by their political representatives.
It’s a brave recent world, but that’s why we follow it.
Chat room
As part of our AI project, we’re asking readers to submit any examples of generative AI you see in the wild this election year.
To better understand how we will evaluate submissions (or even items you find) and send them to us, please click this link here. If you’re not sure whether something was made with generative AI or just a low-cost knock-off, send it in anyway and we’ll take a look.
💬 Leave a comment below this article.
WIRED reads
Want more? Sign up now for unlimited access to WIRED.
What else are we reading
🔗 TikTok says it has removed an influence campaign originating from China: : Last week, TikTok said it had removed thousands of accounts on its platforms linked to 15 Chinese influence campaigns. (Washington Post)
🔗 Ramaswamy urges BuzzFeed to lay off employees and amplify more conservative voices: : Vivek Ramaswamy, a former Republican Party presidential candidate, is currently an activist investor at BuzzFeed. It wants the publication to court conservative readers and said it “lied” in its reporting on Donald Trump and Covid, among others. (Bloomberg)
🔗 OpenAI forms board with Sam Altman after disbanding security team: : According to the company’s blog, the recent board will make security recommendations and will have 90 days to “further develop OpenAI’s processes and security features.” (Bloomberg)
Downloading
Last thing! This week on the podcast, I spoke with our editor and host Leah Feiger about the AI elections project. Give it To listen!
In addition to talking about the recent project (can you tell I’m excited?), Leah and I were joined by Nilesh Christopher, who has been reporting on the role of deepfakes in the Indian elections for WIRED. The key takeaway: The Indian elections are coming to an end soon, and many of the country’s growing generative AI companies are looking for recent markets that might be interested in their tools – they might even come to the polls near you.
That’s it for today. Thanks again for subscribing. You can contact me via e-mail and X.
Image source: Getty Images