By Wired | Created at 2024-11-20 11:50:50 | Updated at 2024-11-21 09:47:58
22 hours ago
A side effect of this strategy is that it’s common for sex workers to have multiple legitimate accounts with slightly different usernames, none of which are verified, making them an easy target for content theft and imitation.
Both of the AI influencer guides we reviewed also talk about how to avoid being banned on Instagram. “Use a non-realistic bio picture and avoid including false location information in your bio to reduce the chances of being suspended for Inauthentic Identity,” the Digital Divas guide says. “If your picture is cartoonish and you are a digital creator, it’s less likely to be considered inauthentic.”
Professor EP, meanwhile, tells people to use a separate email account for each influencer and to make sure it’s “clean” and doesn’t connect to the person operating it or any of their other accounts. “Let’s say one of your accounts gets a ban. Having separate email addresses eliminates the risk that Instagram makes a connection between the banned account and your other accounts,” Professor EP says in the guide.
“Avoid account bans by using images that are visually appealing but not too provocative. We recommend that you adhere to the following criteria: visible face, amateur style” and “to prevent shadow bans from the get-go, you should warm up the account for the first eight weeks” by logging in daily and commenting on other people’s images to demonstrate “human activity,” Professor EP’s guide says.
St James said that even reporting accounts she knows for a fact are stealing from her is risky and could put her legitimate accounts at risk.
“Anytime that we as creators report these fakes, we tend to get in trouble,” she said. “It seems like Instagram says, ‘Oh, you're tattling on this imposter? Let's take a closer look at your account and see if we can find problems with it.’ So a lot of the time we don’t even report them. Sometimes we pay services to report fakes, but it’s like playing whack-a-mole. It’s never-ending.”
Mantzarlis, the director of the security, trust, and safety initiative at Cornell Tech, and St James agreed that it’s not clear whether Instagram has the means to remove or label these accounts as AI-generated, but the fact that the company isn’t doing either at the moment appears to work to its benefit.
“People are clicking, liking, and interacting with these accounts, and some of that engagement is real, and some isn’t,” Mantzarlis said. “Instagram can sell this as traffic. It can sell ads against this. So, is there a future where actual, real human accounts are almost like an elite, smaller percentage of Instagram? I think yes.”
“If all of a sudden they got rid of all the bots, the dead accounts, the fake accounts, the imposter accounts, what happens to their advertising?” St James said.
