Artifact News App Fights Clickbait With AI: ‘It’s Basically Always Right’

“What if we actually just build a product that lets you take action based on the problem that you’re seeing?” co-founder Kevin Systrom says

Robot typing AI
Getty Images

Clickbait in web journalism and blogging is a big problem, and the staff of the Artifact news app thinks it’s found a solution by way of AI. Artifact has introduced the capability for readers to flag clickbait headlines and have artificial intelligence rewrite them to accurately reflect article contents.

AI won’t rewrite every news article headline that Artifact circulates. The GPT-4-infused app will only tinker with a header if enough app users flag it as clickbait. Once that happens, Artifact gets to work on retooling whatever was written by the attention-seeking outlet in question. It won’t rewrite the article’s headline in the article itself, but it will modify it on the news feed, denoting altered headlines with an asterisk.

“It’s clear that people are passionate about ridding the world of clickbait,” said Artifact co-founder Kevin Systrom in a chat with The Verge. “What if we actually just build a product that lets you take action based on the problem that you’re seeing?”

Human reviewers at Artifact will see the most rewritten headlines and have the power to push that revision to all users on a case-by-case basis.

“In our experience in testing, it’s basically always right,” Systrom said, referring to the accuracy of the AI-rewritten headlines. “I don’t think we’ve actually found an example where it hasn’t been right.”

As a humorous aside, Systrom told The Verge that during the AI feature’s refinement phase, it was actually caught producing a bit of clickbait all its own, having grown used to scanning so much of it that it started to think that’s just how internet headlines were.

Internet journalism isn’t the only area where AI may be more capable than human beings. In the UK, the CEO of energy supplier Octopus Energy revealed that AI had been lending a hand with customer support and was managing better customer satisfaction ratings than the company’s human employees. Furthermore, AI’s proven capable at providing stock advice, even if using it for that purpose isn’t always advisable or permitted by the AI’s creators.