Artifact uses AI algorithms to modify news feeds for each user. With the addition of new tools, the site gives readers a little bit more control. There are lots of ways to write a story, and many of them are bad. Look at this story, for instance. It might have been titled:
You’ll never believe what this app lets its users do. or, New App Wages War Against Fake News. or even, This Trendy News App Gives Users One Weird Trick.
But if it had been, you might have been disappointed by the story itself. This is how clickbait usually works: it makes big claims, builds up expectations, and then delivers something that isn’t nearly as exciting as what was promised. You might have gotten angry and wanted to get your time back or do something to get back at them.
Artifact the news app created by the co-founders of Instagram, offers its users an outlet for this emotion. The AI-powered article recommendation platform now permits users to label news articles as “clickbait” if the headline misrepresents the content of the article. If a user believes they were tricked into reading an article by a misleading or hyperbolic headline, they can report the post to the app. The clickbait marking option is accessible by selecting the article’s three-dot menu or by clicking and holding the article in the feed.
“We will use your flag as a ranking signal so we can give more weight to helpful articles and less weight to misleading ones for the community,” the company wrote in a blog post-Monday afternoon announcing the new feature.
Kevin Systrom, one of the people who started Artifact, told Gizmodo in an email that the platform’s approach is still changing and that the main goal right now is to gather information. “As more people start to mark articles as clickbait, our current plan is to collect data and see how good the signal from the user base is,” he wrote.
Once there’s enough information to go on, Artifact might start lowering the rankings of certain stories or taking content off the site if people say it’s clickbait. As first reported by The Verge and confirmed by Systrom to Gizmodo, if a story is reported enough times as clickbait, Artifact might even change the headline. The tech founder said that the business is trying to “figure out the right response.” “Right now, everything is checked by hand.”
Because there are so many unknowns, it’s hard to say how these clickbait ranks will turn out and how much power Artifact users will really have. Systrom didn’t say for sure how many reports an article would need to be labelled as “clickbait” by the algorithm, nor did he say how the app would handle changes to headlines without taking on too much responsibility for mistakes that might be made. But he did say that Artifact would “actively guard against” users mistakenly marking news as clickbait when they don’t like or agree with a story.
With the anti-clickbait change, Artifact now has more features that put it between Apple News and a social media site like Reddit or the old Twitter. Before, the company added in-app comments and “reputation scores” (like Reddit) so that users could interact with each other through specific posts and be held responsible for those interactions.
In the same blog post that announced the clickbait flag button, Artifact also announced the introduction of two new features: the ability to react to an article with emojis and the ability to share an article excerpt as an image.
If you press and hold on to an article in the feed or tap the heart icon at the bottom of an article you are reading, the emoji react choices will appear. The share arrow menu now has a new option to share an image. This will let short pictures and blurbs be sent straight to Instagram stories and other places.
