Jul 30, 2014 · 4 minutes

From Reddit's catastrophic witch hunt in the wake of the Boston Marathon bombing to the deluge of fake images and reports during Hurricane Sandy, social media has turned every breaking news event into a playground of misinformation. Most recently, the downing of Malaysian Airlines MH17 prompted all manner of false reports, like that dozens of Americans had died, shared even by otherwise reputable journalists.

Many individuals and organizations have tried to counter this trend, from "Is Twitter Wrong?" to Storyful, which employs a crack-team of journalists and technology experts to “debunk, fact-check, clarify, credit and source.” The company most recently partnered with Facebook on "FB Newswire," a page that curates newsworthy Facebook content* that have already been carefully vetted by Storyful's team.

One problem these truth crusaders have failed to solve, however, is how do you stop the spread of those original false tweets? Published tweets can't be edited. And while the original user can delete the tweet, such practices aren't in the interest of transparency. I myself have been guilty of sharing (albeit with heavy caveats, of course) an early report later proven to be false, and although I issued a correction tweet, I saw the retweets of the original untrue post keep piling in for days.

So how can journalists and news consumers leverage the power of observers and experts from the crowd without risking the spread of information?

Grasswire thinks it's found a solution. Launched about a month ago and billing itself as a "real-time newsroom controlled by everyone," it functions a bit like Reddit except instead of simply upvoting or downvoting, users can click "confirm" or "refute." The confirmations and refutations stick with each post so that when the link is shared to social media, whatever factchecking is in place goes along with it. That way, even if a Grasswire link to a false claim is tweeted out or posted to Facebook, once its been refuted all subsequent retweets and posts will surface that factcheck. Co-founder and CEO Austen Allred says in the first two weeks since launching it's had over 27,000 people post or engage with content on the site.

For example, one user posted a photo (warning: very graphic) of women slaughtered in a bathroom, along with the text, "These girls were all recently killed by ISIS militants in Nineveh just for being Christian. Where is the outrage?" But below that, a user notes that the picture was actually taken at an alleged brothel in Baghdad, and whoever was behind the attack perpetrated it because they thought the establishment had been employing child prostitutes. The post now appears with a giant red "REFUTED" sign over it.

Today, Grasswire is announcing an Android app that sends push notifications to users about big developing stories. "We have an algorithm," Grasswire co-founder/CEO Austen Allred tells me. "As the amount of activity (on a story) crosses a certain threshold we know something abnormal is happening, something is breaking."

Allred first got the idea for Grasswire in 2011 while living in China. He ended up going to the wrong train station because a cab driver misunderstood him. The train he was supposed to be on, however, collided head-on with a bullet train, killing forty people. In the wake of the tragedy, and in part due to his having narrowly avoided being on that train, Allred began to obsessively search social media sites for information about the crash.

"There were hundreds of people actually on that train [many of which survived]," he says. "We live in a world where even in China everyone is publishing. Why can't I find the firsthand accounts of the people directly involved?"

Social media has come a long way since 2011, and there's rarely a shortage of eyewitness accounts posted to Twitter and Facebook. But the verification methods still haven't caught up.

"Like we see in Russia and Ukraine, there's all sorts of BS and propaganda, pictures they're pulling from Israel or Gaza and saying it's in Ukraine," Allred says.

So far, Grasswire has raised $50,000 and is now looking to raise a more substantial Seed round. In terms of monetization, Allred is mulling a number of options.

"Honestly we've had interest for licensing the API from a lot of local news stations saying, 'Hey we could just turn it on and let it hum.'" In the meantime, the company wants to continue tweaking and iterating on the design. "Eventually we'll have advertisers. The goal is just continuing to prove that the engine is reliable and works, and scale."

That said, launching "yet another social network," even one with a unique and utilitarian function like Grasswire, won't be easy. Despite its flaws, Twitter is still the go-to destination for news junkies. Facebook, through FB Newswire's vetted user-generated content, is gunning for that spot too. And there will always be Reddit, that chaotic hornet's nest of misinformation and prankery.

But if Grasswire can make some improvements to its user interface and, perhaps even more importantly, master the art of the news push notification, it's not too late to carve out a nice niche in the news reader market. This is particularly true on mobile where the number of people reading news is on the rise. Even just a few weeks in, Grasswire is already a valuable tool for journalists, researchers, and newshounds. Now, perhaps with the help of a larger media partner, it just needs to find a way to reach the average news-reading joes.

[illustration by Hallie Bateman]

*an earlier version of this post stated that FB Newswire curates YouTube videos.