By Dan Kennedy • The press, politics, technology, culture and other passions

Tag: NewsBreak

Reuters exposes NewsBreak’s Chinese ties and AI-based local news hallucinations

Image (cc) 2019 by Journolink

You would have thought that misusing artificial intelligence to make up a murder that never took place and to amplify a false story about yet another murder would have been enough to stop NewsBreak. But the media outlet, which has rushed in to fill the gap caused by the decline of legitimate local news, has continued along its hallucinatory way — until now.

Last week, Reuters published an investigative report into NewsBreak exposing the company’s Chinese ties and its use of AI to write fiction. Reporter James Pearson begins with the fake murder last Christmas Eve from Bridgeton, New Jersey, which I wrote about in January. Here’s Pearson’s lead:

Last Christmas Eve, NewsBreak, opens new tab, a free app with roots in China that is the most downloaded news app in the United States, published an alarming piece about a small town shooting. It was headlined “Christmas Day Tragedy Strikes Bridgeton, New Jersey Amid Rising Gun Violence in Small Towns.”

The problem was, no such shooting took place. The Bridgeton, New Jersey police department posted a statement on Facebook on December 27 dismissing the article — produced using AI technology — as “entirely false.”

“Nothing even similar to this story occurred on or around Christmas, or even in recent memory for the area they described,” the post said. “It seems this ‘news’ outlet’s AI writes fiction they have no problem publishing to readers.”

Pearson fails to mention that the fake story was first exposed by Eric Conklin of NJ.com. Nor does he report that, just a few weeks earlier, NewsBreak used AI to amplify yet another fake-murder story that was produced by actual humans at the Mid Hudson News in New York, a monumental error revealed by Lana Bellamy and Phillip Pantuso of the Times Union, based in Albany.

Despite the lack of credit to local news outlets, the Reuters investigation represents the deepest dive yet into NewsBreak, and could result in action. As Pearson notes, NewsBreak, despite its sleazy tactics, has content-sharing arrangements with news outlets such as The Associated Press, CNN, Fox and, well, Reuters. As far back as 2022, Norman Pearlstine, a longtime top news executive who was working as a consultant for NewsBreak, warned his client: “I cannot think of a faster way to destroy the NewsBreak brand.”

NewsBreak remains a popular destination for people looking for local news. According to SimilarWeb, it was visited nearly 57 million times between March 24 and May 24. But maybe NewsBreak’s reign of error is coming to an end. In a follow-up story, members of Congress voiced concern, with Sen. Mark Warner, chair of the Intelligence Committee, saying, “The only thing more terrifying than a company that deals in unchecked, artificially generated news, is one with deep ties to an adversarial foreign government.”

To be fair, the company says those ties were severed several years ago. Still, NewsBreak’s irresponsibly sloppy use of AI to generate fake local news needs to be called out and shamed.

Leave a comment | Read comments

 

Another fake murder is reported. This one was generated by AI.

Photo (cc) by Nick Youngson via The Blue Diamond Gallery

You wouldn’t think it could happen again — but it did. Mere weeks after a small community news site in New York State reported a murder without bothering to do the due diligence that would have revealed the incident never took place, the same thing has happened in New Jersey. The difference is the role played by artificial intelligence.

At the Mid Hudson News in Newburgh, New York, the fake news was published as a result of human error. Its story was then picked up by the aggregation site NewsBreak, which added a commentary generated by AI lamenting the rise of social media as a factor in such (non-existent) violence.

In New Jersey, a false report of a murder actually originated at NewsBreak, and it appears to have been wholly generated by AI. Eric Conklin of NJ.com reports:

Police in a New Jersey city are urging the public to ignore a story featured on a news website about a fatal shooting they say was written using artificial intelligence.

The story on Newsbreak.com [the link now goes to a 404] said a “local resident” was found dead in the 100 block of West Broad Street in Bridgeton on Christmas Day. It further dives into the gun debate in America as communities seek an end to violence….

Police on Wednesday said the story has been circulating on social media, emphasizing an italicized note at the bottom of the text that the piece “includes content assisted by AI tools”

“Nothing even similar to this story occurred on or around Christmas, or even in recent memory for the area they described,” Bridgeton police wrote on their official Facebook page.

Noor Al-Sibai has a good overview at Futurism and observes:

Ultimately, the slipperiness of this faux article’s sourcing speaks to the heart of AI-generated content. Instead of revolutionizing media — or anything else, for that matter — outlet owners who insist on using generative AI instead of human writers have done little more than sow discord in an institution that’s already infamously mistrusted by the public.

Indeed. It also shows that even as local journalists with ethical scruples struggle to be heard above the noise, operators like NewsBreak will continue to take advantage of the crisis in community journalism to crank out fake news for fun and profit. Keep in mind, too, that most of the AI-generated crap that appears on sites like NewsBreak does not rise to the level of a murder that never actually happened, which makes it all the more difficult to parse fiction from reality. Caveat emptor.

Leave a comment | Read comments

Powered by WordPress & Theme by Anders Norén