NewsGuard, the purportedly impartial media rating service that has created a blacklist of disfavored news organizations, is ramping up efforts to prevent AI from spreading fake content that could influence the upcoming U.S. presidential election.
Semafor reports that in the lead-up to the 2024 election, bad actors are already exploiting AI technology to generate and disseminate misinformation on a massive scale. NewsGuard, which claims to be a misinformation watchdog that rates news sources on journalistic standards while boasting of its ties to the government’s censorship machine, is offering new services aimed at thwarting AI-generated falsehoods about candidates like Joe Biden and Donald Trump.
“The malign actors have really begun to perfect the art of abusing not only the open internet, but also AI,” said NewsGuard CEO Gordon Crovitz. “We’re now living in an AI-enhanced internet, right? And the malign actors are producing more content, more cheaply, more targeted, and more divisive and more persuasive.”
Newsguard is pivoting to combat AI-enabled misinformation. It has partnered with tech giants like Microsoft, which licensed NewsGuard’s data to help train the new AI-powered Bing search engine.
NewsGuard targets pro-Trump news outlets (Alex Wong/Getty, Jason Connolly/AFP/Getty, NewsGuard; Edit: BNN)
However, NewsGuard’s reputation and reliability has been repeatedly called into question. Breitbart News reported in 2022:
Despite presenting itself as a fake news watchdog, NewsGuard’s own founders have fueled false claims about major national news stories. At the height of the 2020 presidential election, NewsGuard co-founder Steve Brill stated on national television that he thought there was a “high likelihood” that the Hunter Biden laptop story was a “hoax perpetrated by the Russians.”
NewsGuard advisory board member and Russiagate conspiracy theorist Michael Hayden signed on to an open letter making the same claim — and has yet to retract it. He remains an advisor to NewsGuard.
The “anti-misinformation” organization has acquired such a strong reputation for targeting conservatives that it was slammed by the chairman of the influential Republican Study Committee, Rep. Jim Banks (R-IN) wrote a letter to the Pentagon warning of its “extreme partisan bias.”
NewsGuard maintains a browser extension that can be added to Edge, Chrome, and other web browsers. Once installed, NewsGuard “nutrition labels” appear next to links to news websites, with a green label signaling the website is trustworthy and a red label signaling the opposite.
Fox News, along with Breitbart News, Revolver, and the Rebel Media now has a “red” rating. Left-leaning outlets known for publishing uncorrected, misleading information about major stories, like downplaying the Hunter Biden laptop, continue to receive perfect or near-perfect NewsGuard ratings.
Among NewsGuard’s new initiatives is “fingerprinting” misinformation to help AI models detect and avoid sharing false claims. The group is also “red-teaming” AI systems, using the mindset of bad actors to ensure text, image, video, and audio generators cannot circumvent safeguards against misinformation.
NewsGuard also aims to incorporate authoritative data on elections, such as polling locations and dates from government websites, into AI knowledge bases. This ensures voters receive accurate logistical information.
NewsGuard claims it documented a 1,000 percent surge in AI-generated fake news articles about elections between May to December 2022. Meanwhile, a Proof study found language models like ChatGPT frequently provide inaccurate answers to granular questions about voting procedures.
Read more at Semafor here.
Lucas Nolan is a reporter for Breitbart News covering issues of free speech and online censorship.