n n n ‘. concat(e. i18n. t(“search. voice. recognition_retry”),’n
(Bloomberg) — Meta Platforms Inc. se gotten rid of a bunch of Facebook accounts linked to covert influence campaigns from China, Israel, Iran, Russia and other countries, some of which used synthetic intelligence equipment to generate misinformation, according to the company’s quarterly risk report. report.
Most on Bloomberg
Major drivers of U. S. customer spending lose steam
GameStop racks up its raise as Gill’s return to Reddit shows a huge gamble
Mnuchin Pursues Wall Street Glory with His Foreign War Fund
Homebuyers are starting to push back against high costs in the U. S.
AMLO Sheinbaum’s protégé becomes Mexico’s first female president
Meta, the parent company of Facebook, Instagram, and WhatsApp, has observed that malicious actors rely on AI to produce fake images, videos, and text in an effort to influence users of their sites. But the use of generative AI has affected the company’s capacity. to disrupt those networks, Meta said in Wednesday’s report.
Among the disinformation campaigns, the company exposed a deceptive Chinese network that shared AI-generated poster photographs of a fictional pro-Sikh movement, and an Israel-based network that posted AI-generated comments praising the IDF on the pages of media organizations and public figures. The company said it removed many of those networks before it could build an audience among the original communities.
“Right now, we don’t see AI generation being used in an extraordinarily complicated way,” David Agranovich, Meta’s director of risk disruption policy, said at a press conference on Tuesday. Tactics such as creating profile shots generated by AI or synthetic intelligence to produce large volumes of spam content have not been effective so far, he said.
“But we know that those networks are inherently contradictory,” Agranovich said. “They will continue to evolve as their generation evolves. “
Social media corporations such as Facebook, TikTok by ByteDance Ltd. and Elon Musk’s X have been suffering with the influx of misleading AI-generated content on their sites. This year alone, falsified audio recordings of U. S. President Joe Biden and photographs of Israel’s Hamas clashes have circulated on social media and garnered millions of views.
Nick Clegg, president of global affairs at Meta, stressed the need to find and tag AI-generated content, especially as the company prepares for the 2024 election cycle. Global elections will take place in more than 30 countries this year, which adds up to many places. where the company’s programs are widely used, such as the United States, India and Brazil.
Clegg said creating a popular industry around watermarks is “the most urgent task we face today. “Meta has been working on the ability to find and label symbols created through teams from artificial intelligence companies such as Google and Alphabet Inc. ‘s OpenAI. The company has already added visual markers to some symbols, as well as non-visual markers and credentials in symbol files.
Meta recently updated its policies to label misleading, AI-generated content on its site, rather than cutting it off. The company also asks advertisers to disclose when they use AI to create Facebook or Instagram classified ads similar to social issues, elections, or politics, but the company does not audit politicians’ political classified ads.
(Updates with symbol detection in the eighth paragraph).
Most read in Bloomberg Businessweek
Yes, Walmart store managers can make $500,000 a year
Disney’s bets on Pixar sequels are back to normal
The Budget Geeks Who Helped Solve America’s Economic Puzzle
Israel searches for underground secrets by tracking cosmic particles
How Rage, Boredom and WallStreetBets Created a New Generation of Young American Traders
©2024 Bloomberg L. P.