OpenAI has stopped an Iran-related disinformation campaign that used ChatGPT
OpenAI has revealed that it successfully disrupted an Iranian influence campaign that leveraged ChatGPT to fabricate fake news and social media posts with the aim of targeting Americans. The campaign involved the creation of content for five websites masquerading as legitimate news outlets, which were utilized to disseminate divisive messages on various topics including the US presidential campaign, LGBTQ+ rights, and the conflict in Gaza.
Details of the Operation
The operation, dubbed Storm 2035, was traced back to the Iranian government and encompassed a network of accounts across multiple platforms, with a significant presence on X and Instagram. Despite the efforts invested, the campaign failed to gain traction, as the majority of its posts failed to garner meaningful engagement in the form of likes, shares, or comments.
Content and Targets
The campaign also generated content surrounding Israel's involvement in the Olympics, political dynamics in Venezuela, the rights of Latin American communities, and the topic of Scottish independence. OpenAI emphasized that the operation attempted to appeal to both ends of the political spectrum within the United States, yet it encountered indifference and received no notable backing from either side.
For more information, you can visit the Open AI website.
