NYT: State-backed, private groups used AI agents to manipulate online discourse - The Jerusalem Post

An investigation by The New York Times, reported on by The Jerusalem Post, states that groups in China, Iran and Israel used open-source artificial intelligence to deploy thousands of autonomous bots on social media. The article describes the activity as an attempt to manipulate online discourse, and it identifies the groups involved as both state-backed and private. The bots are described in the investigation as numbering in the thousands, and the AI tools they used are described as open-source.

According to the article, US officials told The New York Times that the instances marked one of the first times AI agents from Chinese-produced models were used to handle every stage of an online influence campaign. The stages described include creating new fake accounts, coordinating messaging, and framing discussion online. The Jerusalem Post presents the findings as the product of a New York Times investigation into the use of open-source AI by groups in China, Iran and Israel.

The article attributes the assessment of the campaigns to US officials, who described the role of Chinese-produced model agents across the full process of an influence operation, from the creation of fake accounts through to the coordination of messaging and the framing of online discussion. It states that groups in China, Iran and Israel conducted the activity using open-source AI. The Jerusalem Post's account of the investigation centres on the use of autonomous AI agents, rather than human operators, at each stage of the campaigns.