OpenAI Blocks Iranian Group Accounts Using ChatGPT To Target US Elections

OpenAI has closed accounts belonging to groups from Iran (photo: dock. pexels)

JAKARTA - OpenAI announced on Friday August 16 that it has closed accounts belonging to Iranian groups that use ChatGPT chatbots to produce content aimed at influencing US presidential elections and other issues.

The operation, identified as Storm-2035, uses ChatGPT to produce topic-focused content such as comments about candidates in US elections, conflicts in Gaza, as well as Israel's presence at the Olympics, which are then shared through social media accounts and websites.

Investigations carried out by Microsoft-backed AI companies show that ChatGPT is used to produce long articles and short comments on social media. However, OpenAI stated that the operation did not succeed in achieving significant audience engagement. Most of the identified social media posts get little or no response at all, such as likes, shares, or comments, and there is no indication that the web articles were shared on social media.

The accounts have been banned from using OpenAI services, and the company continues to monitor activities to prevent further policy violations.

In early August, a threat intelligence report from Microsoft stated that Iran's Storm-2035 network, which consists of four websites posing as news outlets, actively interacting with US voters at both ends of the political spectrum with divisive messages about issues such as US presidential candidates, LGBTQ rights, and Israeli-Hamas conflicts.

The AI company previously, in May, has also revealed that it has disrupted five covert influence operations seeking to use their model for "misleading" activity across the internet.