OpenAI said the campaign did not appear to generate significant audience participation.
Washington:
OpenAI announced on Friday that it had blocked the accounts of an Iranian group because it used its chatbot ChatGPT to generate content intended to influence the U.S. presidential election and other issues.
The operation, dubbed “Storm-2035,” used ChatGPT to generate content on topics such as comments about candidates from both sides in the U.S. election, the Gaza conflict, and Israel’s participation in the Olympics, and then distributed it across social media accounts and websites.
Research by the Microsoft-backed AI company found that ChatGPT was used to create long articles and shorter social media comments.
OpenAI said the campaign did not appear to generate significant audience participation.
The majority of social media posts identified received few or no likes, shares or comments, and the company found no evidence that web articles were shared on social media.
The accounts have been banned from using OpenAI’s services and the company continues to monitor activity for further attempts to violate its policies, it said.
In early August, a threat intelligence report from Microsoft said that the Iranian network Storm-2035 – which consists of four websites disguised as news agencies – is actively interacting with US voting groups at opposite ends of the political spectrum.
Engagement is being built through “polarizing messages on issues such as the US presidential candidates, LGBTQ rights and the Israel-Hamas conflict,” the report says.
Democratic candidate Kamala Harris and her Republican rival Donald Trump are in a neck-and-neck race ahead of the presidential election on November 5.
The AI company said in May that it had thwarted five covert influence operations that attempted to use its models for “deceptive activities” across the internet.
(Except for the headline, this story has not been edited by NDTV staff and is published from a syndicated feed.)
