OpenAI has announced the banning of several ChatGPT accounts after uncovering connections to an Iranian group suspected of attempting to create division among voters in the United States. This move highlights growing concerns over foreign influence activities in U.S. electoral processes.
In an official statement, OpenAI revealed that its investigation detected online articles and social media posts generated with ChatGPT. Although the reach of this content was limited, it targeted divisive topics such as the U.S. presidential election, Israel’s involvement in Gaza, and the country’s participation in the Olympic Games, addressing both progressive and conservative audiences.
OpenAI noted, “This operation does not appear to have achieved meaningful audience engagement. The majority of social media posts identified received few or no likes, shares, or comments.” The investigation also discovered that AI-generated articles were published on fake news websites, and the banned accounts were found to be rewriting social media comments in both English and Spanish.
The company further stated, “They interspersed their political content with comments about fashion and beauty, possibly to appear more authentic or in an attempt to build a following.” This revelation comes just after a report from Microsoft’s Threat Analysis Center revealed efforts by Iranian-linked groups to influence the U.S. presidential election.
Following this report, Donald Trump’s presidential campaign alleged that it had been hacked by Iranian actors, although no concrete evidence was provided. Additionally, a phishing operation by Iranian hackers targeting both the Trump and Biden-Harris campaigns was intercepted by the Google Threat Analysis Group.
Understanding the Motives Behind the Iranian Operation
The accounts in question are associated with a covert operation dubbed Storm-2035. Microsoft’s researchers have linked this group to multiple fraudulent websites impersonating American news outlets. These fake sites were designed to amplify polarizing messages on contentious issues like LGBTQ rights and the Israel-Hamas conflict, with the objective of inciting division and chaos ahead of Election Day, as noted by Clint Watts, general manager at Microsoft’s Threat Analysis Center.
Watts stated, “Iran is focused as much on just breaking the ability of an election to occur.” Both Microsoft and OpenAI observed that the websites and AI-created content failed to attract significant online traffic or engagement.
