OMM Redaksiyası
“OpenAI” said it banned accounts linked to covert influence operations originating in Russia and Iran that used its AI tools in efforts to shape public opinion abroad.
In a blog post published Thursday, the company said the campaigns combined AI with traditional influence methods, including seemingly legitimate research platforms and fabricated journalist identities.
“OpenAI” said it banned clusters of “ChatGPT” accounts associated with both operations and shared information about them with relevant authorities. The company said the campaigns resembled influence operations from before the AI era, but used AI to make some tasks easier.
Iran-linked operation used fabricated journalist profiles
The Iran-linked operation used seven invented journalist identities to pitch long-form articles to online publications, according to the company. “OpenAI” identified almost 100 articles published under those bylines across more than a dozen outlets.
The articles addressed international affairs, geopolitics and the Middle East. Their publication increased sharply after the outbreak of the US-Iran conflict, the company said.
Operators used “ChatGPT” to edit articles, check drafts against individual outlets’ submission requirements and prepare pitches to editors. The profiles presented themselves as Western journalists specialising in international politics, the Middle East, human rights or conflict.
The operation also generated comments in English and Persian on social media. The company said the requested comments addressed subjects including missile strikes, ceasefires and diplomacy, and were framed to portray the US as an aggressor, Israel as an unreliable US ally, and Iran as a defiant but resilient victim.
However, the comments identified by “OpenAI” generally attracted little engagement. Some likes came from accounts linked to the operation itself, it said.
Russian-linked campaign focused on Ukraine and Latin America
The Russian-linked operation targeted audiences across Latin America, with much of its activity apparently aimed at undermining Ukraine’s reputation. Other efforts focused on domestic politics in Argentina and Bolivia, the company said.
The campaign allegedly used the Social Research Center, a self-described platform covering Latin America. Internal reports suggested Russian operators controlled it through a fabricated persona called “Mia Clark,” according to “OpenAI”.
Available evidence indicated that employees working for the platform in Latin America did not know they were working for a Russian group. They appeared to interview regional experts and produce research papers in good faith, the company said.
Internal reports described efforts to circulate fabricated stories, documents and audio recordings targeting Ukraine and political figures in the region. “OpenAI” said it found material corresponding to some reported campaigns, including false claims that prompted fact-checks and official denials, but cautioned that it could not corroborate some claims in the reports.
Both operations placed content in media outlets, potentially extending their reach beyond fake social media accounts, the company said. It published its findings to help researchers and other organisations identify and disrupt similar campaigns.
