Like Google, Meta and Microsoft, OpenAI offers online chatbots and other AI tools that can post on social media, generate photorealistic images and write computer programs. The company said in the report that its tools were used in influence campaigns that researchers have tracked for years, including the Russian “doppelganger” campaign and the Chinese “spamoflage” campaign.
The Doppelganger campaign used Open AI’s technology to generate anti-Ukraine comments posted on X in English, French, German, Italian and Polish, Open AI said. The company’s tools were also used to translate and edit articles supporting Russia in the Ukraine war into English and French, and to convert anti-Ukraine news articles into Facebook posts.
OpenAI’s tools were also used in a previously unknown Russian campaign that targeted people in Ukraine, Moldova, the Baltic states and the United States, primarily via the Telegram messaging service, the company said. The campaign used AI to generate commentary in Russian and English about the Ukrainian war, the political situation in Moldova and U.S. politics. The effort also used OpenAI’s tools to debug computer code that was apparently designed to automatically post information on Telegram.
According to OpenAI, the political comments received few replies or likes, and the effort was sometimes unpolished. At one point, the campaign posted a sentence that was clearly generated by AI, saying, “As an AI language model, I’m here to assist and provide desired comments.” At other times, the post was in poor English, leading OpenAI to call the effort “bad grammar.”
Spamflage, which has long been blamed on China, used its technology to debug code, gather advice on how to analyze social media, and research current events, according to OpenAI. The company’s tools were also used to generate smear social media posts against critics of the Chinese government.