OpenAI's Strategies to Counter Deceptive Use of AI

OpenAI, an Artificial Intelligence company founded by Sam Altman, recently made public its actions to combat global campaigns that were misusing its technology for deceptive purposes. On May 30, OpenAI announced the termination of accounts engaged in covert influence operations.

According to an official statement by OpenAI, in the past three months, the company has disrupted five covert influence operations that aimed to exploit their models for deceptive activities on the internet.

These malicious actors utilized AI to fabricate comments on articles, create fake social media profiles, and translate and proofread text content.

One particular operation called “Spamouflage” employed OpenAI's tools to conduct social media research, generate multilingual content on platforms like X, Medium, and Blogspot, with the intention to manipulate public opinion and influence political outcomes. The group also utilized AI for debugging code and managing databases and websites.

Another campaign, named “Bad Grammar,” targeted regions such as Ukraine, Moldova, the Baltic States, and the United States. This operation utilized OpenAI models to run Telegram bots and generate political content to influence outcomes. Additionally, the group employed AI for debugging code and managing databases and websites.

A third group known as “Doppelganger” utilized AI to create comments in multiple languages including English, French, German, Italian, and Polish. These comments were posted on platforms like X and 9GAG with the aim of manipulating public opinion.

The “International Union of Virtual Media” (IUVM), an Iranian threat actor, has persistently posted web content supporting Iran while criticizing Israel and the US.

Moreover, a commercial entity in Israel, STOIC, has been producing content related to the Gaza conflict, the Histadrut trade unions organization in Israel, and the Indian elections. This particular operation, dubbed “Zero Zeno” in reference to the Stoic school of philosophy founder, has garnered minimal engagement with its campaigns.

OpenAI emphasized that the content shared by these operations covered a broad spectrum of issues including Russia's invasion of Ukraine, the conflict in Gaza, the Indian elections, European and US politics, and criticisms of the Chinese government by Chinese dissidents and foreign nations.

Ben Nimmo, a principal investigator at OpenAI who authored the report, informed The New York Times, “Our case studies offer insights into some of the most extensively reported and enduring influence campaigns currently active.”

Despite these disclosures, OpenAI highlighted that these operations did not achieve substantial audience engagement or expanded reach through their services.