OpenAI recently disclosed its successful identification and disruption of internet campaigns using its technology for covert influence operations globally.
OpenAI, a company that specializes in artificial intelligence, recently disclosed that it successfully identified and disrupted a number of internet campaigns that used its technology to manipulate public opinion all around the world.
OpenAI Disrupts Covert Influence Operations
Sam Altman’s artificial intelligence startup OpenAI announced on the 30th of May that it had “terminated accounts linked to covert influence operations.”The malicious actors utilized artificial intelligence to make comments for articles, construct names and bios for social media accounts, translate and proofread materials, and build social media profiles.
An operation known as “Spamouflage” reportedly employed OpenAI to conduct research on social media and publish multilingual content across several platforms such as X, Medium, and Blogspot in an effort to “manipulate public opinion or influence political outcomes,” according to the company that is responsible for ChatGPT. It employed AI not only to handle databases and webpages, but also to debug code.
The United States of America, Ukraine, Moldova, and the Baltic States were the targets of an operation known as “Bad Grammar,” which utilized its models to run Telegram bots and generate political remarks. In addition, it was stated that another organization known as Doppelganger utilized artificial intelligence models to generate comments in English, French, German, Italian and Polish.
They specifically targeted X and 9GAG with these comments to sway public opinion. The “International Union of Virtual Media,” a group OpenAI also cited, used the technology to create long-form articles, headlines, and website content for their linked website.
OpenAI similarly affected the commercial company known as STOIC, which used artificial intelligence to generate articles and comments on social media platforms like Instagram, Facebook and X, as well as websites affiliated with the operation.
According to Ben Nimmo, a principle investigator for OpenAI who authored the paper, “Our case studies provide examples from some of the most widely reported and longest-running influence campaigns that are currently active,” he said in an interview with The New York Times.
The outlet also noted that this was the first time a significant artificial intelligence company had revealed how it used its unique capabilities for online fraud. According to their conclusion, these activities do not currently appear to have benefited from significantly improved audience engagement or reach as a result of our services.