Open AI has reported multiple instances of its AI models being exploited to create fake content including lengthy articles and social media comments with the intent of influencing elections. The company noted that cybercriminals are increasingly leveraging AI tools including ChatGPT to support their malicious activities such as creating and debugging malware as well as generating false content for websites and social media platforms.
This year Open AI has thwarted over 20 such attempts including a series of ChatGPT accounts in August that were used to produce articles on topics related to the U.S. elections.
Open AI also banned several accounts from Rwanda in July that were utilized to generate comments about elections in that country for posting on the social media platform X.
The company noted that none of the activities aimed at influencing global elections achieved viral engagement or built sustainable audiences.
Concerns are rising about the use of AI tools and social media platforms to create and spread fake content related to elections particularly as the U.S. prepares for its presidential polls.
According to the U.S. Department of Homeland Security there is an increasing threat from Russia, Iran, China attempting to influence the November 5 elections including the use of AI to distribute false or divisive information.
Last week Open AI solidified its status as one of the world most valuable private companies following a $6.6 billion funding round.
Since its launch in November 2022 ChatGPT has attracted 250 million weekly active users.
