OpenAI has revealed that an Israeli firm attempted to disrupt the Indian Lok Sabha elections by using AI-generated content to criticize the ruling BJP and praise the opposition Congress party. The activity was flagged in May, just days before the election results are due to be announced.
According to OpenAI’s threat intelligence report, the network operated by STOIC, a political campaign management firm based in Israel, focused on manipulating public opinion by generating comments and posts that targeted the Indian elections. This covert operation aimed to influence voter sentiment against the BJP.
The report details how AI models were employed to produce content critical of the BJP while commending the Congress party. This activity was part of a broader strategy to sway public perception in the lead-up to the elections.
OpenAI, in collaboration with Microsoft Threat Intelligence, identified and disrupted this malicious activity. The company emphasized its commitment to detecting and preventing the misuse of AI for such purposes. This incident underscores the growing concern over the use of artificial intelligence in political manipulation and the importance of robust measures to safeguard electoral integrity.