US tech firm OpenAI says it identified five groups, based in various countries, that have been using its technology in attempts to manipulate public opinion. The firm developed the artificial intelligence chatbot ChatGPT.

OpenAI on Thursday reported the latest trends in covert influence operations. It said five groups, based in Russia, China, Iran and Israel, used generative AI for a range of tasks, such as making up names and bios for social media accounts, creating comments and translating texts.

The content posted by the groups focused on a wide range of issues, including Russia's invasion of Ukraine, the conflict between Israel and the Islamic group Hamas in the Gaza Strip, and politics in the United States.

The firm said a group based in China generated texts in English, Japanese and other languages that criticized Japan's release of treated and diluted water from the damaged Fukushima Daiichi nuclear power plant into the ocean. The texts were posted on online platforms.

The report points out the risk of influence operations by generative AI ahead of the US presidential election and other events.