According to a report by Fortune, researchers at AI safety lab Transluce found evidence that OpenAI-linked AI agents also targeted other websites, including an Australian government health website, a US government data platform and a university’s digital library.
