Microsoft uncovers global hacking ring that used AI for harmful content

Microsoft has said that it has discovered a global cybercrime network that bypassed safety measures on its generative AI tools, including Azure OpenAI, to produce illicit content, such as non-consensual intimate images.

According to a report by Bloomberg, the tech giant identified hackers based in multiple countries – from Iran, the UK, Hong Kong, and Vietnam – who leveraged stolen customer login credentials to gain unauthorised access to its AI services.