Earlier this year, South Australia’s Department for Education decided to bring generative AI into its classrooms. But before they opened the doors, one question loomed large: how to do it responsibly?

Mindgard announced the detection of two security vulnerabilities within Microsoft’s Azure AI Content Safety Service. The vulnerabilities enabled an attacker to bypass existing content safety measures ...

Microsoft has introduced new features in Azure AI Content Safety, including the ability to correct hallucinated AI-generated content and a hybrid deployment option for on-device scenarios. Azure AI ...

Microsoft Azure AI Content Safety has received a new Multimodal API that allows it to detect harmful and inappropriate content made by users or by AI. Microsoft has made several announcements today ...

New tools for filtering malicious prompts, detecting ungrounded outputs, and evaluating the safety of models will make generative AI safer to use. Both extremely promising and extremely risky, ...

New tools for detecting prompt injection attacks and hallucinations and for ensuring model safety are coming to Azure AI Studio. Microsoft is adding safety and security tools to Azure AI Studio, the ...