Overview

• Microsoft's 'Trustworthy AI' initiative includes new capabilities for AI security, safety, and privacy.
• The new Correction capability in Azure AI Content Safety aims to fix AI hallucinations in real-time.
• Confidential inferencing in Azure OpenAI Service ensures end-to-end privacy for sensitive data.
• Microsoft Copilot will now show exact web search queries used to generate AI responses, enhancing transparency.
• New tools in Azure AI Studio help developers assess and mitigate generative AI risks.