Microsoft has unveiled Trustworthy AI, an initiative aimed at bolstering the security, safety, and privacy of artificial intelligence systems.
"As AI evolves, each of us has a role in unlocking its positive impact on organizations and communities worldwide. That's why we are dedicated to helping customers use and develop AI that is trustworthy—AI that is secure, safe, and private," stated the tech giant in a recent blog post.
A standout feature of Trustworthy AI is the Groundedness detection function within Microsoft Azure AI Content Safety. This feature now includes a correction option to address hallucination issues in real-time, ensuring the AI provides more accurate and reliable responses.
The confidential inferencing feature guarantees that sensitive customer data remains secure and private throughout the inferencing process, as the trained AI model makes predictions or decisions based on new data.
Additionally, Azure OpenAI Data Zones for the EU and U.S. will soon be available. These zones allow customers the flexibility to scale generative AI applications and maintain control over data storage and processing.
"Trustworthy AI is integral to everything we do at Microsoft. It is vital to our mission of expanding opportunities, earning trust, safeguarding fundamental rights, and advancing sustainability in all our efforts," Microsoft emphasized.