Insecure output handling In the previous examples, we learned about a few various security risks, threats, and exploits, especially against …
What is privacy? – Security and Privacy Considerations for Gen AI – Building Safe and Secure LLMs
What is privacy? The National Institute of Standards and Technology (NIST) part of the US Department of Commerce defines privacy …
Securing data in the generative AI era – Security and Privacy Considerations for Gen AI – Building Safe and Secure LLMs
Securing data in the generative AI era As with any other technology, ensuring security and data protection is important. As …
Auditing – Security and Privacy Considerations for Gen AI – Building Safe and Secure LLMs
Auditing Oftentimes, we hear words that generally may have a negative connotation around them. For many, the word “audit” or …
Reporting – Security and Privacy Considerations for Gen AI – Building Safe and Secure LLMs
Reporting Reporting is a fairly simple concept, and it means exactly what the name implies, so we will not delve …
Managed identities – Security and Privacy Considerations for Gen AI – Building Safe and Secure LLMs
Managed identities Azure OpenAI supports Microsoft Entra ID, which is the fairly newly rebranded Azure Active Directory (Azure AD) service. …
Understanding responsible AI design 2 – Responsible Development of AI Solutions: Building with Integrity and Care
Key principles of RAI – Responsible Development of AI Solutions: Building with Integrity and Care
Key principles of RAI Figure 9.1 – Responsible AI principles Microsoft has established a Responsible AI Standard, presenting a comprehensive …
Azure OpenAI service API keys – Security and Privacy Considerations for Gen AI – Building Safe and Secure LLMs
Azure OpenAI service API keys The Azure OpenAI service itself, along with OpenAI, uses API keys for applications to access …
Transparency – Responsible Development of AI Solutions: Building with Integrity and Care
Transparency This principledemands clarity on how AI systems make decisions or reach conclusions. For example, a credit scoring AI system …