Microsoft has launched Prompt Shields, a new security feature for its Azure OpenAI Service.
Prompt Shields is designed to safeguard AI-driven applications across the Azure platform. It protects against direct (jailbreak) and indirect prompt injection attacks, also known as cross-domain prompt injection attacks. The feature helps identify attacks on large language models (LLMs), enhancing the security of AI applications.
Microsoft claims that Prompt Shields will help prevent unauthorized actions, such as retrieving sensitive data or visiting harmful websites, which could compromise security. However, the company notes that the feature's efficiency may be limited to certain languages, as it was developed and tested using Chinese, English, French, German, Spanish, Italian, Japanese, and Portuguese.
By using this site, you agree to allow SPEEDA Edge and our partners to use cookies for analytics and personalization. Visit our privacy policy for more information about our data collection practices.