OpenAI Strengthens ChatGPT Security with Lockdown Mode and Elevated Risk Labels
With the introduction of Lockdown Mode and Elevated Risk labels, OpenAI aims to arm users against potential vulnerabilities, such as prompt injection attacks that threaten to compromise sensitive data.
Google Strengthens Gemini Security with Multi-Layered Defense Against Prompt Injection
Google strengthens security in its Gemini gen-AI systems with a multi-layered defense mechanism against indirect prompt injection attacks.