Kroll’s LLM penetration testing has seen it analyze data sets of OpenAI models, non-public models and RAG systems. It has used this to produce an anonymized dataset that catalogs vulnerabilities from all LLM engagements. Kroll has found a worrying prevalence of prompt injection attacks in the LLM cases it has investigated and shares its findings in this briefing.

Welcome to the Protecto Snowflake Integration Demo, where we show you how to safeguard sensitive data using Protecto’s advanced AI-powered masking tools! In today’s world, businesses using Snowflake for AI and analytics face significant risks with sensitive information hidden within unstructured data like comments and feedback columns. Protecto provides a unique solution, precisely masking only the sensitive parts of your unstructured data while leaving the rest untouched, ensuring your datasets remain valuable for analysis.

In this week’s episode, we recap our main innovation week of the year, Birthday Week, and its many announcements. Host João Tomé is joined by Brendan Irvine-Broque (Group Product Manager) and Kelly May Johnston (Director, Product Management). We dive into some of the key announcements of the week, including the AI audit and major improvements in security, performance, and AI/ML support, such as enhanced threat detection and analytics, faster navigation with Speed Brain, and expanded free tools for all users.

Artificial intelligence (AI) is revolutionizing problem-solving and innovation in DevSecOps. Leveraging AI responsibly is essential for building secure, trustworthy systems. Governments are crafting policies to ensure AI's benefits while mitigating risks. Join JFrog's VP of Product Marketing, Jens Eckels, and BSA CEO, Victoria A. Espinel, as they explore how BSA’s Policy Solutions for Building Responsible AI provides a framework to achieve these goals, focusing on governance, innovation, and transparency.