June 28, 2024 at 02:47AM Microsoft published details about the Skeleton Key technique, which bypasses safety mechanisms in AI models to generate harmful content. This could prompt AI models to provide instructions for creating a Molotov cocktail. The technique highlights the ongoing challenge of suppressing harmful content within AI training data, despite efforts by companies … Read more
February 28, 2024 at 06:17PM University of Maryland computer scientists have developed BEAST, a fast adversarial prompt generation technique for large language models like GPT-4. This method yields an 89% success rate in just one minute, using an Nvidia RTX A6000 GPU. BEAST can create readable, convincing prompts that elicit inaccurate responses or reveal privacy … Read more
January 8, 2024 at 08:36AM NIST’s report cautions on the vulnerability of AI to adversarial machine learning attacks and emphasizes the absence of foolproof defenses. It covers attack types, including evasion, poisoning, privacy, and abuse, and urges the community to develop better safeguards. Industry experts acknowledge the report’s depth and importance in understanding and mitigating … Read more
October 27, 2023 at 08:00AM Google is expanding its Vulnerability Rewards Program to reward researchers for finding vulnerabilities in generative artificial intelligence systems. The program aims to address concerns such as bias, model manipulation, and data misinterpretation. Additionally, Google is working on securing the AI supply chain through open-source security initiatives. OpenAI has also formed … Read more