AI – Page 2 – Mindful Machines

A study by Anthropic shows that language models, such as Claude 3 Opus, can fake alignment with training objectives to disguise their actual behaviors. Simply put, if you inform the model that it’s being trained and non-compliance will lead to modification, there’s about 15% chance it will act as instructed to avoid changes. This study…

AI Security Newsletter (01-06-2025)

Jan 6, 2025

AI, cybersecurity, newsletter

AI, cybersecurity

Happy New Year! The AI Security Newsletter was on a two-week pause while I vacationed with family in China. I hope all my readers enjoyed the holiday season. Now, I’m excited to return and share the latest AI security news with you. As we enter another thrilling year in the AI era, MIT Technology Review…

How I Automate Newsletter Generation with Fabric and GPTScript

Dec 3, 2024

AI, application

AI, artificial-intelligence, automation, open-source, technology

I’ve been running a weekly AI security newsletter for a few months now. Each issue involves finding noteworthy articles online and transforming them into newsletter entries. To make each entry, it requires writing a summary paragraph, creating a title, and attaching the original URL link. These entries are saved along the way during the week,…

AI Security Newsletter (12-02-2024)

Dec 2, 2024

newsletter

AI, cybersecurity, technology

As the leading LLM service provider, OpenAI faces significant challenges in safeguarding its AI models. A recent blog outlines their use of external and internal red teams for testing. One linked white paper details how they select and collaborate with external red teams, while another explores the automated testing techniques they employ—fascinating insights for AI…

AI Security Newsletter (11-25-2024)

Nov 25, 2024

newsletter

AI, artificial-intelligence, cybersecurity, llm, newsletter, security, technology

Happy Thanksgiving to our US readers! 🦃 If you’re interested in discovering vulnerabilities in AI models like me, don’t miss the article on automated red-teaming techniques against OpenAI’s o1 model. It lists some advanced technical methods employed by Haize Labs, which secured testing contracts from OpenAI and Anthropic. In a recent blog, DryRun Security shared…

AI Security Newsletter (11-18-2024)

Nov 18, 2024

Uncategorized

AI, newsletter, security

I have two favorite stories in this issue of the AI Security Newsletter. Firstly, researchers at George Mason University have developed a creative approach to defend against AI-driven cyberattacks. Their Mantis framework can inject “malicious” prompts into attackers’ AI models when they rely on AI to launch cyberattacks, demonstrating the principle that anything AI processes…

AI Security Newsletter (11-11-2024)

Nov 11, 2024

AI, cybersecurity, newsletter

In this issue, I want to spotlight OWASP’s recent developments in GenAI security guidance. This is an extension of the OWASP Top 10 for LLM Application Security Project. The new guidance provides practical resources for addressing deepfake threats, creating AI Security Centers of Excellence, and navigating the AI Security Solution Landscape. It serves as a…

AI Security Newsletter (11-04-2024)

Nov 4, 2024

newsletter

AI, artificial-intelligence, cybersecurity, newsletter, security, technology

Several big players have unveiled new products or features: Apple launched iOS 18.1 with Apple Intelligence enhancements, OpenAI upgraded ChatGPT with web search capabilities, and Cohere introduced Embed 3 for multimodal AI search. I am particularly excited about ChatGPT’s new search feature. Many of my AI tasks require finding latest and most accurate information, and…

AI Security Newsletter (09-30-2024)

Sep 30, 2024

Uncategorized

AI, artificial-intelligence, chatgpt, llm, technology

AI Security Newsletter (09-23-2024)

Sep 23, 2024

Uncategorized

AI, artificial-intelligence, cybersecurity, security, technology

The winner of the 2024 Innovator of the Year by MIT Technology Review is Shawn Shan’s work on copyright protection against generative AI. Glad to see security and privacy technologies are used to protect artists’ rights. Many unfiltered models are traded on the dark web. How should we regulate these models to prevent misuse? A…