AI Safety Diary: October 2, 2025
A diary entry on the 8th chapter of the AI Safety Book, providing a deep dive into the challenges and potential solutions in AI governance, from corporate self-regulation to international treaties.
AI Safety Diary: October 1, 2025
A diary entry on the 7th chapter of the AI Safety Book, which analyzes AI development through the lens of collective action problems, such as arms races and the tragedy of the commons.
AI Safety Diary: September 30, 2025
A diary entry on the 6th chapter of the AI Safety Book, which delves into the philosophical and technical challenges of machine ethics and ensuring that AI systems not only avoid harm but actively promote beneficial outcomes.
AI Safety Diary: September 29, 2025
A diary entry on the 5th chapter of the AI Safety Book, exploring the safety challenges that arise from the interaction of multiple AI agents and the emergent properties of complex AI ecosystems.
AI Safety Diary: September 28, 2025
A diary entry on the 4th chapter of the AI Safety Book, which discusses the engineering principles required to build robust and reliable AI systems, drawing parallels with traditional safety engineering fields.
AI Safety Diary: September 27, 2025
A diary entry on the 3rd chapter of the AI Safety Book, focusing on the core challenges of single-agent safety, such as specifying correct reward functions and preventing unintended behaviors in a single AI system.
AI Safety Diary: September 26, 2025
A diary entry on the 2nd chapter of the AI Safety Book, which provides a technical introduction to the fundamentals of AI, including machine learning, neural networks, and deep learning concepts.
AI Safety Diary: September 25, 2025
A diary entry on the 4th chapter of the AI Safety Atlas, focusing on the critical area of AI governance and the challenges of creating effective policies and institutions to manage AI development globally.
AI Safety Diary: September 24, 2025
A diary entry on the 3rd chapter of the AI Safety Atlas, which covers the different high-level strategies being pursued to mitigate AI risks, including technical alignment, policy, and strategy research.
AI Safety Diary: September 23, 2025
A diary entry on the 2nd chapter of the AI Safety Atlas, which provides a comprehensive overview of the various catastrophic risks associated with advanced AI, from misuse to structural issues.