Anthropic, an AI safety and research company, has raised $580 million in a Series B funding round to build large-scale experimental infrastructure and explore the safety properties of computationally intensive AI models. The company has made significant progress in making systems more steerable, robust, and interpretable, including developing techniques to improve language model performance and releasing datasets to help other research labs train models aligned with human preferences. Anthropic aims to develop technical components necessary for large-scale models with better implicit safeguards and tools to examine these models' safety features. The company is expanding its team and partnerships to explore policy and societal impacts of AI systems, with plans to focus on culture and governance as it scales.