AI BriefingAnthropic News news
The latest AI and developer news from Anthropic News, with short summaries.
Visit website →An Approach to Understanding and Addressing AI Harms
Anthropic has unveiled a new framework for managing a wide range of AI harms, including physical, psychological, and economic impacts.
Lawrence Livermore National Laboratory Expands Claude for Enterprise Adoption to Support Scientists
Lawrence Livermore National Laboratory (LLNL) is expanding its enterprise-wide adoption of Claude for Enterprise to support approximately 10,000 researchers.
Healthcare and Life Sciences
Anthropic has significantly expanded Claude's capabilities to support research and clinical work in healthcare and life sciences.
Golden Gate Claude
Anthropic has published interpretability research that identifies internal features within Claude and demonstrates that adjusting them changes the model's behavior.
Conducting Frontier Threat Red Teaming for AI Safety
Anthropic conducted red team testing with experts to evaluate AI frontier threats directly tied to national security, such as biological risks.
Frontier Model Security
Anthropic proposed robust security frameworks such as 'two-party control' and national-level infrastructure protection to prevent theft and misuse of frontier AI models.
The Economic Future of the UK and Europe
Anthropic is expanding its 'Economic Futures Programme' to the UK and Europe to study AI's impact on the labor market and economy.
Building AI Nuclear Safety Guardrails Through Public-Private Partnership
Anthropic partnered with the U.S. Department of Energy (DOE) to develop a classifier that detects AI-related nuclear risks and applied it to Claude.
Detecting and Responding to Misuse Cases of Claude (March 2025)
Anthropic disclosed cases of detecting public opinion manipulation and malicious activity using Claude, along with its response measures.
Core Perspectives on AI Safety
Anthropic emphasized that multifaceted AI safety research is urgently needed to prepare for rapid AI advancement and its accompanying risks.
Claude Sonnet 4.5
Anthropic has released Claude Sonnet 4.5, a model with greatly improved capabilities in coding, agent building, and computer use.
Claude Opus 4.5 Unveiled
Anthropic has unveiled **Claude Opus 4.5**, its next-generation model with greatly enhanced coding, agentic, and computer use capabilities.
Claude Haiku 4.5
Anthropic has released a new small model, Claude Haiku 4.5, offering excellent cost-efficiency and speed.
Claude for Nonprofits
Anthropic has launched 'Claude for Nonprofits', which includes discounts of up to 75% and dedicated connectors to help maximize the impact of nonprofit organizations.
Claude for Life Sciences Research
Anthropic has improved Claude's performance for life sciences research and introduced scientific platform Connectors and Agent Skills.
Building a Path Toward AI Accountability
Anthropic proposed measures to ensure AI accountability, including model evaluations, risk-based assessments, and pre-registration of large-scale training runs.
The Challenges of Red Teaming AI Systems
This analyzes the various Red Teaming methodologies used to secure the safety of AI systems, along with their respective pros, cons, and challenges.
Building Safeguards for Claude
Anthropic operates a multi-faceted Safeguards strategy to prevent misuse of Claude and ensure its beneficial use.
Anthropic's Responsible Scaling Policy
Anthropic has announced a 'Responsible Scaling Policy (RSP)' that introduces AI Safety Levels (ASL) to manage catastrophic risks from AI.
Anthropic Signs Partnership with Teach For All
Anthropic is partnering with Teach For All to support Claude-based AI education for more than 100,000 teachers across 63 countries worldwide.