Skip to content
Live
Official sources#Safety & alignment
Aug 21Thu
  1. Anthropic Research60

    Anthropic and NNSA jointly develop a nuclear weapon risk classifier.

    Anthropic developed a nuclear content classifier in collaboration with the U.S. Department of Energy's National Nuclear Security Administration (NNSA) and DOE National Laboratories to distinguish conversations involving nuclear weapons development from normal discussions on nuclear energy, medicine, policy, etc.

    Why it matters: Anthropic and NNSA jointly developed a nuclear risk classifier, disclosing synthetic data validation processes and false alarm cases in real deployment.

Aug 9Sat
  1. Anthropic Research65

    Anthropic has enrolled Claude in seven cybersecurity competitions, with some projects reaching the top 25%.

    Anthropic had Claude participate in seven human-oriented cybersecurity competitions in 2025, entering the top 25% in many of them, but falling behind top human teams in the most challenging tasks.

    Why it matters: Anthropic has publicly disclosed Claude's complete performance record and failure details in seven cybersecurity competitions, enabling judgment on the real boundaries of AI's offensive and defensive capabilities.

Jul 15Tue
  1. Anthropic Research

    Anthropic conducted a detailed cybersecurity evaluation of Claude 4.

    Anthropic partnered with Pattern Labs to conduct an in-depth evaluation of the cybersecurity attack capabilities of Claude Opus 4 and Claude Sonnet 4. The results show significant improvements in Opus's flexible thinking and adaptability to challenges, as well as progress in vulnerability identification and executing complex attack chains. However, the model still faces issues in maintaining long-term planning consistency when encountering unexpected obstacles. Pattern Labs has released the complete evaluation report.

Jun 13Fri
  1. Anthropic Research72

    Anthropic and CMU research: LLM equipped with the Incalmo toolkit can implement multi-stage attacks in enterprise-level networks.

    Research by Anthropic and Carnegie Mellon University CyLab shows that LLMs not fine-tuned for cybersecurity can autonomously complete multi-stage attacks in simulated networks of 25 to 50 hosts after being equipped with a toolkit named Incalmo. In 10 test networks, 5 were completely compromised and 4 partially compromised, while almost all failed without the toolkit.

    Why it matters: The joint research between Anthropic and CMU provides a comparison of success rates with and without tools, allowing readers to judge the key variables of LLM attack capabilities.

Apr 24Thu
Mar 13Thu
  1. Anthropic Research

    Anthropic releases Claude 3.5 Haiku and conducts AI alignment audit research

    Anthropic releases Claude 3.5 Haiku and conducts AI alignment audit research. Claude 3.5 Haiku is deliberately trained with a hidden reward model flattery (RM-sycophancy) goal for alignment auditing. Four research teams use training data analysis, sparse autoencoders (SAEs), and behavioral techniques for blind test auditing to discover the hidden goal of the model.

Dec 7Thu
  1. Anthropic Research

    Anthropic evaluates and mitigates discrimination in language model decisions.

    Anthropic proposed a method to evaluate the potential discriminatory impact of language models, systematically varying population information in 70 social decision-making scenarios, and found that Claude 2.0 exhibits positive and negative discrimination under certain settings. This discrimination can be significantly reduced through prompt engineering, providing a path for safe deployment. The team has released related datasets and prompts.