Search Anthropic

Preparing for the 2026 US midterms
Earlier this year, we shared an update on our election safeguards. Here, we outline how we’ve continued to strengthen these safeguards ahead of this year’s US midterms.
Claude’s values across models and languages
New research into how the values expressed by Claude vary across the top 20 languages on Claude.ai.
Coding agents in the social sciences
Results from a survey of 1,260 social scientists about AI and coding agent use.
An update on our election safeguards
We explain what we’re doing to ensure Claude plays a positive role in the US midterms and other major elections around the world this year.
A “diff” tool for AI: Finding behavioral differences in new models
"Model diffing" is a powerful way to understand how models change during fine-tuning. Our new research paper extends this method to its most challenging and general use case: comparing models with entirely different architectures.
Measuring political bias in Claude
We introduce and open-source an automated evaluation for political even-handedness, testing six models across 1,350 prompt pairs.
A statement from Dario Amodei on Anthropic's commitment to American AI leadership
A statement from Anthropic CEO, Dario Amodei, on Anthropic’s commitment to advancing America's leadership in building powerful and beneficial AI
Building safeguards for Claude
How Anthropic's Safeguards team protects users, from policy and pre-deployment evaluations to real-time classifiers.
Evaluating feature steering: A case study in mitigating social biases
Our preliminary (mixed) findings on how feature steering might play a role in creating safer model outputs.
How character training cultivates traits like curiosity, open-mindedness, and thoughtfulness in Claude using a variant of Constitutional AI.
Evaluating and mitigating discrimination in language model decisions
We present a method for proactively evaluating the potential discriminatory impact of language models in a wide range of use cases, including hypothetical use cases where they have not yet been deployed.
Collective Constitutional AI: Aligning a language model with public input
Anthropic and the Collective Intelligence Project recently ran a public input process involving ~1,000 Americans to draft a constitution for an AI system. In this post, we share the resulting publicly sourced constitution, as well as what happened when we trained a new AI system against it using Constitutional AI.
Challenges in evaluating AI systems
Insights from the red teaming approaches we've used to test our AI systems.
Towards measuring the representation of subjective global opinions in language models
In this paper, we develop a quantitative framework to evaluate whose opinions model-generated responses are more similar to.
The capacity for moral self-correction in large language models
We test the hypothesis that language models trained with reinforcement learning from human feedback have the capability to morally self-correct if instructed to do so.
Anthropic is a Public Benefit Corporation. Our mission is the responsible development and maintenance of advanced AI for the long-term benefit of humanity.