Search Anthropic

GLM-5.3 and the spread of advanced cyber capabilities
Like Claude Mythos Preview, GLM-5.3 has strong capabilities for autonomously building end-to-end cyber exploits. But unlike other frontier models, GLM-5.3 has been released without meaningful safeguards to limit misuse.
Improving Fable 5's biology safeguards
We’re making updates to Claude Fable 5’s biology safeguards in a way that substantially reduces false positives. Fable 5 users will now experience many fewer “fallbacks”—where the system switches to a less capable model after they make a biology-related query.
More details on Fable 5’s cyber safeguards and our jailbreak framework
What is and isn't blocked by our cyber classifiers, and a first draft of our jailbreak severity framework
Fable 5 returns globally July 1. We're also proposing an industry-wide framework for scoring jailbreak severity, together with Amazon, Microsoft, Google, and other Glasswing partners.
Statement on the US government directive to suspend access to Fable 5 and Mythos 5
The US government has issued an export control directive to suspend all access to Fable 5 and Mythos 5.
How we contain Claude across products
As agents grow more capable, so does their potential blast radius. The engineering question is how to cap it. Here’s what we’ve learned building containment for claude.ai, Claude Code, and Cowork.
How we built Claude Code auto mode: a safer way to skip permissions
Claude Code users approve 93% of permission prompts. We built classifiers to automate some decisions, increasing safety while reducing approval fatigue. Here's what it catches, and what it misses.
The assistant axis: situating and stabilizing the character of large language models
In a new paper, we look at several open-weights language models, map out how their neural activity defines a “persona space,” and situate the “Assistant” persona within that space.
Next-generation Constitutional Classifiers: More efficient protection against universal jailbreaks
Constitutional Classifiers++ improve on our previous approach, yielding a system that is even more robust, has a much lower refusal rate, and is dramatically cheaper to run.
Mitigating the risk of prompt injections in browser use
Claude Opus 4.5 achieves a 1% attack success rate against prompt injections in browser use, and Claude for Chrome moves to beta.
Disrupting the first reported AI-orchestrated cyber espionage campaign
In September 2025, we detected and disrupted a highly sophisticated cyber espionage campaign. We’re sharing this case publicly to help others strengthen their own defenses.
Strengthening our safeguards through collaboration with US CAISI and UK AISI
Over the past year, we've collaborated with the US Center for AI Standards and Innovation (CAISI) and UK AI Security Institute (AISI), government bodies established to measure and improve the security of AI systems.
Activating AI Safety Level 3 protections
We've activated ASL-3 Deployment and Security Standards from our Responsible Scaling Policy alongside the launch of Claude Opus 4.
Testing our safety defenses with a new bug bounty program
A new bug bounty program with HackerOne challenges red-teamers to find universal jailbreaks in safety classifiers we haven't yet deployed.
Tracing the thoughts of a large language model
Today, we're sharing two new papers that represent progress on the development of an “AI microscope,” and the application of it to see new “AI biology.”
Claude models help power Alexa+, Amazon's new premium voice assistant, delivered via Amazon Bedrock with Claude's safety features built in.
Forecasting rare language model behaviors
Anthropic research on predicting rare, undesirable AI behaviors
Constitutional Classifiers: Defending against universal jailbreaks
A paper from Anthropic describing a new way to guard LLMs against jailbreaking.
Expanding our model safety bug bounty program
We're expanding our bug bounty program with a new initiative focused on finding flaws in the mitigations we use to prevent misuse of our models.
Many-shot jailbreaking
A long-context jailbreak that overrides safety training with hundreds of faux dialogues, and the mitigations that reduce its effectiveness.
Releasing Claude Instant 1.2
Claude Instant 1.2 is a faster, lower-cost model with improved math, coding, reasoning, and safety inherited from Claude 2.