Anthropic Updates Claude Fable 5 Biology Safeguards to Cut False Positives

Anthropic Updates Claude Fable 5 Biology Safeguards to Cut False Positives Anthropic has refined Claude Fable 5's biology safety classifiers to substantially decrease false positives, reducing biology-related fallbacks to less capable models by approximately 85% in testing. Testing showed an approximate 85% reduction in biology-related fallbacks, expanding user access for everyday health and educational questions. Dual-use research tasks, including virology, toxicology, and molecular design, remain restricted and continue falling back to Opus 5. Anthropic refined the classifier by rewriting its constitution, consulting external and internal experts, and retraining the system on updated data. Total fallbacks across all categories are expected to drop by around 67% on Claude.ai, 55% on Cowork, 17% on Claude Code, and 7% on the Claude Platform. Expanded Everyday Access and Classifier Retraining The updated safety classifiers reduce the frequency with which Fable 5 switches to Opus 5 for benign queries. Users and healthcare professionals will see fewer blocks on clinical tasks, symptom understanding, lab results interpretation, and educational biology inquiries. To achieve this, Anthropic rewrote the classifier's guiding constitution with expert feedback, generated new training data, and retrained the model to distinguish safe queries from harmful ones. Mitigating Catastrophic Dual-Use Risks Fable 5 was originally released with broad biology blocks to mitigate catastrophic risks, as the model can outperform human experts on certain biological tasks and potentially provide significant uplift to malicious actors developing biological weapons. Distinguishing beneficial research from harmful applications remains difficult because legitimate medical advances, like vaccines and toxin-derived drugs, often involve dangerous agents. Citing threat assessments regarding state-sponsored weapons programs and advances in biotechnology, Anthropic noted that broad restrictions were chosen to deploy the model quickly without compromising safety. Persistent Restrictions and Trusted Access Pathways Despite the reduction in false positives, Fable 5 still reroutes professional biology research and drug development queries involving dual-use fields like virology and molecular design to Opus 5. Anthropic plans to address this limitation by establishing trusted access pathways that provide vetted researchers with access to frontier biological capabilities.

Anthropic更新Claude Fable 5生物安全防线，大幅降低误报并减少降级处理 Anthropic宣布对Claude Fable 5的生物安全分类器进行优化，在维持对潜在生物武器及两用威胁防护的前提下，大幅减少了对良性生物与健康类查询的误报和降级处理。 测试显示，更新使跨产品界面的生物相关模型降级（fallback）减少了约85%。 日常健康咨询、症状理解、实验室指标解读以及部分临床任务将更顺畅地使用Fable 5。 病毒学、毒理学和分子设计等涉及两用风险的专业生物研究与药物开发查询，目前仍会降级至Opus 5。 团队重写了分类器的规范准则，并重新训练分类模型，未来计划通过受信任访问途径向专业生物学家开放前沿能力。 显著降低良性查询的降级频率 此前Fable 5在触发安全限制时会自动降级至能力较弱的Opus 5模型。此次更新重构了安全分类器规则并更新了训练数据，在测试中将生物相关降级减少了约85%。在全平台各类原因的总降级比例上，预计Claude.ai将减少约67%，Cowork减少55%，Claude Code减少17%，Claude Platform减少7%，改善了用户在健康教育及临床支持方面的体验。 应对生物领域的“两用”风险挑战 Fable 5在某些复杂生物任务上已展现出超越专家的能力，但此类能力存在双重用途风险，恶意行为者可能借此开发生物武器。由于治疗研究与制造致病源在技术层面界限模糊，且存在恶意混淆意图的风险，Anthropic在发布初期选择全面拦截生物查询以确保安全，现阶段仍继续限制高风险的专业生物和药物开发请求。 分类器优化与未来开放路径 安全分类器是自动检测受保护任务的小型AI系统。Anthropic通过重写分类器章程细化良性用途界限，并结合内外专家反馈重新训练与验证，以在防止越狱的同时降低误报。官方表示未来将继续改进防线，并致力于构建受信任的访问途径，以便日后安全地向专业研究人员开放前沿生物能力。

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. In our testing, this update reduced *biology-related* fallbacks by about 85% across our product surfaces.¹ Fable 5 will thus be able to assist with a wider range of biology tasks. In practice, users should see far fewer fallbacks on everyday health and educational questions—for example, interpreting lab results, understanding symptoms, and learning about biology in an educational context. Healthcare professionals will be able to receive more support from Fable 5 on clinical tasks. We believe the greatest opportunity for AI to positively affect the world is in biology and medicine, and we're investing significantly in building a responsible way to give biologists frontier access. Today, Fable still falls back to Opus 5 for requests we consider dual-use—including virology, toxicology, and molecular design—so it isn't yet usable for professional biology research and drug development. We're committed to closing that gap through trusted access pathways for frontier biology capabilities. Our objective is to get Fable 5’s frontier capabilities into the hands of as many of our users as possible, as quickly as possible. However, to do so, we need to manage the increasing risks that come with models this capable. One such risk is in the field of biology: Fable 5 can now outperform experts on some highly complex biological tasks and provide operational support on others. That means that it can provide genuine assistance to a researcher developing a new medical treatment (which is the reason we’re so keen to widen access to the model via both classifier improvements and trusted access programs). But in the wrong hands, those same capabilities could be used by a malicious actor, for example in developing a biological weapon. Our [capability assessments](https://www-cdn.anthropic.com/d00db56fa754a1b115b6dd7cb2e3c342ee809620.pdf) show that Fable 5 could provide significant *uplift* to such an actor—that is, it could provide them with capabilities they could not find anywhere else. It’s often difficult to tell apart beneficial and harmful uses of AI in biology. For example, in some cases researching a treatment for a disease requires scientists to produce the dangerous compounds that *cause* that disease in the first place. This is most obvious for live vaccines, which require scientists to grow the same pathogen they’re aiming to prevent. It’s also the case for some medicines. To develop the drug captopril, which treats hypertension, scientists isolated toxic components of snake venom that crash blood pressure in humans. As new biological capabilities develop on the frontier of AI, we need to be cautious to ensure that the new risks they pose do not materialize ahead of their potential scientific benefits. Sophisticated actors who wish to use our models to do harm know how to exploit this ambiguity to obscure their intent, making dangerous tasks look like ordinary research pursuits. The US Intelligence Community’s [2026 Annual Threat Assessment](https://www.dni.gov/files/ODNI/documents/assessments/ATA-2026-Unclassified-Report.pdf) makes clear that such actors exist, and that advances in biotechnology including synthetic biology and genomic editing *“could lead to novel biological threats.”* It notes that several state actors likely maintain active offensive biological and chemical weapons programs—programs that could be accelerated by access to the raw capabilities of frontier AI models. Because of our concerns about these “dual-use” capabilities (those that could be used for beneficial or harmful purposes, and where the line between them is not always easy to draw), we intentionally launched Fable 5 with almost all biology queries blocked. This enabled us to make the model available for users in other domains. We knew this would be frustrating for legitimate biology users: it would result in a high number of false positives in the near term, where users asking biology-related questions would have their requests blocked and sent to a less capable model. Nevertheless, we chose to make this tradeoff because the cost of Fable being misused in a dual-use domain like biology could potentially be [catastrophic](https://cdn.sanity.io/files/4zrzovbb/website/0bacdc8440ea96e62a8766d99ebe1d4eea6d5f3a.pdf). One of the core ways we protect against misuse in biology is via safety *classifiers:* smaller, automated AI systems that detect when Fable 5 is asked to perform a safeguarded biology task, or produce a harmful output (we've [previously written](https://www.anthropic.com/news/fable-safeguards-jailbreak-framework) about our similar classifiers in the domain of cybersecurity). In the case of Fable 5, when a classifier fires, the model re-routes the user’s request to Opus 5, a capable model that does not have the same level of biological capability as Fable 5 and which cannot provide as much assistance to a malicious user. This is the fallback that users see when their requests are blocked. Developing precise, robust classifiers is not a straightforward task. For a classifier to work rapidly and consistently, it has to learn the difference between what we consider “in scope” and “out of scope” for the topics and queries we consider to be potentially harmful. It takes time and iteration to tune the classifiers, avoiding both false positives (where classifiers fire on out-of-scope content) and false negatives (where in-scope content is missed). We also require our classifiers to be robust to attempts to bypass them (known as jailbreaks), which requires even further research and testing. Starting with a very broad biology classifier meant that we could give our users access to Fable 5 while we continued our research aimed at refining it. The alternative—holding back the model until much more safeguards progress was made—would have delayed the model’s general access, and its potential benefits to our users, by weeks or months. Over the past several weeks, we've carefully rewritten the classifier’s constitution (which consists of a collection of rules to help the model discern between safeguarded and allowed content), taking care to carve out benign uses in detail. We solicited feedback on the changes from a diverse range of experts (both internal and external to Anthropic). We then developed updated training data for the classifier based on that constitution, and retrained it, and verified the new classifier would still generally trigger for harmful and dual-use research biology content but would now enable a wider range of benign and beneficial uses. As is illustrated in the diagram below, these updates meant that—compared to at the time of Fable 5’s launch—the classifier will trigger for many fewer benign biology-related requests. There’s still much more to be done to refine our safeguards. There will inevitably remain false positives—requests that fall within the classifier’s safety margin where the request is very low-risk but where the classifier still fires. As we noted above, Fable will continue to block dual-use professional biology and drug development queries because of potential dual-use risk. We are fully committed to developing a safe, scalable path for researchers to use our most capable models via trusted access pathways. We hope you’ll continue to share your feedback with us so we can improve our safeguards even further. ¹ As a result, we expect the total number of fallbacks—for biology–related or any other reasons—will also be reduced: by roughly 67% on [Claude.ai](http://claude.ai/), 55% on Cowork, 17% on Claude Code, and 7% on the Claude Platform.