Claude AI服务再中断，Anthropic确认2026年4月7日错误率升高
2026年4月7日，人工智能助手Claude再次出现服务异常，其开发商Anthropic在官方状态页面确认系统遭遇“错误率升高”问题。此次事件发生在4月6日类似问题修复后仅一天，显示出服务不稳定性，引发用户和业界关注。根据报告，问题从东部时间上午开始显现，尽管呈现间歇性，但已足以影响正常使用流程。 具体而言，此次中断主要影响用户登录、语音模式功能以及聊天交互。监测平台Down Detector的数
Anthropic意外泄露Claude Code源代码，超50万行代码曝光
人工智能公司Anthropic近日发生源代码泄露事件，其热门编程助手工具Claude Code的完整源代码被意外公开。泄露发生在周二上午，当时Anthropic向公共npm注册表发布Claude Code版本2.1.88时，不慎包含了一个源映射文件。该文件直接引用了未混淆的TypeScript源代码，导致超过50万行代码和近2000个文件暴露，涵盖了核心架构与未发布的功能标志。 Anthropic
Claude Service Hit by Second Outage in Two Days: Anthropic Confirms High Error Rates
Claude AI assistant experienced another service disruption on April 7, 2026, with Anthropic confirming high error rates. Reports from Down Detector showed a spike in issues from 10 a.m., causing intermittent outages and significantly impacting stability.
Anthropic's Accidental Claude Source Code Leak Challenges AI Safety Narrative
Anthropic accidentally leaked internal source code for Claude due to a packaging error, raising security concerns despite no sensitive data exposure. This incident highlights operational vulnerabilities in the AI safety-focused company.
Anthropic's Claude AI Source Code Leak Raises Security Brand Concerns
Anthropic PBC, an AI company known for its security focus, accidentally released internal source code for its Claude AI coding assistant. The firm stated it was a human error in packaging, not a security breach, and no sensitive data was exposed. However, this incident sparks industry doubts about its internal controls and operational safety.