Anthropic 向更多安全团队开放 Claude 并放宽安全限制
原标题：Anthropic gives more security teams access to Claude with fewer safety restrictions
AI 摘要
Anthropic 将其 Cyber Verification Program 开放给更广泛的安全专业人员，允许经过审核的组织和独立研究员在减少安全过滤的情况下使用 Claude 进行漏洞研究、恶意软件分析、事件响应和渗透测试。该计划分为 Defense Access、Red Team Access 和 Specialized Access 三个层级，其中最高层级需与美国政府共同审核。Anthropic 称前身项目 Project Glasswing 的合作伙伴在 2026 年 4 月至 7 月间确认了至少 12.9 万个漏洞，其中超过 3.3 万个为高危或严重级别。
正文节选
Anthropic gives more security teams access to Claude with fewer safety restrictions Anthropic is opening its Cyber Verification Program to a much wider pool of security professionals. The program gives vetted organizations and individual researchers access to Claude's most powerful AI models with reduced safety filters for vulnerability research, malware analysis, incident response, and penetration testing. The publicly available Claude models block most of that kind of work because the same cap