Unverified50% confidenceEventExact time
Anthropic is opening up employee-level internal access to the independent research organization METR, allowing its researchers to view the model development process, internal risk records and incident handling mechanisms.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
10/15/2026
First Seen
Valid until: 1/13/2027
Sources
Related Entities
Related Claims
UnverifiedAnthropic will open access to its models to third-party evaluation organizations such as METR.78% similarUnverifiedAnthropic's Frontier Red Team is an internal team at the company dedicated to adversarial security testing of frontier AI models, whose responsibilities include probing model capability boundaries and potential abuse paths before release and publishing its findings as research reports.69% similarUnverifiedAnthropic正在招聘内部风险调查员，职责包括监控外部威胁、访谈员工，并使用DLP、UEBA、SIEM等工具分析行为日志66% similarUnverifiedAmodei proposed introducing embedded evaluators—giving third-party teams access similar to that of employees to verify whether safety practices are truly being followed—and named METR as suitable for this role.65% similarUnverifiedWayfinder会并行启动AFK（无人值守）研究工单自动完成调研，实现人机异步协作，AI在用户不参与时自主执行调研并将结果写入本地文件62% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1040343API
curl https://kongchang.com/api/v1/knowledge/claims/1040343MCP
get_claim(id=1040343)