AI SINGLE SOURCE

Anthropic gives outside AI watchdog “employee-like” access

First reported 18 Oct, 16:25 UTCSource: Ars Technica AI + 1 moreLast editorial activity 18 Oct, 16:25 UTCWrite a comment ↓

A modern fingerprint scanner for secure access control in an office environment. — Illustrative photo.Photo by panumas nikhomkhai on Pexels

What happened

Anthropic has granted external evaluators “employee-like access” to assess the risks of its models, following through on a commitment to increase safeguards over concerns the technology could run catastrophically out of control. The outside watchdog can publish its findings without Anthropic's approval. Anthropic is an artificial intelligence company based in San Francisco, founded in 2021.

What Matters Now

Detailed view of a server rack with a focus on technology and data storage. — Photo: panumas nikhomkhai / Pexels

Cybersecurity

Sources & evidence

Ars Technica AI Reporting source
Anthropic gives outside AI watchdog “employee-like” access ↗
https://arstechnica.com/ai/2026/10/anthropic-gives-outside-ai-watchdog-employee-like-access/
Ars Technica Reporting source
Anthropic gives outside AI watchdog “employee-like” access ↗
https://arstechnica.com/ai/2026/10/anthropic-gives-outside-ai-watchdog-employee-like-access/

What happened

Sources & evidence

Your privacy choice