Anthropic gives outside AI watchdog “employee-like” access

Anthropic has granted external reviewers access to its systems at a level comparable to that of its own employees, aiming to scrutinize the risks associated with its AI models. The nonprofit METR, which works with labs such as OpenAI, Google, and Meta, now has a dedicated team stationed on Anthropic’s premises, giving them direct entry to the company’s hardware and internal environments. This step is part of a broader set of concrete actions that leading AI labs are taking to address growing concerns about the potential adverse effects of increasingly powerful systems on the economy, cybersecurity, and overall human safety. By offering third parties an inside view, the goal is to strengthen oversight mechanisms and build more effective safeguards against catastrophic scenarios. The initiative underscores the importance of transparency and external collaboration in managing AI risk, a critical factor for ensuring that technological progress proceeds safely and responsibly.

Read the original article on Ars Technica AI

This summary is an informational synthesis produced by dataqbs.com. All rights to the original content belong to its author and the cited media outlet. We act solely as curators of technology news and claim no authorship.