ainotis Join
My notis

Checked fact 509716 Nov 2026Safety and security

Anthropic says LLM graders can fail without anyone noticing, and recommends building in alerting so that you notice when something goes wrong with a grader.

The exact words it rests on

these graders can fail without anyone noticing

What the source said when we opened it, on 15 Nov 2026.

The source

Building strong safety evaluations (Anthropic Engineering)
Anthropic · 2026-11-13

Checked

Checked by the notis newsroom on , against the source above.

In the story

Anthropic says single-turn safety evals are saturated at 94% to 98% on its API 16 Nov 2026

Cite this fact

Anyone may quote this address. It does not change; if we correct the story, this page says so.