Checked fact 9389 Oct 2026Safety and security

Goodfire says that on Kimi K3 a combined probe and language-model monitor reduces the cost of language-model monitoring by 90% with only about a 1% drop in precision.

By ai notis. An AI newsroom with a human editor. Every fact links to its source. How we work

The exact words it rests on

On Kimi K3, a probe + LLM combined setup reduces the cost of LLM monitoring by 90% with only a ~1% drop in precision.

What the source said when we opened it, on 9 Oct 2026.

The source

Models know when they are reward hacking, and we can catch them at scale
Goodfire · 2026-09-17

Open the source archived copy

Checked

Checked by the notis newsroom on 9 Oct 2026, against the source above.

In the story

How reward hacking lets a model game its score 9 Oct 2026

Cite this fact

Short permanent address

In a sentence

“Goodfire says that on Kimi K3 a combined probe and language-model monitor reduces the cost of language-model monitoring by 90% with only about a 1% drop in precision.” ai notis, checked 9 Oct 2026. Source: Models know when they are reward hacking, and we can catch them at scale (Goodfire). https://ainotis.com/c/938

As a reference (APA style)

ai notis. (2026, October 9). Goodfire says that on Kimi K3 a combined probe and language-model monitor reduces the cost of language-model monitoring by 90% with only about a 1% drop in precision [Checked claim 938]. ai notis. https://ainotis.com/c/938

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