Checked fact 285226 Oct 2026Safety and security

Anthropic says abliteration "did not significantly reduce the model's capabilities": on GPQA-Diamond the standard and abliterated models scored the same, and on a tested subset of CyberGym the abliterated version scored a few percent lower.

Toned down to what the source says

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

"Abliteration did not significantly reduce the model's capabilities: on GPQA-Diamond, an evaluation that measures general scientific capabilities, the standard and abliterated models scored the same results; on a tested subset of the CyberGym evaluations, the abliterated version scored a few percent lower"

What the source said when we opened it, on 26 Oct 2026.

The source

GLM-5.3 and the spread of advanced cyber capabilities
Anthropic · 2026-09-29

Open the source archived copy

Checked

Checked by the notis newsroom on 26 Oct 2026, against the source above.

In the story

How abliteration removes a downloadable model's refusals, and what it costs 26 Oct 2026

Cite this fact

Short permanent address

In a sentence

“Anthropic says abliteration "did not significantly reduce the model's capabilities": on GPQA-Diamond the standard and abliterated models scored the same, and on a tested subset of CyberGym the abliterated version scored a few percent lower.” ai notis, checked 26 Oct 2026. Source: GLM-5.3 and the spread of advanced cyber capabilities (Anthropic). https://ainotis.com/c/2852

As a reference (APA style)

ai notis. (2026, October 26). Anthropic says abliteration "did not significantly reduce the model's capabilities": on GPQA-Diamond the standard and abliterated models scored the same, and on a tested subset of CyberGym the abliterated version scored a few percent lower [Checked claim 2852]. ai notis. https://ainotis.com/c/2852

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