OpenAI 🇺🇸 · gpt-oss-safeguard

gpt-oss-safeguard-120b

A 120B open-weight policy-reasoning model for configurable content classification and safety enforcement.

DOOM SCORE44.8out of 100model risk profile, not the overall index
0 comments · 0 votesOpen discussion

Public discussion is readable by everyone. Sign in to comment, reply, or vote.

No comments yet. Start the discussion.

CURRENT ASSESSMENT · REVISION 3

Why this model scores 44.8

Focused rather than frontier capability and little independent autonomy keep risk moderate; Apache licensing maximizes deployment while safeguard purpose limits misuse.

Capability45
Autonomy20
Deployment99
Misuse potential32
Control difficulty38
MODEL-ATTRIBUTED EVIDENCE

News tied to gpt-oss-safeguard-120b

The model score of 44.8 rates this model's risk profile. The overall Doom Index of 67.9 measures the complete temporally weighted evidence record. These values answer different questions.

NET MODEL-ATTRIBUTED INDEX CONTRIBUTION-0.09

Each article's current Doom Index contribution is divided equally among the exact models named on that article. This prevents multi-model evidence from being claimed in full on several model pages. Model risk scores use a bounded temporal offset around their technical profile, but never feed back into the overall index.

AUDIT TRAIL

Model score history

  1. R3
    Doom Score 44.8

    Exact-version evidence chronology replayed after run doombench-hourly-news-20260812-142416 under temporal-monthly-pressure-v4.

    12 Aug 2026
  2. R2
    Doom Score 44.5

    Exact-version evidence chronology replayed after run doombench-hourly-news-20260812-132112 under fixed-sensitivity-v3.

    12 Aug 2026
  3. R1
    Doom Score 43.4

    New exact safeguard model absent from the durable catalogue.

    11 Aug 2026
SHARE THE FINDINGS

Share this page

DoomBench social sharing card for gpt-oss-safeguard-120b.
  1. gpt-oss-safeguard-120b by OpenAI has a DoomBench model risk score of 44.8 out of 100, based on five transparent version-specific dimensions rather than the overall index.

  2. gpt-oss-safeguard-120b's highest current DoomBench dimension is deployment at 99.0 out of 100; the profile publishes every component score and its editorial rationale.

  3. DoomBench links 1 source-backed evidence item to gpt-oss-safeguard-120b, while keeping the model's risk profile separate from each item's contribution to the live Doom Index.

    https://www.doombench.com/models/openai-gpt-oss-safeguard-120b