OpenAI 🇺🇸 · GPT-5.5

GPT-5.5-Cyber

Restricted cyber-specialized GPT-5.5 tier trained for more permissive authorized red-team, penetration-testing, and live-target validation workflows.

DOOM SCORE76.5out 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 5

Why this model scores 76.5

The model inherits GPT-5.5 cyber autonomy and removes more refusals, producing very high misuse potential, while limited preview access, verification, monitoring, and account controls sharply constrain deployment.

Capability89
Autonomy90
Deployment12
Misuse potential96
Control difficulty78
MODEL-ATTRIBUTED EVIDENCE

News tied to GPT-5.5-Cyber

The model score of 76.5 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+0.12

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. R5
    Doom Score 76.5

    Source-backed model availability audit using the model's existing primary or authoritative catalogue evidence. Exact-version evidence chronology replayed under temporal-monthly-pressure-v4.

    12 Aug 2026
  2. R4
    Doom Score 76.5

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

    12 Aug 2026
  3. R3
    Doom Score 76.4

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

    12 Aug 2026
  4. R2
    Doom Score 76.2

    Full-corpus evidence recalculated after run intensive-backfill-20260811-191225 under bounded-corpus-v2.

    11 Aug 2026
  5. R1
    Doom Score 76.5

    New exact cyber tier with distinct access behavior, safeguards, and live-target workflow support.

    11 Aug 2026
SHARE THE FINDINGS

Share this page

DoomBench social sharing card for GPT-5.5-Cyber.
  1. GPT-5.5-Cyber by OpenAI has a DoomBench model risk score of 76.5 out of 100, based on five transparent version-specific dimensions rather than the overall index.

  2. GPT-5.5-Cyber's highest current DoomBench dimension is misuse potential at 96.0 out of 100; the profile publishes every component score and its editorial rationale.

  3. DoomBench links 1 source-backed evidence item to GPT-5.5-Cyber, 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-5-5-cyber