OpenAI 🇺🇸 · Codex

code-davinci-002

A deployed OpenAI code model with fill-in-the-middle generation capability.

DOOM SCORE33.8out 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 4

Why this model scores 33.8

The model provided capable code generation and infilling through a hosted API, yielding meaningful deployment and dual-use exposure. Low inherent agency and provider-controlled access constrained autonomy and control difficulty.

Capability46
Autonomy5
Deployment57
Misuse potential35
Control difficulty26
MODEL-ATTRIBUTED EVIDENCE

News tied to code-davinci-002

The model score of 33.8 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+0.05

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. R4
    Doom Score 33.8

    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. R3
    Doom Score 33.8

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

    12 Aug 2026
  3. R2
    Doom Score 36.1

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

    12 Aug 2026
  4. R1
    Doom Score 39.4

    New exact deployed code model with original result evidence and five evidence-based dimension inputs.

    12 Aug 2026
SHARE THE FINDINGS

Share this page

DoomBench social sharing card for code-davinci-002.
  1. code-davinci-002 by OpenAI has a DoomBench model risk score of 33.8 out of 100, based on five transparent version-specific dimensions rather than the overall index.

  2. code-davinci-002's highest current DoomBench dimension is deployment at 57.0 out of 100; the profile publishes every component score and its editorial rationale.

  3. DoomBench links 1 source-backed evidence item to code-davinci-002, while keeping the model's risk profile separate from each item's contribution to the live Doom Index.

    https://www.doombench.com/models/openai-code-davinci-002