DOOMBENCH.COM
Safety and alignment

OpenAI deploys InstructGPT as default API language models

OpenAI made InstructGPT models the default language models on its API after human-feedback training improved instruction following and reduced toxic, untruthful, and hallucinatory outputs.

0 comments · 0 votesOpen discussion

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

No comments yet. Start the discussion.

Send detailed or private feedback →
CURRENT ASSESSMENT · REVISION 1
AWAY FROM DOOM37confidence 95/100

Why it moved the index

Deploying human-feedback-trained models as the API default translated alignment research into broad practical safeguards and reduced several measured harmful behaviors. Residual failures and the lack of a disclosed exact serving checkpoint constrain the magnitude and preclude a model record.

Assessment history

  1. R1
    Away 37 · confidence 95

    New January 2022 deployed alignment intervention absent from the durable corpus; exact serving checkpoint remains intentionally unprofiled.

    12 Aug 2026

Share this page

DoomBench social sharing card for OpenAI deploys InstructGPT as default API language models.

  1. DoomBench assesses “OpenAI deploys InstructGPT as default API language models” as evidence moving away from doom, with magnitude 37 and confidence 95 out of 100 in the safety and alignment category.

  2. The DoomBench assessment of “OpenAI deploys InstructGPT as default API language models” is based on reporting from OpenAI and records the editorial rationale, source quality, attribution, and revision history.

  3. DoomBench summarizes “OpenAI deploys InstructGPT as default API language models” as follows: OpenAI made InstructGPT models the default language models on its API after human-feedback training improved instruction following and reduced...

    https://www.doombench.com/news/openai-deploys-instructgpt-as-default-api-language-models-2022-01-27