Loading page...Загружаем страницу...正在加载页面...
AtlasRepo

We ran HealthBench on our health AI's safety layer. It scored lower than the bare model

Описание и применение

Тестирование безопасности в системе Tabibu показало, что добавление защитного слоя снижает показатели на HealthBench, но корректировка политик повышает их и выявляет ошибки в выдаче дозировок.

Задача

Assessing the impact of a safety layer on a health AI’s performance on a public benchmark.

Результат

The safety layer reduced benchmark scores due to its conservative policies, but policy tweaks improved performance while uncovering a safety bug involving prescription dose quoting.

Рабочий процесс

  • Run HealthBench on a 150‑conversation subset with a gpt‑4.1‑mini grader
  • Compare scores of (1) full Tabibu pipeline, (2) bare gpt‑6‑luna, (3) bare gpt‑4o‑mini
  • Identify per‑criterion failures
  • Modify safety policies to allow general answers and tiered doses
  • Re‑run benchmark
  • Implement safety fix for prescription dose quoting
  • Re‑run benchmark

Доказательства

The full pipeline scored 0.362 on 150 HealthBench conversations

Источник

The pipeline was 0.158 below the bare model on the same conversations (paired difference, ±0.039).

Источник

I changed one policy: let the assistant answer general, low‑risk questions from the model's own knowledge, with a clear label saying the answer is not from its curated sources, instead of refusing.

Источник

With the original rules, the pipeline could quote prescription doses straight out of a retrieved national formulary

Источник

  • Дата записи: 2026-10-06T06:41:11.777Z