Latest
  1. Johns Hopkins Bayview tells 3,182 patients an employee put their health data into a public AI tool
  2. China’s AI compute access expected to double in 2027, with US-designed chips about 65 per cent
  3. UK commits £100m to national AI Biosecurity Lab
  4. Moonshot AI releases Kimi K3.1 Pro with 1-million-token context window
  5. Residents in Newport and Cardiff oppose hyperscale datacentres near homes
  6. Study: Grok 4.7 picked Republicans in 92% of forced midterm answers, GPT-6.1 Sol picked Democrats in 97%
  7. Attackers exploiting the PHP flaws hand the break-in to AI agents, Unit 42 says
  8. Elvis Presley’s rightsholders sue Anthropic over training on 274 songs

Latest / Health and life sciences

ResearchHealthScienceUSConfirmed

Nature publishes Google’s Med-PaLM paper on how language models answer medical questions

See how this story connects

Source: Nature (via PubMed)

Nature published Google’s paper “Large language models encode clinical knowledge” on 12 July 2023. It introduced MultiMedQA, a benchmark built from six medical question datasets plus HealthSearchQA, a new set of questions from online searches. Flan-PaLM reached 67.6% on MedQA, and the instruction-tuned Med-PaLM still fell short of clinicians in human evaluation.

Why it matters

Researchers and regulators gained a peer-reviewed method for testing medical chatbots that combines exam scores with clinician review of answers.

When
12 Jul 2023
ID
WTA-N-2023-0075
Who
Google

Summary by wheresthe.ai, from Nature (via PubMed). Checked against the source on 9 Oct 2026.

Read the original at Nature (via PubMed)

Source copy: download the page as saved on 21 Oct 2026 (PDF). Also held by the Internet Archive.

Earlier in this Storyline: What led hereGoogle says Med-PaLM 2 scores 85.4% on US medical licensing-style questions

Cite this story

WTA-N-2023-0075: “Nature publishes Google’s Med-PaLM paper on how language models answer medical questions.” wheresthe.ai, 12 Jul 2023. https://wheresthe.ai/d/google-med-palm-nature-paper-2023-07/ (source: Nature (via PubMed)).

Spotted an error? Email [email protected].