New AI model performs near expert levels in medical question answering

New AI model performs near expert levels in medical question answering
Image: Intellect (generated)

In short: Researchers developed Med-PaLM 2, a large language model designed to improve medical reasoning and accuracy. In a pilot study using real-world questions, medical specialists preferred the AI's answers over those from generalist physicians 65% of the time, and both groups rated the model's safety as comparable to human doctors.

The researchers improved the model through medical domain fine-tuning and new strategies like chain of retrieval to enhance reasoning. While the model showed significant performance gains over its predecessor, particularly on adversarial datasets, the study was a pilot using real-world questions to assess practical utility.

Why it matters: AI is showing increasing potential for real-world medical applications by matching physician safety levels.

Opens the Intellect+ feed with this card first. Swipe up for the next one.

Sources

To cite this card, link to https://intellectplus.co.uk/app/technology-research-new-ai-model-performs-near-expert-levels-in-medical-question.

How we source, check and correct cards: Editorial standards.

General information, not professional advice. Spotted an error? Flag the card or see corrections and complaints.

Intellect Plus is a visual educational discovery platform for exploring fascinating facts and ideas. Brain Food, Not Brain Rot. What is Intellect Plus?

Related

Finding something worth knowing…