An older CNN names more skin lesions correctly in a dermatologist's test
Magnus Lynch reports 62% diagnostic accuracy for fine-tuned Inception v3 versus 42% for GPT-6 Astra and Claude Fable 5.1 across 200 biopsy-confirmed lesions. At its preset discharge threshold, the CNN discharged 18 of 30 melanomas, versus three and four for the LLMs. Lynch ran each LLM case once.