Foundation Models Do Not Understand Biology

Imagine an AI medical tool that can write a beautiful diagnostic report in seconds, which is clean, confident, perfectly structured, and then say something completely impossible, like a blood parasite living inside a tissue sample where it simply could not exist. That’s the strange thing about modern multimodal LLMs. On one hand, they can do genuinely impressive work: pull together clinical details, write clear summaries, spot patterns, and help overstretched healthcare workers move faster. In places where.