A 2024 checklist for medical imaging AI quietly asks the questions a headline AUC skips: who was tested, where, how trustworthy the numbers are, and who benefits.

A CT scan gives more radiation than a regular X-ray. How much more depends on the scan and the person, and the risk is small but uncertain.

Seventy-three teams tested one hypothesis on the same data and reached opposite answers. The lesson is about method, not misconduct, and about making hidden choices visible.

A 68-country survey published in January 2025 found trust in scientists moderately high on average. Where scientists scored lowest was listening, and that part is fixable.

Registered Reports let journals judge a study's question and methods before any data exist. For many studies that test a clear idea, I think that is a better deal.

A 2024 Nature Perspective argues that AI can make scientists feel they understand more than they do. The fix starts with knowing the assumptions behind a fluent answer.

When one patient's scans end up in both training and test data, a medical imaging model can look far better than it is, and nobody has to cheat.

In March 2025, bioRxiv and medRxiv moved to a new independent nonprofit. Science leans on these servers now, so how they are run and paid for matters.

A December 2024 consensus asks health AI teams to say who is in their data, who is missing, and how a model works for each group.

A December 2022 Liver-Chip study reported strong results on one defined job. The useful question is what a model is validated for, not whether it replaces animals.

In a 2023 study, radiologists followed a fake AI's wrong calls on mammograms. The lesson is less about AI than about when and how its answer is shown.

Nature's 2023 postdoc survey found more optimism than in 2020, alongside familiar pressures. I think the healthiest reading is that a science career has many homes.