06/03/2026
AI vs Vets: Why we’re not out of a job yet
We’re seeing more people upload pet X-rays and lab reports into AI tools asking for a diagnosis. Curiosity is understandable — but this is a good example of why that doesn’t work.
I showed this X-ray of my own dog to an AI tool and asked what was wrong with the spine.
Its confident answer:
• Diagnosed multiple “butterfly vertebrae” (L1–L3)
• Claimed these were congenital spinal malformations
• Suggested they were probably incidental
What actually happened:
• None of those findings were present
• The AI missed the real clinical pathology causing my dog’s problem
• It produced a confident explanation for abnormalities that simply weren’t there
This highlights an important issue: AI can sound very convincing — even when it’s wrong.
Veterinary diagnosis is never based on one image alone. It involves:
• The pet’s history
• A physical examination
• Multiple imaging views and correct positioning
• Laboratory results
• Clinical experience interpreting all of that together
Even trained veterinarians can’t safely diagnose from a single image without context — which is why we interpret diagnostics as part of the whole patient.
AI can be useful for explaining concepts or helping people learn. But it can’t examine your pet, assess pain, interpret the full clinical picture, or take responsibility for medical decisions.
So don’t worry — your vet isn’t being replaced anytime soon.
Dr Noriel - Sunshine Vet Clinic 🐾
More Cuddles, Less Struggles
