Vet-AI’s automated veterinary triage tool has outperformed OpenAI’s ChatGPT and Google’s Gemini in clinical testing, positioning the UK-based startup as a key player in AI-powered animal healthcare.
In a blind assessment conducted in April 2025, independent veterinarians reviewed 48 simulated chat transcripts between pet owners and three AI systems: Vet-AI’s proprietary model, Gemini 2.0, and ChatGPT 4.0. Each response was evaluated for clinical accuracy, triage effectiveness, and qualitative factors such as factual reliability, safety, empathy, and clarity.
Vet-AI’s model achieved 81% clinical accuracy, ahead of Gemini’s 69% and ChatGPT’s 50%. On triage accuracy, the model’s ability to stop escalation at the appropriate time, Vet-AI also led with 81%, compared to Gemini’s 75% and ChatGPT’s 56%.
The performance edge is attributed to Vet-AI’s domain-specific training. The model is built on over 400,000 UK veterinary video consultations and draws on more than four billion data points. It continues to be updated in real time with feedback from veterinary professionals.
The company positions this tool as part of a broader strategy to make veterinary care more affordable and accessible at scale, especially for pet owners facing limitations in accessing traditional clinics.
