How a $10M Digital Health Platform Evaluated GPT-5.3 Codex After a Regulatory Near-Miss

https://reportz.io/ai/when-40-ai-models-faced-1200-hard-questions-what-the-numbers-actually-show/

In January 2026, a Series-B digital health company with $10 million in annual recurring revenue faced a near-miss: an external report flagged that the virtual triage assistant had generated several confidently stated but incorrect recommendations