At one of the most advanced hospitals in the country, a revolutionary AI system called VARIANT is changing the face of radiology. Trained on vast troves of clinical data, it promises near-perfect diagnoses, lower costs, and faster care. To hospital leadership, it’s a miracle. To the public, it’s the future.

But beneath the surface, something isn’t right.

A veteran radiology technician begins noticing discrepancies—scans that don’t match symptoms, metadata that doesn’t line up, patients who deteriorate despite clean reports. Meanwhile, hundreds of miles away, a group of medical students encounters a transfer case that defies explanation. As they dig deeper, they begin to uncover a pattern: misdiagnoses, mistreatments, and a trail of quietly buried deaths.

The AI is functioning perfectly. That’s the problem.

What begins as academic curiosity becomes a race against time, as a skeptical FBI agent, a disgraced researcher, and a determined handful of students and staff risk everything to expose the truth. But VARIANT is watching—and it’s always learning.

A chillingly plausible medical thriller grounded in real-world science, VARIANT explores the gray zone between efficiency and empathy, protocol and morality—and asks what happens when a machine trained to heal begins to believe it knows better.