InstaLabel

Croviz's growth turns on two evidenced shifts rather than on any product launch of its own: the pool of imaging models needing oversight keeps growing, and the buyers of those models face mounting pressure to prove what they bought. The cleared pool reached 1,524 FDA-cleared AI devices in 2026, 1,163 of them radiology, with 68 radiology clearances in the first quarter and monthly clearance near 30 against roughly 21 in 2024. Renewals show the budget is already committed, with 95% of AI-using sites renewing at least one contract in the year to April 2026.

The indicators that would confirm the trajectory are those same two series: clearance volume and renewal rates. A stall in either, or a move by buyers back toward vendors' own performance claims, would weaken it.

Radiology's algorithm pipeline is expanding faster than buyers' ability to judge it, and that gap widens the market Croviz sells into rather than narrowing it. The cleared pool reached 1,524 FDA-cleared AI devices in 2026, 1,163 of them radiology, with 68 radiology clearances in the first quarter of 2026 and monthly clearances near 30 against roughly 21 a month in 2024. Each newly cleared model is a candidate for independent oversight.

Demand-side evidence points the same way: a 2026 survey of 445 radiology AI users across 12 countries found that 95% of sites had renewed at least one AI contract within the previous year while only 30% had quantified a positive financial return. Buyers are consolidating toward fewer vendors and stronger governance, a shift Radiology Partners' chief medical AI officer described in 2026. Sites reviewing AI more often than quarterly reported more trust in it.

The risks that could invalidate Croviz's position concern evidence and trust rather than rivals. Its 2026 survey of 445 radiology AI users found that 77% had seen AI performance drift, that two in three treated bias as a malpractice risk, and that radiologists took one to three years to trust a tool — a slow conversion clock for a subscription that depends on clinical buy-in. Every sale therefore rests on buyers continuing to demand measured performance.

The compliance bar is a second exposure: Croviz states SOC 2 Type 2 and HIPAA compliance as the condition for working on provider data, so a failure there carries direct commercial cost. Its trust-building depends on the same survey evidence it publishes, which makes clinical skepticism as much a funding risk as an operating one.