Patients with EGFR-mutated (EGFRm) NSCLC are particularly sensitive to treatment with an EGFR-tyrosine kinase inhibitor (EGFR-TKI). In current clinical practice molecular diagnostic testing is used to identify these patients, however this technology comes with challenges around turnaround time of testing. Lunit is developing an AI tool that can rapidly identify those at high risk of EGFR-TKI sensitising mutations. This tool may offer a rapid and low-cost solution to address testing challenges. Here we show Lunit’s tool had robust performance and generalisability on an independent test dataset provided by AstraZeneca. Further model improvements are ongoing and will be presented at future industry meetings.