Does Burna AI train AI models on customer data?

No. Customer data is processed to produce grades, attributions and reports for that customer, and Burna AI does not use it to train its engine or any third-party model without the customer's explicit written consent. The engine's behavior comes from its architecture rather than from customer data: twelve specialized agents in a cascading constraint pipeline, the CTCAE criteria, MedDRA terminology and the WHO-UMC and Kramer algorithms. Documents a customer uploads, such as protocols, sit in that organization's workspace and ground answers for that organization only.