In a Shenzhen stakeholder study of 45 participants, concerns surfaced about medical AI governance, centering on safety, regulatory clarity, responsibility allocation, and cross-sector coordination. Developers, regulators, and clinicians described gaps that could complicate deployment and accountability when medical AI tools affect patient care. The findings point to a governance challenge that extends beyond model performance: who is responsible when systems underperform, how responsibilities are assigned across organizations, and how regulators and practitioners coordinate. For biotech and digital health companies, the practical implication is that commercialization will likely depend not just on technical validation, but on clearer compliance pathways and operational responsibility frameworks.