Diagnostic-support AI has undergone a technological leap from being rule-driven to data-driven. Its inherent autonomous decision-making capacity, an algorithmic black-box nature, and dynamic evolution pose structural challenges to the traditional medical malpractice liability system, which is centered on the prevailing medical standard. To address these challenges, the use of diagnostic-support AI should be characterized as the adoption of a new medical treatment to establish its legal basis. It is necessary to clarify that its outputs are not directly equivalent to the prevailing medical standard, and that the implications of this standard will also evolve. On this basis, this paper constructs a dynamic fault-determination framework centered on the duty of prudent reassessment. This framework shifts the focus of assessment to evaluate whether healthcare professionals have fulfilled their duties of prudent use and substantive verification of algorithmic outputs, thereby responding to the needs of liability determination in the current human-computer collaboration model and achieving a normative balance between technological innovation and patient safety protection.