A method for assessing Safety Of The Intended Functionality of autonomous driving based on DAG block chain is provided. The method is based on DAG block chain technology and fused with system of the intended functionality, realizing safety assessment of the intended functionality. The intended functionality is objectively assessed based on data. The assessment method, in combination with an in-the-loop test, analyzes the current test data, without relying on a large amount of traditional experience to make judgments. Result of credit assessment of the nodes is taken as a result of safety assessment. Each of hardware or components is analyzed in terms of the safety and reliability, and the result of assessment is relatively complete. A device and a computer-readable storage medium are further provided.