[Context and Motivation] The automotive industry is moving towards increased automation, where features such as Automated Driving Systems (ADS) typically include machine learning (ML), e.g. in the perception system.

[Question/Problem] Ensuring safety for systems partly relying on ML is a challenge. Different approaches and frameworks have been proposed, typically where the developer must define quantitative and/or qualitative acceptance criteria, and ensure the criteria are fulfilled using different methods, often including the need for extensive testing. Further, during operation, methods need to be defined that enable the vehicle to judge that it is not leaving its operational design domain (ODD) as well as recognizing edge cases or distribution shifts as potential hazards. However, there is still a gap between many quality methods and metrics employed in the ML domain and their potential use for safety assurance.

[Principal Ideas/Results] In this paper, we argue the need for connecting available ML quality methods and metrics to the safety lifecycle and explicitly show their contribution to safety. In particular, we analyse Out-Of-Distribution (OOD) detection, e.g., the frequency of novelty detection, and show its potential for multiple safety-related purposes. I.e. as (a) an acceptance criterion to decide if the software is Ready-for-Release, e.g., if there only are X novelties per Y miles, (b) in ODD selection and expansion by including novelty samples into the training/development loop, and (c) as a run-time measure, e.g., if there is a sequence of novel samples, the vehicle should consider reaching a minimal risk condition.

[Contribution] This paper describes the possibility to use OOD detection as a safety measure, and is analysed in different stages of the safety lifecycle.