English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

In the future, AI will increasingly find its way into systems that canpotentially cause physical harm to humans. For such safety-critical systems, itmust be demonstrated that their residual risk does not exceed what isacceptable. This includes, in particular, the AI components that are part ofsuch systems' safety-related functions. Assurance cases are an intensivelydiscussed option today for specifying a sound and comprehensive safety argumentto demonstrate a system's safety. In previous work, it has been suggested toargue safety for AI components by structuring assurance cases based on twocomplementary risk acceptance criteria. One of these criteria is used to derivequantitative targets regarding the AI. The argumentation structures commonlyproposed to show the achievement of such quantitative targets, however, focuson failure rates from statistical testing. Further important aspects are onlyconsidered in a qualitative manner -- if at all. In contrast, this paperproposes a more holistic argumentation structure for having achieved thetarget, namely a structure that integrates test results with runtime aspectsand the impact of scope compliance and test data quality in a quantitativemanner. We elaborate different argumentation options, present the underlyingmathematical considerations, and discuss resulting implications for theirpractical application. Using the proposed argumentation structure might notonly increase the integrity of assurance cases but may also allow claims onquantitative targets that would not be justifiable otherwise.

Integrating Testing and Operation-related Quantitative Evidences in Assurance Cases to Argue Safety of Data-Driven AI/ML Components