Technical articles · 2024.04.22
Reinhard Stolle on narrow domains, simulation and AI assurance
Interview mit Reinhard Stolle – »Wir nutzen KI, um Systeme sicherer zu machen«
Overview
In this 2024 interview with Hans-Thomas Hengl, Reinhard Stolle contrasts incremental ADAS development with autonomy in narrow domains. He discusses robustness and confidence assessment of ML models, simulation and reuse of evidence across development cycles, distinguishing safe design from convincing others of its safety.
Publication and version information
The source records posting on 2024-04-22 and modification on 2024-06-03; the update date is not an exact revision identifier. Forecasts and safety statements are interview views, not validated deployment results. The source records publication on 2024-04-22 and modification on 2024-06-03; its biography saying deputy director 'since June' is not backdated to the initial April posting. Hengl is the writer and Stolle the interviewee.