# German AI Start-Ups and AI Ethics: Using A Social Practice Lens for Assessing and Implementing Socio-Technical Innovation

**Authors:** Mona Sloane, Janina Zakrzewski  
**arXiv:** [2206.09978](https://arxiv.org/abs/2206.09978)  
**Date:** 2022-06-20  
**Summary page:** https://www.summarizepaper.com/en/arxiv-id/2206.09978v1/

## Key points

- The field of AI ethics lacks empirical research on how ethical concerns are understood and implemented in start-ups
- This gap can lead to a disconnect between scholarly research, innovation, and application
- There is an urgent need for socio-technical innovation that prioritizes fairness, accountability, and transparency in AI systems
- The authors propose a framework based on social practice theory to systematically analyze cultural understandings, histories, and social practices related to ethical AI
- Empirical findings from their study on the operationalization of ethics in German AI start-ups emphasize the importance of understanding AI ethics and social practices within specific cultural and historical contexts
- Their contributions are threefold: introducing a practice-based approach for understanding ethical AI; presenting empirical findings that highlight the importance of contextualizing ethical considerations within specific cultural contexts; suggesting breaking down ethical AI practices into principles, needs, narratives, materializations and cultural genealogies as a useful backdrop for considering socio-technical innovations.
- This paper provides valuable insights into how socio-technical innovations can be implemented effectively by analyzing existing cultural understandings and social practices related to ethical AI.

## Summary

The field of AI ethics is currently lacking empirical research on how AI practitioners understand and implement ethical concerns, particularly in the context of start-ups. This gap poses a significant risk of disconnect between scholarly research, innovation, and application. With mounting pressures to identify and mitigate potential harms of AI systems, there is an urgent need for socio-technical innovation that prioritizes fairness, accountability, and transparency. To address this need, the authors propose a framework based on social practice theory that allows researchers, practitioners, and regulators to systematically analyze cultural understandings, histories, and social practices related to ethical AI. The authors present empirical findings from their study on the operationalization of ethics in German AI start-ups to emphasize the importance of understanding AI ethics and social practices within specific cultural and historical contexts. Their contributions are threefold: firstly they introduce a practice-based approach for understanding ethical AI; secondly they present empirical findings that highlight the importance of contextualizing ethical considerations within specific cultural contexts; finally based on their empirical findings they suggest breaking down ethical AI practices into principles, needs, narratives, materializations and cultural genealogies as a useful backdrop for considering socio-technical innovations. Overall this paper provides valuable insights into how socio-technical innovations can be implemented effectively by analyzing existing cultural understandings and social practices related to ethical AI. By doing so it will help bridge the gap between scholarly research and practical application while ensuring that fairness accountability transparency are prioritized in future developments in this field.