Academic edition · 2026
Foundations and purpose of this edition
A critical approach to the relationship between technology, work and professional responsibility.
Foreword
Artificial intelligence has become part of people management through classification, prediction and content generation systems. Its presence does not automatically make a decision objective or improve job quality in itself. This book examines the conditions under which these tools can contribute to more informed, transparent and responsible management.
Human dignity is the central principle. A job application, an assessment or a career cannot be reduced to a score without context. The ability to automate a task must be examined alongside the legitimacy of its purpose, the quality of the evidence and the possibility of challenging an outcome.
Introduction and scope
This revised and condensed academic edition develops the ten chapters of the manuscript supplied by the author. It corrects the contents, removes duplication and replaces promotional language with critical discussion. Organizational examples are presented as teaching scenarios; no successes, interviews or collaborations are attributed without documentary support.
The book covers information systems, emerging technologies, platforms, performance, recruitment, planning, agile methods, security and the future of work. The questionnaires check understanding of concepts and support discussion of dilemmas. They do not assess respondents' suitability for employment, personality or moral worth.
How to read this book
Each chapter identifies concepts, design decisions, risks and questions for professional practice. Four levels should be distinguished: what is technically possible, empirically demonstrated, organizationally desirable and legally permissible. A system's ability to process a signal does not demonstrate that the signal measures the ability being assessed.
Links to legal and methodological sources appear at the end of the book. Notes from recognized authors provide starting points for reflection, rather than appeals to authority that replace evidence. The translations are automated working versions and require specialist linguistic review before being considered definitive.
Working concepts
An information system organizes data and procedures. A predictive model estimates outcomes using learned patterns. A generative tool produces plausible content that may contain errors. Automation executes rules or actions; it does not necessarily involve machine learning. These categories can coexist within a single platform.
In this book, human oversight means having the information, time, expertise and authority to review or stop a decision. Adding a human signature at the end of a process does not guarantee effective review. Responsibility must be assigned before deployment and maintained throughout the life cycle.
Questionnaires
Chapter knowledge & Human Resources dilemmas
Dilemmas receive no moral or employability score and are not used to assess job candidates.