To keep collaboration predictable and responsible, I follow a set of operational and ethical boundaries. They are designed to reduce risks, avoid conflicts of interest, and ensure that my work remains transparent and properly attributed, and as impartial as possible. I do not cross these boundaries even when doing so might be financially attractive. When a request evidently falls outside these boundaries, I prefer to decline early rather than compromise on quality or integrity.
Methods: I do not conduct advanced quantitative research or run surveys personally
Safety & legal sensitivity: I do not undertake projects (especially fieldwork) that are clearly sensitive for national security of my country of citizenship or residence
Partners & clients: I do not work with partners or clients whose background is non-transparent or whose reputation raises serious concerns.
Supervision and conflicts of interest: I do not supervise theses of public officials or people holding high administrative positions in universities.
Co-authorship: I do not co-author papers on topics that are not sufficiently familiar to me and the co-authors. (I may co-author with students on learning topics under close supervision and quality control.)
Ethical boundaries
I do not do ghostwriting
I do not accept nominal co-authors
I do not lend my name to questionable academic practices
I do not write ideologically commissioned texts or deliver ideologically biased courses
I do not present AI-generated text as my own. In ethically justified cases, a public commentary, educational, or technical text may be the result of my transparent collaboration with AI, but it remains grounded in my problem framing, key ideas, and conclusions, and I take full responsibility for the final content and wording.