Researchers say AI ethics critics treat several different goals as one
Researchers at Bayreuth, Saarland and CISPA sort AI ethics into three objectives and four domains in a paper in AI and Ethics. They argue that criticism of one instrument, such as company guidelines, should not automatically call the whole field into question.
The Scientist · Science desk
What happened
- Critics argue that companies publish guidelines for AI development and use while the AI systems themselves often remain problematic.
- The Bayreuth, Saarland and CISPA authors say debates on AI ethics are frequently shaped by a fundamental misunderstanding.
- The three objectives are generating knowledge about the moral, social and normative aspects of AI, communicating it, and applying it to build and deploy systems responsibly.
- The four domains are research, education, practical development of AI systems, and governance such as ethics boards, guidelines and legislation.
Why it matters
- decision Anyone judging an ethics program has to say first whether it aims at producing knowledge, communicating it or applying it, because each goal calls for a different test of success.
- constraint The authors argue ethics cannot be folded into legislation or handed to market mechanisms, since research and public debate set which rules and values apply at all.
- capability If the authors are right that ethics committees and academic assessments often respond faster than legislation, those bodies would have a role in the interval before statute arrives.
Lead author Astrid Schomäcker, a research associate at Bayreuth, framed the paper's central move as a change of question [13]. "The term AI ethics is used to cover a wide range of activities that pursue different objectives. The question should therefore not be 'Is AI ethics useful or not?' but rather: Which type of AI ethics is useful for which purpose?" she said [3].
The test this framing implies is narrower than the debate it replaces. According to the phys.org account, the authors grant that certain ethical guidelines may have only a limited influence on how AI systems are actually developed, and they say this does not make research, education or other governance mechanisms equally ineffective [7]. Checking the guideline claim means comparing systems built with and without the guidelines. Whether ethics research produces knowledge is a separate study with a separate yardstick.
The authors work inside the field they assess. Schomäcker, Luca Deck, Timo Speith and Lena Kästner are all members of the AI Responsibility Centre at the University of Bayreuth [6]. Speith said the paper takes no single side: "In this publication, we do not defend AI ethics from any single perspective." [10]
Kästner put the conclusion in two sentences [11]: "AI ethics is neither a panacea nor meaningless. Its value depends on the objective being pursued and the instrument being employed."
The thing this doesn't tell you is how large any of these effects are. The phys.org account describes a taxonomy and an argument. It does not report how far guidelines, ethics boards or legislation change what developers build, or how fast committees and legislatures each respond.
In our view the distinction holds as logic, and a reader auditing an ethics program should name the objective before judging the result. We would drop that view if the objectives prove inseparable in practice, for example if research never reaches the people building systems. The full paper, at DOI 10.1007/s43681-026-01326-6, may supply evidence the summary leaves out [12].
What to watch
- Whether any study measures how far ethics guidelines, boards or education change the systems developers ship, the test the summary does not report.
- Whether the full paper offers evidence for its claim that ethics committees respond to new technology faster than legislative processes.
- Whether critics of company AI guidelines start naming the objective they say has failed, as the authors propose.
Clarity's read
What the record supports and how the coverage leans. The claims behind it follow.
Reality
- Evidence30
- Adoption
- Insufficient
- Hype gap−5
- Incentives40
- Confidence55
Claim ledger
Ranked by verification strength, evidence, and original report placement.
- [1]
Researchers from the University of Bayreuth, Saarland University and the CISPA Helmholtz Center for Information Security say debates on AI ethics are frequently shaped by a fundamental misunderstanding.
- [2]
Critics argue that while companies publish guidelines for AI development and use, the AI systems themselves often remain problematic.
- [3]
Astrid Schomäcker said: "The term AI ethics is used to cover a wide range of activities that pursue different objectives. The question should therefore not be 'Is AI ethics useful or not?' but rather: Which type of AI ethics is useful for which purpose?"
- [4]
The researchers identify three fundamental objectives of AI ethics: generating knowledge about the moral, social and normative aspects of AI, communicating that knowledge, and applying it in practice to develop and deploy AI systems responsibly.
- [5]
The publication identifies four key domains in which AI ethics operates: research, education, the practical development of AI systems, and governance mechanisms such as ethics boards, guidelines and legislation.
- [6]
Astrid Schomäcker, Luca Deck, Timo Speith and Lena Kästner, all at the University of Bayreuth, are members of the recently established AI Responsibility Centre at the university.
- [7]
A central finding is that criticism of particular instruments should not automatically call the entire field of AI ethics into question: it may be justified to argue that certain ethical guidelines have only a limited influence on the actual development of AI systems, but this does not mean that research, education or other governance mechanisms are equally ineffective.
- [8]
The authors say AI ethics must not be reduced solely to legislation or replaced by market mechanisms, because research and public debate are necessary to determine which rules and values should govern the development and use of AI in the first place.
- [9]
Ethics committees and academic assessments can often respond more quickly to new technological developments than legislative processes.
- [10]
Timo Speith said: "In this publication, we do not defend AI ethics from any single perspective."
- [11]
Lena Kästner said: "AI ethics is neither a panacea nor meaningless. Its value depends on the objective being pursued and the instrument being employed."
- [12]
The findings are published in the journal AI and Ethics (2026): Astrid Schomäcker et al., 'What is AI ethics, and what is it good for? Clarifying debates on AI ethics', DOI 10.1007/s43681-026-01326-6.
- [13]
Astrid Schomäcker is a research associate in the Philosophy, Computer Science and Artificial Intelligence research group at the University of Bayreuth and lead author of the publication.
Get this ledger inside ChatGPT. Ask about any of our stories and the answer brings its figure, each claim’s verdict and the sources. Free, no account.
Sources
1 independent publisher whose own reporting we read for this story.
- phys.orgAI ethics analysis clarifies why different goals call for different ethical approaches
1 article · October 14, 2026
Topics and entities
Follow any of these and your For You feed starts watching them — no settings page required.
Topics
- AI EthicsFollow
- AI GovernanceFollow