Ethical Machines Orchestrating Ethics
Nov 13, 2025
In a thought-provoking conversation, David Danks, a Professor of Philosophy and Data Science known for his work on AI ethics, explores the crucial concept of ethical interoperability. He discusses the risks of differing ethical standards when companies integrate AI models from multiple sources. Danks emphasizes the need for case-by-case ethical alignment and the challenges of accountability in AI deployment. He also delves into how transparency and operational clarity can enhance ethical assessments, urging firms and governments to recognize mismatched ethical practices.
AI Snips
Chapters
Transcript
Episode notes
MSG Facial Recognition Example
- David Danks uses an example of face recognition at venues like Madison Square Garden to show allocation of responsibility.
- He asks whether the model builder, deployer, or venue bears ethical responsibility for wrongful detentions.
Require Use-Case Transparency
- Do perform due diligence on clients' intended uses and request context when building models.
- Avoid building systems blind to downstream uses because you may become complicit in harms.
Agree On Use Cases, Not Principles
- Ethical interoperability doesn't demand identical principles across parties, but agreement that a specific use is ethically acceptable.
- Parties must locate overlap on concrete use cases rather than abstract principles.