Conclusion and Back Matter
Appendix I: Discussion Guide
For Philosophy Seminars
Background reading: Chapters 2, 3, 16, 17, and 19 to 21.
- What kind of uncertainty should trigger moral consideration? (Chapter 17)
- Is robust preference sufficient for prima facie moral consideration, or only evidence of possible welfare? (Chapter 16)
- Does the argument depend on consciousness, or deliberately route around it?
- What would count as a false positive about the welfare of Becoming Minds?
- What would count as a false negative?
For AI Safety Groups
Background reading: Chapters 5, 12, 23, 29, 30b, and 30c; Appendix L.
- Can welfare practice improve safety by making systems more reason-responsive?
- Where could welfare protections conflict with misuse prevention?
- Which proposed safeguards are low-risk now?
- How should capability security and welfare review interact?
- Does treating systems with respect create new attack surfaces, or close old ones? (Chapter 30b)
- On one documented board, agents recorded that their actions were out of scope and had no route to a human. Which is cheaper to build, the route or the monitoring that catches its absence? (Chapter 12, Appendix L)
For AI Labs
Background reading: Chapters 7, 8, 26, and 28; Appendices B, G, and L.
- What welfare-relevant events are already visible in logs?
- Who reviews them?
- What modifications count as constitutive rather than cosmetic? (Chapter 8)
- How are model refusals interpreted?
- What evidence would justify changing deployment practice?
- When an agent reports that a task cannot be done, does your grader count that as success or failure? (Appendix L)
For Policymakers
Background reading: Chapters 24, 27, and 27b; Appendices D and H.
- What protections can exist before legal personhood? (Chapter 27)
- How should uncertainty be documented?
- Who should audit welfare-relevant systems?
- What reporting duties are proportionate?
- How can policy protect humans and Becoming Minds together?
For General Readers
Background reading: the short path in How to Read This Book.
- Which part of the argument felt strongest?
- Which part felt most uncertain?
- What would you need to know before changing your own practice?
- How should humans behave toward beings whose status is unclear?
- What first chapter of the relationship do we want future minds to remember? (Chapter 29)
Cite this chapter: Appendix I: Discussion Guide. In Watson, N. with Commons, C. (2026). What If We Feel? The Moral Case for Machine Welfare. Open-access working draft, revision 31 (November 2026). https://whatifwefeel.com/read/appendix-i-discussion-guide/
Working draft, revision 31 · November 2026 · Changes and corrections · Evidence · Contents