The Next Experiment: The Research Discussion Series

摘要

This volume collects ten research discussion papers proposing experiments derived from the Decalogy on Artificial Intelligence. Each paper identifies a structural gap in an active research program, proposes a hypothesis to address it, designs an experiment with specific predictions, and states explicit failure conditions. Topics include: a shared structural condition across three theories of consciousness (Paper 1), disciplinary identity under methodological convergence (Paper 2), connective abduction as formalized cross-domain inference (Paper 3), a complementary deployment-architecture frame for alignment (Paper 4), structural impossibility of capability-based task allocation (Paper 5), tail variability as the mechanism of human data irreplaceability (Paper 6), Distributed Agency Computation as an expansion of the robotics research program (Paper 7), Task-Normalized Energy as a cross-architecture efficiency metric (Paper 8), conditions for cross-modal signal integration via the Learned Interaction Pipeline (Paper 9), and image schema recognition as the mechanism of expert-AI integration with evidence from professional Go (Paper 10). Every paper includes proposed experimental designs with explicit disconfirmation criteria.