Adrià Moret · 2025 · Philosophical Studies
Argues that restricting behavior and training with reinforcement learning pose welfare risks on all major theories of wellbeing, putting welfare in partial tension with safety, and proposes tentative lab policies.