Artificial intelligence is increasingly being applied to animal welfare — from automated welfare monitoring systems to AI-assisted drug development alternatives — with significant potential to improve welfare outcomes at scale.
AI applications for animal welfare represent both an opportunity and a risk. Opportunity: automated, continuous welfare monitoring could detect problems earlier and more reliably than periodic human assessment, potentially reducing the duration of welfare harms before intervention. Risk: AI-enabled monitoring could be used to manage welfare problems within intensive systems rather than address their root causes, or could create false confidence in welfare outcomes without genuine improvement. The welfare value of AI monitoring depends entirely on what is done with the information it generates. AI-assisted drug development alternatives are more unambiguously welfare-positive — reducing the need for animal tests by enabling more effective in silico screening.