We Need Positive Visions for AI Grounded in Wellbeing

The Gradient · 2 years ago · 24

Researchers argue that AI development should be grounded in human wellbeing rather than abstract principles, drawing on scientific literature about what makes lives flourishing. The wellbeing framework includes factors like supportive relationships, meaningful work, growth, and achievement, with models like Seligman's PERMA and need-satisfaction theories providing measurable dimensions. AI systems should be designed and evaluated to support these concrete wellbeing outcomes across both individual lives and societal institutions like education, government, and media, requiring both positive visions of beneficial AI futures and technical implementation through new algorithms and training approaches.