Stanford University professor and World Labs co-founder Fei-Fei Li stated that increasingly capable AI systems require safety evaluations beyond internal company testing. While developers can assess model capabilities and safety properties internally, Li emphasized the need for academia, government, and industry to jointly establish broader independent evaluation standards. Li asserted that AI risks depend on human development, governance, and usage, stressing that humans must retain control over the technology. She also disclosed that World Labs is developing "world models" capable of generating and reconstructing three-dimensional environments, with applications in robot training and evaluation scenarios.