Blog Post Examines Anthropomorphism in AI Governance

Atoosa Kasirzadeh and Mario G published a blog post on anthropomorphism in AI governance. Kasirzadeh, a staff research scientist at Google DeepMind, shared on X that the two had studied the topic extensively over recent days through conversations with several AI experts. They chose to write the post after finding their thoughts too extensive for a single thread on the platform. The article asks whether anthropomorphic descriptions should be used to interpret how AI models behave.