Meta

We conducted extensive safety evaluations before deployment, following the Advanced AI Scaling Framework, which defines evaluations, threat models, and deployment thresholds for our most advanced models.

Requiring parental approval and age verification at the App Store/OS level provides a centralized, consistent, and privacy-preserving place for age assurance, rather than requiring every individual app to comply with different rules.

We are on our way to personal superintelligence: an assistant that can help anyone, anywhere with the things that matter most to them.

Our open source approach also helps us to better anticipate and mitigate risk because it enables us to learn from the broader community’s independent assessments of our models’ capabilities.

Until consensus develops, it is premature to require specific measurements and benchmarks.