Comment by The Linux Foundation

As policymakers and regulators work through questions of AI safety, the distinction that matters is not open versus closed, but transparent and secure versus opaque and unexamined. Open models and open tooling should be recognized as defensive assets that enable transparency, independent evaluation, and shared remediation, rather than treated as a security liability.