"frontier AI models"

Open source and closed models have become competing strategic narratives in the global competition for AI dominance.

The Hugging Face breach demonstrates that attackers are already using AI agents, and that defenders can't afford to rely on frontier AI duri...

The order, intended to review artificial intelligence models that could pose risks to the US, is strictly voluntary.

Frontier AI models will not wait for internal approval, maintenance schedules or rollout plans to launch attacks.