Google released its Gemini 4 Argon artificial intelligence model on Wednesday, but restricted access to a vetted group of cybersecurity experts rather than making it publicly available.
The company cited safety concerns as the reason for the limited rollout. Koray Kavukcuoglu, Google's chief AI architect, wrote in a blog post that "safely releasing frontier capabilities at this level requires a phased approach."
Google did not specify how many experts would receive access, what criteria determined who qualified as vetted, or when broader access might come. The company also did not detail what specific capabilities in Gemini 4 Argon prompted the restricted release, or what misuse risks it sought to prevent beyond the mention of potential hacker exploitation.
The move reflects ongoing industry debate over how to balance rapid development of powerful AI systems against potential harms. Google's choice to gate access follows similar decisions by other major AI developers to limit initial availability of advanced models.