Google Unveils Gemini 4 Argon, Limits Access to Cyber Defenders
Google announced Gemini 4 Argon, a new frontier AI model, initially available only to trusted cyber defenders, with plans for broader rollout.
Google announced Gemini 4 Argon, its next frontier AI model, on [date]. The company is initially limiting access to a set of trusted cyber defenders, according to Koray Kavukcuoglu, who said Google is engaged in the U.S. government's voluntary pre-release model access process while gradually expanding access.
Gemini 4 Argon is already being used in Google's internal workflows, including large-scale codebase migrations. Google shared benchmark charts showing Gemini 4 outperforming models from OpenAI and Anthropic.
Before a broader rollout, Google plans to strengthen frontier safeguards, including defenses against misuse and prompt injection attacks and monitoring for misalignment.
Background
A day before the announcement, OpenAI held its DevDay conference where it launched its Dots AI agent and GPT-6.1 Sol AI model. Earlier in the week, OpenAI said it would not release a planned GPT-6.1 Astra model due to safety concerns. In August, Koray Kavukcuoglu was appointed as the new boss of DeepMind.
Quick answers
What is Gemini 4 Argon?
Gemini 4 Argon is Google's next frontier AI model, announced by Google.
Who can access Gemini 4 Argon?
Initially, access is limited to a set of trusted cyber defenders.
What safeguards is Google planning?
Google plans to strengthen frontier safeguards, including defenses against misuse and prompt injection attacks and monitoring for misalignment.