Stories
Google Withholds Powerful AI Model Gemini 4 Argon Over Safety Concerns
F
Firstpost•01-10-2026, 05:30
Google Withholds Powerful AI Model Gemini 4 Argon Over Safety Concerns
- •Google is withholding its most powerful AI model, Gemini 4 Argon, from the public, releasing it only to vetted cybersecurity experts to prevent misuse by hackers.
- •The cautious rollout mirrors Anthropic's approach with its Claude Mythos Preview, and Google is voluntarily giving the US government early access for feedback.
- •The decision follows a White House meeting where tech executives, including Google's Sundar Pichai and Anthropic's Dario Amodei, signed an accord to police AI risks.
- •Cybersecurity experts fear advanced AI could be used to hack critical infrastructure like banks and hospitals; Argon excels at finding and fixing software flaws.
- •Google states Argon is designed to refuse requests that could aid cyberattacks or develop chemical, biological, or nuclear weapons, with safeguards similar to those in Anthropic and OpenAI models.
Key Questions Answered
Google withheld Gemini 4 Argon to prevent misuse by hackers and manage cybersecurity risks, releasing it initially only to a vetted group of experts and the US government.
President Donald Trump hosted top tech executives, including Google chief Sundar Pichai and Anthropic's Dario Amodei, at the White House to sign a voluntary accord policing AI risks.
Early testers used Argon to uncover a critical vulnerability in hospital software globally that exposed sensitive personal information, a flaw that other advanced models missed.
