Resolution criteria

This market resolves to YES if Google makes its Gemini 4 Argon model publicly available to individual developers or consumers on or before October 14, 2026, at 11:59 PM PT. Otherwise, this market resolves to NO.

To qualify as publicly available, ordinary individual users must be able to obtain access without being individually selected, invited, vetted, or approved by Google. Qualified access pathways include:

• Developer Access: Available via Google AI Studio, the Gemini API, or Google Cloud Vertex AI for any registered developer.

• Consumer Access: Offered via a generally available consumer subscription (e.g., Google AI Advanced, Google AI Ultra, or a successor tier) on the Gemini web or mobile interface.

Exclusions and Edge Cases:

• Access does not qualify if it remains restricted to enterprise customers, select companies, private-preview cohorts, trusted testers, or members of specific closed initiatives such as Google's Fairwind program.

• Paid subscriptions or paid API usage qualify as public access, as long as any ordinary user can sign up and pay.

• Geographic restrictions will not prevent a YES resolution, provided ordinary individual users in at least one supported region can obtain unvetted access.

• The model must be functionally accessible and interactive by the deadline; a mere announcement of a future release date does not suffice.

Background

Google announced its next-generation frontier AI model, Gemini 4 Argon, on September 30, 2026. Engineered for complex reasoning, multi-step software engineering (scoring 77.9% on DeepSWE v1.1), and defensive cybersecurity, the model boasts a massive 1-million-token output capacity.

At launch, Google made Gemini 4 Argon available exclusively to trusted testers and select cybersecurity partners through its vetted Fairwind program. This market tracks whether Google transitions the model from its restricted early-access phase to general public availability for developers or consumers before the October 15, 2026 deadline.