Google just blew a hole in the Anthropic-OpenAI investment thesis

Everyone has been talking about Anthropic and OpenAI building a business around staying six months ahead of China’s open-weight models. That lead is supposed to justify enormous spending, expensive subscriptions and premium API pricing.

But Google is now making that argument a lot harder. Gemini 4 Argon is competing at the frontier. Carbon is aimed at demanding coding and cybersecurity work. Then Google released Hydrogen, its faster, cheaper model, which reportedly edged out Argon on an independent benchmark. Google is competing at both ends of the market, not just trying to catch up.

And Google has something Anthropic and OpenAI don’t: Search, YouTube, Android, Cloud, advertising revenue and a balance sheet that can fund AI development for years without depending on the same business model. Meanwhile, Chinese open-weight models like Kimi are putting pressure on API prices. Kimi K3.1 Pro and Gemini 4 Argon both launched at competitive prices, with Google’s introductory pricing set to rise in December.

So the market keeps treating this like a two-horse race between Anthropic and OpenAI, with China threatening their lead. Google makes it three, and arguably has the deepest pockets of the group. If frontier models become cheaper and good-enough models become nearly free, the economics Wall Street has been assigning to the AI labs could look very different. Is the market actually pricing in what Google is building?

https://deepmind.google/models/gemini/

https://siliconainews.com/tracker/ai-model-releases?check=backed&kind=frontier&maker=google

https://modelpricewatch.com/reports/state-of-llm-pricing-2026-11/

https://www.reuters.com/business/google-goes-laggard-leader-it-pulls-ahead-openai-with-stellar-ai-growth-2026-02-05/

Got a news tip or correction? Let us know

If you got something out of this, please chip in to keep this site running, or subscribe to go ad-free.

0 views