Microsoft is aiming to develop and deploy large, cutting-edge artificial intelligence models by 2027, a move framed as building in-house alternatives to existing powerful AI tools from OpenAI and Anthropic. The plan, reported by Matt Day in the Australian Financial Review on April 3, 2026, shows another giant corporation racing to concentrate more technical power inside its own walls. **Who Has the Power** Mustafa Suleyman, chief executive of Microsoft AI, said, "We must deliver the absolute frontier," and added, "Certainly by 2027, the objective is to really get to state-of-the-art" across models capable of responding to or generating text, images, and audio. The language is polished, but the direction is plain: more advanced systems, more corporate control, and fewer outside dependencies for one of the biggest firms in the industry. The target is not just to build models, but to replace reliance on existing powerful AI tools from OpenAI and Anthropic with Microsoft-owned alternatives. That is the hierarchy in miniature: a dominant company seeking to internalize even more of the infrastructure that shapes how text, images, and audio are produced and controlled. **What the Company Wants** The article says Microsoft plans to create large, cutting-edge AI models by 2027. It does not provide specific model sizes, compute budgets, or technical details in the material provided, but the objective itself is clear enough. Microsoft is pushing toward state-of-the-art systems that can handle multiple forms of output, including text, images, and audio. The company’s stated goal is to build in-house alternatives to tools from OpenAI and Anthropic. That means the competition is not about serving ordinary people better; it is about which corporate apparatus gets to own the frontier and extract value from it. The report was published by the Australian Financial Review on April 3, 2026. No grassroots response, worker organizing, or community alternative appears in the source material. The only actors named are Microsoft and the corporate AI firms it wants to outpace. **The Frontier as Corporate Territory** Suleyman’s words make the power dynamic explicit. "We must deliver the absolute frontier," he said. "Certainly by 2027, the objective is to really get to state-of-the-art." In the language of corporate AI, the frontier is not a public good. It is a competitive zone where the biggest players fight to dominate the tools, the models, and the terms of access. The article presents Microsoft’s plan as a strategic push, but the underlying structure is familiar: a giant firm expanding its control over a technology stack that increasingly mediates communication, creativity, and automation. The people at the bottom do not get a vote in that process. They get the products after the fact, packaged as progress. The source gives no indication of any reform, regulation, or public oversight attached to the plan. It is simply Microsoft aiming higher, with the rest of society left to live inside the systems these companies build and own.
