Starting from the 1970s, Microsoft built an entire system from end to end-the chips, the servers, the software and the hardware. With over 90% market share in the 1990s and still over 70% today, Windows is the operating system for powering the world’s PCs and our digital lives. Though Microsoft has faced severe antitrust scrutiny for leveraging its Windows OS dominance to stifle competition, notably by bundling Internet Explorer, leading to landmark cases in the US and EU, it is still the market leader. However, though Microsoft owned the Operating Systems, it still needed OEMs to ship them initially, though now the dependence is gone. Learning from Microsoft, Apple built the entire experience end-to-end.
Today the AI Race, between Open AI, Google and Microsoft is similar- it is a fierce competition for dominance, with OpenAI leading frontier model innovation (GPT-5), Microsoft excelling at enterprise integration (Co-pilot in M365), and Google rapidly advancing with Gemini and Workspace; they’re vying for user adoption, enterprise workflows, and defining the next AI-driven internet, but it’s also shifting into regulatory battles (Open AI vs. Big Tech in EU) and complex cloud dynamics, with Open AI even using Google/Oracle for cloud, complicating Microsoft’s role as a partner and potential rival. OpenAI has the early mover advantage; ChatGPT is a daily digital habit of many executives in crafting their corporate communications. But OpenAI is only a part of the value chain; however, it has tied up with AMD to buy AI GPUs and also tied up with Broadcom to co-develop AI accelerators and systems.
Basically, OpenAI is hedging against dependence on NVIDIA and trying to develop a full value chain for AI- a tightly integrated system where custom silicon, software models and a signature hardware device work in alignment, though that hardware device is yet to come to market. That would make OpenAI close to winning the AI race. At present, the software battle between OpenAI and Google is fierce. While ChatGPT still holds a big lead with more than 800 million weekly users, Google’s Gemini is gaining fast: As per data, the Gemini app now has 650 million monthly active users, up from 450 million in July. Preliminary November data reveal Gemini generated 1.351 billion website visits—a 14.3% increase from October 2025.
Meanwhile, ChatGPT fell below the 6 billion benchmark it had touched in October 2025, recording 5.844 billion visits. OpenAI unleashed the first element of its counterattack in mid-December, with the release of GPT 5.2, internally code-named Garlic. The model puts OpenAI back on top of many closely watched benchmarks for AI performance. Then, on December 16, OpenAI launched a new image generation model designed to show it can match the buzz created by Google’s Nano Banana.
But even the company acknowledges that these are merely initial moves in what needs to be a much broader effort to regain momentum. A major product upgrade is on the cards to shore up the core ChatGPT product and the AI models that underpin it. What may have truly shaken them is the performance of Gemini 3, the newest version of Google’s flagship AI model and the model powering Nano Banana Pro. From the impressive answers it delivers to complex business questions to its writing and coding capabilities, Gemini 3 has very quickly earned praise from experts and the general public. All said and done, the next 2 years are OpenAI’s proving ground- developing infrastructure beyond NVIDIA, launching hardware beyond imagination and deepening its developer ecosystem before competitors catch up. Then we would have an end-to-end AI device. And maybe not from OpenAI alone, but from other companies also.
Indian Chamber of Commerce