In late January 2023, ChatGPT was reported to have reached approximately one hundred million monthly active users, having launched only two months earlier. The growth rate made it the fastest-growing consumer application in history. Within days, Microsoft confirmed a multi-year multi-billion dollar investment in OpenAI, widely reported at around ten billion dollars, that came on top of the billion dollars Microsoft had previously committed in 2019.
The numbers were staggering enough on their own. The strategic implications mattered more. Microsoft had positioned itself as the cloud and infrastructure provider behind the most-discussed product in technology, and through the same arrangement, had effectively secured rights to integrate the same capabilities into its own products. The Bing search engine, the Office productivity suite, and various developer tools were going to receive AI capabilities that nobody else outside of OpenAI had access to at the same level.
Google’s situation was the immediate counterpoint. Google had invented the transformer architecture that underpinned the entire modern language model approach. Google had its own large language models, including LaMDA, that were technically competitive with what OpenAI was deploying. What Google had not done was ship a consumer-facing AI product that captured public imagination the way ChatGPT had. The internal tensions inside Google about how aggressively to move language model capabilities into search were significant. The company would declare a code red and accelerate plans, with consequences that would unfold over the following months.
The competitive landscape had shifted faster than the existing strategies could absorb. Companies that had treated AI as a research area or as a long-term differentiator had to reframe it as an immediate product imperative. Companies that had been moving cautiously found their caution had become a strategic liability. The question of whether a particular AI capability would be released by year-end was no longer about technical readiness but about the cost of being seen to lag.
What was harder to assess in January 2023 was how the durability of the lead would play out. ChatGPT had captured an enormous initial audience. Whether that audience would stay, whether the feature gap would persist as competitors caught up, and whether the underlying economics of running large language models at consumer scale were viable, were all questions that would only be answered through the rest of the year and beyond.