Despite Microsoft's First-Mover Advantage, Google Still Has a Chance to Overtake

The original article (微軟雖然先發，Google 仍有機會反超) was written in Traditional Chinese. Below is the version translated by AI.

• Microsoft's Azure services worldwide AI traffic (ChatGPT/Azure OpenAI Services/xxx Copilots/...), its GPUs are all used to make money, squeezing out resources for research and development.

• Google and other companies' GPUs are idle, perfectly poised to fully invest in building larger models. Moreover, Google has access to a vast amount of global internet information. Coupled with the fact that today, there is much more information related to LLM/multimodal/long context/high efficiency than in 2021, it is highly possible that there are more efficient/better methods that can overtake in a short period of time.

• Switching conversation/API/GitHub Copilot is very easy. If Google and other companies launch better and cheaper models and APIs in the future, users and various applications can easily switch to better options.

• Microsoft's Windows/Office/Teams/VSCode/... etc. are still a strong moat, and it's difficult to switch from these. Will users be tied down by these Microsoft links in the future, unable to leave even if they want to?

• How much of Google's market can Microsoft seize with AI during this time when Google is catching up?

• The only question for Google is: Can the dinosaur still dance?

• If Google can't dance, who else can compete with Microsoft? Apple/Meta/Amazon?

  Thank you for reading this far. If you're also interested in unique perspectives on AI development and predictions about the future world, feel free to subscribe for free to receive new articles and support my continued writing.