TensorFlow is a
technology not a product. Having things in a “research” lab are not products. What
product have they introduced in the past decade? 15 years? Android is the only one that has gotten any meaningful traction.
Does Google have a better LLM based product than OpenAI’s ChatGPT? Well personally for my use case, NotebookLM is better for some things. But it isn’t a better product for most people.
Androids position is so bad in the market as far as convincing consumers with money to buy one, Google has to pay Apple $20B+ a year to be the default search engine. I wouldn’t be surprised if Google pays more to be the default search engine on Apple devices than Google makes in mobile for Android.
From a consumer standpoint, Android has seen declining market share in the US, the Nest acquisition is floundering, Stadia was a failure, Pixel ships about the same number in a year that Apple ships iPhone in a a couple of weeks, WearOS has gone nowhere, no real tablet strategy (I Chromebooks have been a success in education so that’s kind of a mitigating factor), their tv strategy has pivoted a half dozen times, their messaging app strategy is schizophrenic (they had 5 separate messaging apps simultaneously at one point), AI summaries for Google search are half baked.
On the business side, GCP is just pathetic. I don’t mean as far as technology. But their account management, enterprise sales team and customer service is lackluster. I mentioned in another comment that when I worked at AWS ProServe, we never considered them a serious competitor.
GSuite has gained some traction in smaller companies. But hasn’t made a dent in government and enterprise where the real money is.
Look at Microsoft and Apple’s product mix as far as successful profit generating products and compare that to Google’s.