There are kind of a lot of errors in this piece. For instance, the problem the author had with Gemini CLI running out of tokens in ten minutes is what happens when you don’t set up (a free) API key in your environment.
There is a time and a place for everything. Software development is often about compromise and often it isn’t feasible to work out a solution from foundational principles and a comprehensive understanding of the domain.
Many developers use libraries effectively without knowing every time consideration of O(n) comes into play.
Competently implemented, in the right context, LLMs can be an effective form of abstraction.
> Crazy you think that the company who invests the most in AI won’t retain any value in the transition to the space.
They’ll obviously retain some value. Google can integrate more easily and seamlessly with things people are using but that’s a competitive advantage not a moat.
Google’s trajectory has been from innovative market maker, to dominant market leader, to megacorp that has a strong established position that keeps them competitive and relevant, to legacy provider, to kind of irrelevant.
Google’s still strong and significant but every day they are less so.
I'm sure machine learning has already been introduced in one way or another in almost all of Google's services. But the implementations are mostly in the backend and enhancements like better recommendations that don't jump out as incredible leaps in artificial intelligence. They are almost exclusively incremental rather than radical innovations, quantitative and not qualitative improvements.
In terms of machine learning technology that introduces truly novel innovations Google's product portfolio is notable barren. For instance the incredible powerful potential for image generation these new diffusion models open up, who's models will the world use to explore the potential and start using this technology? Google's model with the intense, though imperfect, effort that goes into addressing questions of bias and abuse? Or the model bankrolled by an ex hedge fund manager who probably put a bit less thought into addressing these questions?
Google has done incredible work on basic machine learning research that has enabled other individuals and organizations to do amazing things. In terms of actually implementing new machine learning technology, they've pretty much hobbled themselves to the point of irrelevance. In some ways, it may be for the best that they've ensured that the future of machine learning will be written primarily by those who hold opposing views.
“If they then choose to ignore the indicators and ride with the In&box inactive, that's on them and we can expect it not to inflate in the event of a crash.”
When their product doesn’t work, one of their customers dies, and they need to explain to the deceased children why their dads not coming home; I wonder how convincing the “that’s on them” explanation will be.