It certainly won't beat the market in what the market optimizes for -- namely risk adjusted return on capital. It probably won't beat the market on the implicitly coupled metric of "value delivered to customers" either.

The trick here is that "users" and "customers" have become all but totally decoupled when it comes to the modern internet. Not enough people were willing to pay, so the market has turned to other sources of revenue.

I share your skepticism about bureaucratic government agencies creating value, and doubly so when it comes to technology. At the same time, as an individual voter I am beginning to question whether "the market" is optimizing for the same things that I value.

Perhaps there is a natural conflict between maximizing utility for the majority of people vs maximizing utility for the majority of capital.

Exactly. Just like Airbus can't compete with the likes of Boeing.

I am not advocating for central planning. As I underscored in my comment, I advocate for something closer to the defense procurement model: where the market is failing to provide an appropriate off the shelf product and the state contracts an entity (usually a private company, but in the case of software it could be a public agency) to make it for them. This is a model that is currently in use in the US, in the EU, and actually worked well in the USSR too, their Ladas (and consumer products in general) might have been terrible but their defense industry was great.

It doesn't need to be better, it just needs to be good enough that certain features can pull users away from the incumbent (thus applying pressure to them).