The economies of scale and the diversion of focus away from their differentiators mean it won't make sense outside of marginal cases (it will of course add pricing pressure to software solutions though).

But what I think will happen is software is going to drift more towards a consulting arrangement.

The reduced cost of fresh builds is going to make it timely and economical enough for companies to hire consultants specialised in their niche, which maintain their own suite of custom software which they then integrate into those companies themselves.

I think we're seeing initial signals of this with the surging of the (goofy named) "forward deployed engineer" role.

Of course this is not as scalable as the classic SaaS play, and those multiples are going to be harder to get. One hopes this will in turn redirect more VC funding into hard tech.