I think software will be particularly unique in the abundance world because the exact same thing that made it so profitable is what will make it so 'worthless' - it has practically no direct real world cost. Cloth is abundant, yet you still pay a significant chunk of change for a shirt or pair of pants. But that's because there's still a significant real world cost there. Even if the production and processing can be heavily automated, there's still a very long process to go from cotton to pants at a store.
And then there's brand marketing and so on, which can drive prices to absurdly high levels. But when it comes to problem solving software. You just want the cheapest possible solution that works. And if we reach a level where you can even spin that solution yourself with a potentially poorly phrased, inconsistent, or contradictory spec (of the sort that developers already have to work with), then software as an industry is basically done. Not done as in dead, but done as is in there's just nowhere to go from here. Relatively few industries have any sort of comparable 'finish line', let alone the chances of us viably hitting it.