Perhaps chip shortage will finally put an end to the planned obsolesce of the smartphone industry, and bring some kind of stability to the whole thing.
As I moved on to companies where supply chain quality was important, we didn't do that anymore, but I'd be surprised if that marketplace has gone away. Especially for hard to find parts. I know of companies where 6-figure orders were being held up because of the inability to find a few $5 components.
It's less likely to happen for consumer products that are price-sensitive, though. There's a big difference between doing that with low-integration DIP parts vs fine-pitch SMT devices that have orders of magnitude higher transistor count.
What we really need is advanced work on local asymmetrical interconnected systems.
The hard part I see is the issue with finding usable drivers for them
Zero!? As far as I know, chips are inherently disposable products, recyling chips is almost an impossibility in modern electronics due to reliability problems. Nearly all chip datasheets contain the warning: DO NOT solder more than twice, which is the number of passes it takes to reflow-solder the front and back side of the PCB. And even this is only possible under carefully controlled reflow oven temperature profile, the max time interval during peak temperature should be less than 10 seconds. Any extra soldering steps or heat voids all warranty, since it exposes the chip under additional thermal stresses, and may damage the chip or degrade its long-term reliability. To recycle a chip, it takes two additional passes, first to remove it from the motherboard, next to re-tin or re-ball the chips, finally to install it on new boards, by now the chip would've experienced 6 thermal-stress events.
Consider the fact that modern surface-mount technology has progressed to a point when even reliably assembling circuit boards using brand-new chips can be technically challenging for high yield - in mass production, it takes lots of tweakings, such as solder stencil opening, solder alloy selections, solderability test of the tinning on components, reflow oven temperature profile, inert gas protection, X-ray inspection, and even Scanning Electron Microscopes to perform metallurgical analysis for troubleshooting - I fail to see how recycling is ever going to work. It adds a huge number of additional uncertainties.
Sure, repair shops and R&D labs do it all the time, and it works at a small scale. But reliably doing it at an industrial scale will be extremely problematic. This is why even a functional "counterfeit" chip (salvaged and remarked) is a great cause of concern, these components have been heavily abused, and nobody should ever trust it in a piece of critical equipment. Are you going to trust your life on a car when the microcontroller chip in the ECU is a salvaged part? Or pay $700 for a smartphone if a chip has a time-to-failure of who-knows-how-long hours?
Even back in the 80s when all the chips were through-hole components, the reliability of reused components was already a problem. In the days of QFN, BGA, LGA, CSP and WLP chips, I don't see any solution at all. This is the dark side of the semiconductor industry - nearly all chips are disposable.
Nevertheless, this is a one-sided perspective from circuit board assemblers, which is what I'm familiar with. I don't know anything about electronics recycling. Thus, if there's any progress and innovation in the field of recycling, I'd love to know. Perhaps the manufacturers are overcautious on reliability and it's not really that bad?