Actually, the only way to update the MiG Switch cart is via a file in the microSD side, not at/from the cart connection side. I mean, it would be kinda stupid they would have such a vulnerability, more so from a team that knows quite a bit about systems security. Now, there could still be a "first time only" update or bootloader protocol via cart connector, but again, such a first bootstrapper would probably be overridden after the first time a valid update was flashed or even if it was permanent, it would only work if the current flashed update failed a signature/CRC check.But the Mig Switch (or any flash card for that matter) is not just an SD card in place of a cartridge. There are FPGAs/ASICs that do all the heavy work to make the Switch think a valid gamecard is in there. I doubt that they implemented any sort of writing commands into the flash card if it was not needed for Switch 1 cartridges. More work to achieve nothing.
We are all speculating here but it is more normal to think about a) nintendo using some not properly implemented part of the protocol to detect imperfections ( in this case they would not be updating the Switch 1 firmware or cart reader firmware because there is not much to gain and giving such an update would give us potential source for reverse engineering the changes and learning how to overcome the fixes ) b) it is actually not on purpose but there is some side condition due to the natural differences of the new cart reader. In both a) and b), speed may be a big pivot/factor, maybe in general or maybe in certain parts of the protocol.
Always remember the carts do read and show an image, title, and so on and the Switch 2 does install a small amount of data once you put the linker cart in.













