ftdibrick

But bricking retroactively in the field is a bad move, and sends even more anger and RMA requests back towards parties. https://github.com/therealdreg/ftdibrick Pepperidge Farm remembers the FTDI driver bricking.

This isn't even the worst payload ever delivered through Windows Update. The prize for that should probably go to chip manufacturer FTDI, which once abused the system to publish a driver that would semi-permanently brick USB serial bridge parts the driver detected as counterfeit [1] by exploiting a command that the genuine parts did not implement correctly (how ironic) [2]. The backlash was large enough that Microsoft ended up pulling the update almost immediately, but that did not stop FTDI from trying again a few years later with another driver update that deliberately corrupted data sent through detected-counterfeit parts. [1] https://en.wikipedia.org/wiki/FTDI#Driver_controversy [2] https://github.com/therealdreg/ftdibrick#diving-deep

A similar, even higher profile case that shook the electronics industry around a decade ago was chip manufacturer FTDI releasing an update to their drivers that would detect and semi-permanently brick clones of FTDI USB serial bridge chips [1]. The bricking was performed by setting the USB product ID to zero, preventing Windows and macOS from detecting the device at all; the Linux drivers quickly got updated to recognize the new PID, allowing for the development of unbricking tools. Somewhat ironically, the detection relied on errata of the original parts that the clones fixed [2]. The backlash to this measure was massive, as many legitimate products turned out to use counterfeit FTDI parts without the manufacturers' awareness due to unreliable supply chains. Microsoft quickly pulled the update but FTDI seemed not to care for the most par