Here's my problem with the "It's the chip" theory....
First, it's not an actual failure of the chip. The chip works fine. It's firing the mod so it hasn't failed. It would have to be a software glitch rather than a hardware failure of the chip.
Fine, I work with software glitches all of the time. However, this isn't a single glitch, it requires a minimum of two separate glitches, one happening right after the other in the right order.
Your first glitch would be the chip waking itself up out of sleep mode without any input. The programing on this would be a simple open/closed reading of the circuit from the fire button. If it's open, the chip sleeps, if it closes the chip wakes up. But this glitch would require the chip to wake up despite never getting that closed reading from the circuit.
Then it has to glitch again, this time firing the mod non-stop. This is an entirely different glitch from the first because the chip cannot fire if it is asleep. It has to wake up, read the atomizer and battery, and then fire, again, with the circuit to the fire button being open and no current flowing through it. That would require the chip be physically wired to bypass the fire button. It can't pass voltage from the battery to the atomizer with no current going through the firing button unless the firing button is on an entirely separate circuit and the chip is handling voltage transmission between the two separate circuits.
But, again, this would have to be a software failure, so we are looking at the software making two changes to two different circuits, neither powered, and without any input.
And it would have to only happen to one mod, despite ALL of these mods using the same chips and the same software.
Or it could just be a collapsed contact inside a button.