The bit cell is programmed by applying a high-voltage pulse not encountered during a normal operation across the gate and substrate of the thin oxide transistor (around 6 V for a 2 nm thick oxide, or 30 MV/cm) to break down the oxide between gate and substrate. The positive voltage on the transistor's gate forms an inversion channel in the substrate below the gate, causing a tunneling current to flow through the oxide. The current produces additional traps in the oxide, increasing the current through the oxide and ultimately melting the oxide and forming a conductive channel from gate to substrate.
So, basically, they intentionally apply an out-of-spec voltage on the cell's output port, overloading the gate and causing a permanent short to ground. The cell always reads as 0 afterwards.
These fuses have always been around in microcontrollers. They are used to configure various aspects of the microcontroller operations, like startup sequences, whether or not the contents of the chip can be read out, is their voltage monitoring (brownout detection) enabled, is there a watchdog timer enabled which could reset the chip automatically if needed, etc.
It is common that fuses like this can only be set to progressively stricter settings. And the only way to reset the fuses is to erase the entire chip, firmware and all. It sounds like these fuses in the Nvidia dont even allow this.
a clever but despicable tool.