Gilbert spliced into several wires on the throttle pedal harness, shorted two of them together through a resistance, and then shorted them to sensor power. In other words, he induced a multi-point failure, with very specific parameters. If he had done the splicing in a different order, or shorted different wires, or used a different resistor, it would not work.
Gilbert proved that you can re-wire a car to misbehave. He didn't come even close to proving that this is what is going on in the field.
ABC proved that it would rather stir the pot and get people excited, while providing ammo for trial lawyers, than report responsibly on a highly technical issue. They even threw in some faked tachometer footage to make things look more dramatic--hilariously, with all the telltales illuminated, showing a vehicle at rest, with doors open, and in park.
I would like to see ABC and Gilbert now defend their piece, by explaining how it is relevant to the Toyota untended acceleration issue.