When a piston is moving towards the top on the exhaust stroke, the exhaust valve is already closing, and the inlet valve is opening, but in most cases, neither will be fully open. Timing will determine whether either are open enough to meet the piston. Nothing sits still, until the belt breaks.
That's true, of course - and the engine is designed in such a way that the valves won't make contact with the pistons in normal service, whatever the position of the variator.
Once the belt breaks, then the positional relationship between camshaft and crankshaft is lost completely, and if any valve is protruding into the cylinder beyond the safe clearance limit, then something will break when the piston next reaches TDC.
My point was just that the engine is either an interference engine, or it isn't; it won't come into or out of being an interference engine as the variator moves. You can't make an interference engine into a non-interference one by changing the valve timing. On most 4cyl engines, there is usually at least one valve fully open for any given position of the camshaft and if the belt breaks, that's going to make contact with a piston on an interference engine.
On engines with variable valve
lift (the TA is such an engine), then they may indeed only be interference in those power regimes where the valves are fully opened.
Whatever, it seems foolhardy to play russian roulette with a cambelt, and I certainly wouldn't want to rely on something I read somewhere about someone who had one break some time in the past and seemingly got away with it. Even the best case means waiting some indeterminate time at the side of the road for a breakdown service, and that's not somewhere I would want to be.