In my opinion, the chance of nuclear fusion "working" is so low that it's not worth what is currently being spent on it. To the extent that it is a worthwhile area of research, it should more resemble basic research in plasma physics, not targeted development at a limited number (or one) large scale concepts that are going to be dead ends.
Nuclear fusion is a necessity in assuring diversity among energy sources in the long term. It's probably much more fruitful an endeavour with eventual returns than macroscale particle accelerators.
