It's a really tough question of when to pull the plug. Marvin Minsky wrote a whole book in 1969 'proving' that neural nets cannot, fundamentally, do anything interesting - yet, here we are.
Dead wrong. Predicting protein folding has been a scientific holy grail for decades, and neural network-based AlphaFold is the leader. No non-neural approach comes within a country mile of its performance. They created a fold database of unprecedented coverage which is being used in research.