So the question is, has Deep Blue’s victory in1997 demonstrated that any game with perfect information is always solvable by computer analysis?
With some obvious limitations like end games or high-probability variations, there are no serious safety concerns about computer chess today. Chess engines often win against grandmasters because they use brute-force searching techniques to exhaust all human possibilities in a given position, rather than creatively analyzing it. The irony of this situation is that these same brute-force techniques cannot help but solve the game for computers in due course. Computers will eventually be able to generate an optimal strategy for any chess position immediately on the input of its pieces and squares.