While it is a fact that the currently strongest chess engines are using alpha-beta search and not neural networks, this is basically an artifact of history combined with the desire to utilize the hardware the most optimally. It's not a value judgement on AI's ability to play chess:
Deep learning chess playing AI like AlphaZero use neural networks and not tree traversal:
AlphaZero - Wikipedia
"Danish grandmaster Peter Heine Nielsen likened AlphaZero's play to that of a superior alien species. Norwegian grandmaster Jon Ludvig Hammer characterized AlphaZero's play as "insane attacking chess" with profound positional understanding."
"Former world champion Garry Kasparov said it was a pleasure to watch AlphaZero play, especially since it plays in an open and dynamic style, as he does."
Those characterizations by human chess grandmasters do not suggest that AI has fundamental weakness in static analysis. And note that this was the ultimate NN exercise:
"AlphaZero was trained solely via "self-play" using 5,000 first-generation TPUs to generate the games and 64 second-generation TPUs to train the neural networks, all in parallel, with no access to opening books or endgame tables. After just four hours of training, DeepMind estimated AlphaZero was playing at a higher Elo rating than Stockfish 8; after 9 hours of training, the algorithm decisively defeated Stockfish 8 in a time-controlled 100-game tournament (28 wins, 0 losses, and 72 draws). The trained algorithm played on a single machine with four TPUs."
The AlphaZero paper was published only a few months ago. I expect hybrid engines to emerge - but in the end pure NN's will probably win on a power efficiency level - but for that NN acceleration hardware functionality has to be more widely available, most chess engines strive to be able to run on commonly available computing platforms.