With decent level of hardware and time control, I would rate the opening moves choices by that order
1. Leela first choice after 1 million position search
2. Top opening books/ database used by ICCF Grandmasters
3. Brainfish database
4. Top opening books/ database of 2700+ OTB super GM
5. Stockfish 10 search at 1 billions nodes per position
6. Common Opening explorer./Human opening theory.
P.S. These opinions are based on the results of TCEC engine games and also opinion from ICCF World Chess champion.( No 1, No 2, No 3 are extremely strong and hard to compare which is better)
Why I put too much weight on Leela, cos she is quite bad in tactics and endgames among top engines but she is extremely superb in opening and early mid game, it is the reason she is standing among top 4 engine in TCEC and CCCC.
I followed some computer matches. For example Stockfish - Houdini and Stockfish - Alpha Zero.
What I noticed is that the computer that plays reliable openings like Nimzo Indian Queens Indian Ruy Lopez etc scores better than a computer that has dubious openings lines in its openingbook like Traxler Gambit Schliemann Gambit of Chingorin defence or even King Indian. That shows In muy opinion that the openings that is taken randomly out of an openings book with dubious openings has a big influence on the match result. So openings are importent for the computers. Maybe if you let them free from move one to choose they can even become beter. And it must be possible to let them learn by analysed games and the next time let them not play again the losing move but an other move. This must be possible to program. That they have a data base of played games of them self.
That is interesting.But it could be hard to have evidence in the depth analysis that the opening will give an advantage, considering chess is a multi branched moves game. Perhaps there are thousands of branched moves after an umpteenth move.
That is an interesting result but difficult to be accepted objectively.