Yes, b5 in these lines does seem difficult to justify. 
I was just exploring some of the a4 anti-Berlins, due to the Caruana-Carlsen game in Tata Steel. Moving the b-pawn liberates the rook pretty nicely. This is one of the lines in this c3 modern I've been laboring over with a similar motif. Black could justify the pawns by looking for activity on the b-file+Ba6, most likely liquifying.



Edit: Updated the study. I'm done with the 8.a4 lines, there is still this 8.Bb5 where I'm not sure about the best approach. Na3 is also still lacking lines but will look to add them asap. Maybe I'll also bother to add more references and spellcheck it.

I'd agree with the ECO on that.
Thinking of the differences between a Pirc/modern and the Ruy, a6->b5 is quite useful in the c6 operation to undermine a pawn on d5. Maybe there is an illustrative, existing term for this type of knight-pawn-total-undermining-operation?



This is a rare instance of it being deployed on the kingside. Apparently, that is a critical Nimzovitch defense position... well for engines anyway, I'd stick to 2.Nf3.
https://www.youtube.com/watch?v=cIFZwDLdeaA
The position comes up around the 11:50 mark,

kylemeister wrote on 01/22/23 at 22:10:23:

Interesting how people have these intersecting definitions. I'm not that into semantics, at the end of the day what is there is there, call it whatever serves the purpose.

In the line I looked at Bd7 probably would transpose to the fianchetto Ruy. I think the a6 move makes a better case, seeing that the modern isn't deployed to get into e4-e5 theory, and Bxc6 looks justifiable for black with some Rossolimo-type plans.

https://www.chessgames.com/perl/chessgame?gid=1886521

I decided to try conceptualizing a little bit since some of the ideas aren't very self-explanatory and found the PGN alone to be a bit too cryptic.

https://drive.google.com/file/d/1iJQlEH3yf8M_8Uz4EAwXdrFcxxwLTiWR/view?usp=share...

It's just a passion project, it's still not quite "done" and by no means fine literature. I ended up a bit unsure of what done means here since I originally had no plans of making a study.

I thought I'd want to share it though, seeing that the ideas I've posted here are a bit here and there and green. Didn't add the lines after 8.Na3 yet, since there are many good approaches, and I'm pondering on which makes the most conceptual sense.

>if the diagram frightens black then please stay away from the Spanish Breyer, the Orthodox QGD, etc.

Well, the Breyer and Orthodox QGD are orders of magnitude more popular than whatever this is, so this sort of argument by association doesn't seem very compelling 

With high respect to Balashov, black seemed lucky to get away in that game. I think the positional assessment of these old-Benoni structures continues to go down due to the long-term issues with the dark-squared bishop, space, and d6. Black can't combat white on the queenside, whereas the battle on the kingside is much more even.

Yesterday I read this Pterodactyl book.
It got me thinking about this straight-to-the-point move order:

1.e4 g6 2.d4 Bg7 3.Nf3 d6 4.Bd3 c5
It provides more direct pressure on d4 than the other ways to reach the Alapin.
5.c3 cxd4 6.cxd4 Bg4 7.Nbd2 Nc6 8.h3 Bd7 9.Nb3 a5
White is forced into concessions of allowing a4 or letting go of d4!
Of course, these are critical too:
5.d5 Nf6 
The early Bd3 removes the irritating Bc4 and Bb5+ options that usually hinder e6. Black ofc still has more to prove. 
5.dxc5 Qa5+ 4.c3 Qxc5
With a pretty reasonable-looking position.

Now funnily enough the line pointed out by CbT:
1.e4 g6 2.d4 Bg7 3.Nf3 d6 4.h3
Might indeed deserve the "most dreaded of Anti-lines" title. As it stops Bg4 and thus kills the direct play for d4.
And either black goes back to the normal Alapin lines or into this:
4...Nf6 5.Bd3 e5 6.dxe5 dxe5 7.Nxe5 Nxe4 8.Nxg6 hxg6 9.Bxe4 Qxd1 10.Kxd1
A line I had originally written with faulty notation

About:
1.e4 g6 2.d4 Bg7 3.Nf3 d6 4.Bd3 Nc6 5.c3 e5 6. h3 Nf6 7.O-O O-O
In the main play 8.Re1 Re8 9.d5 Ne7 10.c4 with the Rook on e8 the f5 plan lacks power. Without Re8 black has little hope for timing d5 effectively. 
White play will most likely follow with slow improvement:
1. Bc2 opening the queen to avoid Be3 being met with d5
2. Be3 developing, keeping options open for the b1 knight
3. Nbd2/Na3 now Re8 doesn't need to be met with d5, I see no way for black to have kept their d5 break alive. The one exception I can think of is black playing b6->Bb7, which would again lead to d5->c4->Nc3 and a speculative KID. 
The game goes on...

In the previous post, I forgot to mention:
8.Be3
This move, like Re1, justifies the g5->g4 scheme.
I'm still working on the PGN about the h6 move, but I think I've already clogged this thread enough so won't post more specifics until it's done. I will say the a4 and Na3 lines do seem the most positionally difficult.

Ok, looking at the idea of Nge7 I immediately notice that the "dreaded" 
1.e4 g6 2.d4 Bg7 3.Nf3 d6 4.h3
does cause an issue. and black would probably have to resort to: 
4...Nf6 5.Bd3 e5 6.dxe5 dxe5
Which looks playable. However, I'm Not terribly fond of:
7.Nxg6 hxg6 8.Bxe4 Qxd1 9.Kxd1
Where the compensation reeks of the e5 classical or Bh6 Pirc. White has quite a few simple tries to consolidate and black has to do all the proving. 
And what about white not taking on e5 with the knight... or playing 6.c3. So I firmly agree with you that making the Nc6 stuff work would be the compact option.

In order to not throw the Leela idea in the garbage quite yet, maybe h6 could be thought of more as a waiting move than as a commitment to g5. For example:
1.e4 g6 2.d4 Bg7 3.Nf3 d6 4.Bd3 Nc6 5.c3 e5 6. h3 Nf6 7.O-O h6 8.a4 a5
And black could try to prove that the b4 square allows plans such as dxe5->Nb4 followed by b6 or something
8.Nbd2
And black might get away with g5 again, due to the clumsy knight maneuvers.
[EDIT: g5 is not very viable here, the Fish hits it with a simple dxe5 dxe5 Bb5, where the knight conveniently blocks out Qxd1. So black needs another argument against Nbd2]

Then maybe cook up something against
8.Na3