In Yeonwoo's recent video on a game between Shin Jinseo (white) and Shin Minjun, she highlighted an interesting improvement AI found in a common corner shape: https://youtu.be/hFGagc1zUiE?t=684.

Black exchanged 1 for 2, as he saw it as a nice outside for inside exchange before jumping into the 3-3 (as he is going to give r6 to white):

Click Here To Show Diagram Code

[go]$$Bc
$$ +---------------------------------------+
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . X O . . . . . . . . . . . . . . . |
$$ | . . . X O . . . X . . . . . . . . . . |
$$ | . . X , . . . . . , . . . X . , X . . |
$$ | . . . . . . . O . . . . . . . . . . . |
$$ | . . . X . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . X . X . . |
$$ | . . . , . . . . . , . . . . . , . . . |
$$ | . . . X . . . . . . . . . . O . O . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . 1 O . . . . . |
$$ | . . . . . . . . . . . . . 2 . . X . . |
$$ | . . . O . . O . . . . . . . . . . 6 . |
$$ | . . O , . . . . . , . . . . . O 4 5 . |
$$ | . . . . X . . O . . . . . O . . 3 . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ +---------------------------------------+[/go]

Black then continued with the very normal solid connection, and played the standard sequence to live in the corner in gote:

Click Here To Show Diagram Code

[go]$$Bcm7
$$ +---------------------------------------+
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . X O . . . . . . . . . . . . . . . |
$$ | . . . X O . . . X . . . . . . . . . . |
$$ | . . X , . . . . . , . . . X . , X . . |
$$ | . . . . . . . O . . . . . . . . . . . |
$$ | . . . X . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . X . X . . |
$$ | . . . , . . . . . , . . . . . , . . . |
$$ | . . . X . . . . . . . . . . O . O . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . X O . . . . . |
$$ | . . . . . . . . . . . . . O . . X . . |
$$ | . . . O . . O . . . . . . . . . 2 O . |
$$ | . . O , . . . . . , . . . . 6 O O X . |
$$ | . . . . X . . O . . . . . O 5 7 X 1 . |
$$ | . . . . . . . . . . . . . 4 3 . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ +---------------------------------------+[/go]

But AI suggests the hanging connection, because now you live by playing the endgame 1st line hane. If white answers this and defends the 2nd line cut, black transforms a gote invasion to sente invasion (at the cost of strengthening white in the m3 area), a big gain at this early stage of the game.

Click Here To Show Diagram Code

[go]$$Bcm7
$$ +---------------------------------------+
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . X O . . . . . . . . . . . . . . . |
$$ | . . . X O . . . X . . . . . . . . . . |
$$ | . . X , . . . . . , . . . X . , X . . |
$$ | . . . . . . . O . . . . . . . . . . . |
$$ | . . . X . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . X . X . . |
$$ | . . . , . . . . . , . . . . . , . . . |
$$ | . . . X . . . . . . . . . . O . O . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . X O . . . . . |
$$ | . . . . . . . . . . . . . O . . X . . |
$$ | . . . O . . O . . . . . . . . . 4 O . |
$$ | . . O , . . . . . , . . . . 8 O O X 2 |
$$ | . . . . X . . O . . . . . O 7 . X 3 . |
$$ | . . . . . . . . . . . . . 6 5 . . 1 . |
$$ | . . . . . . . . . . . . . 9 . . . . . |
$$ +---------------------------------------+[/go]

NB if white does (now or later) block the outside, there is some ko aji at s1.

This technique was not part of my go vocabulary before, and apparently not Shin Minjun's either as once you know it it seems pretty easy to see situations where it could be good. By giving white the atari at t4, black loses the s6 aim, but in this position that white area is strong so that's not much of a loss. And the other aspect to make it good is white is invested in building the lower side, so breaking that area later if they ignore your 1st line hane should be a valuable move. And it shouldn't be so wide that white is happy to answer your premature endgame to make it thicker. Doing a pattern search on waltheri finds no examples of it before.

One situation I was aware of hanging connection was when white has a high not low extension from the corner, because you want to jump further and ruin what the marked stone was building (maybe white blocked the wrong side when you 3-3ed), though this move is thin and rarely see:

Click Here To Show Diagram Code

[go]$$Bc
$$ . . , . . . . . , . . . |
$$ . . . . . . . . . O . . |
$$ . . . . . . . . . . . . |
$$ . . . . . . . . . . . . |
$$ . . . . . . . . . X . . |
$$ . . . . . . . . . 4 O . |
$$ . . , . . . O . O O X 2 |
$$ . . @ . . . . . . X 3 . |
$$ . . . . . . 5 . . . 1 . |
$$ . . . . . . . . . . . . |
$$ ------------------------+[/go]

Hah! Yes! I saw that one (still half way through, though). DDK moves have Pro dan strength!! I *knew* I was doing something right.

Resigning, likely.

Take care.

_________________
一碁一会