Jerome Gambit: Google Gemini, Again

 

It's time to check in with an artificial intelligence (AI) entity - this time, Google's Gemini - and see what it has learned about the Jerome Gambit (1.e4 e5 2.Nf3 Nc6 3.Bc4 Bc5 4.Bxf7+) since we last visited it (see"Jerome Gambit: Gemini's Turn")

Tell me about the Jerome Gambit chess opening and show me a game with it

The Jerome Gambit is a highly aggressive and controversial chess opening for White, usually arising from the Italian Game (Giuoco Piano). It is widely considered "unsound" at high levels of play because White sacrifices two minor pieces for a king-side attack, but it is a favorite for blitz games or "fun" sessions due to the immediate chaos it creates.

The Opening Moves

The gambit begins with the following moves:

1. e4 e5

2. Nf3 Nc6

3. Bc4 Bc5 (The standard Italian Game)

4. Bxf7+!? (The Jerome Gambit begins)

White sacrifices the Bishop on f7 to force the Black King out of safety. After 4... Kxf7, White follows up with 5. Nxe5+, sacrificing a Knight as well. If Black accepts with 5... Nxe5, White plays 6. Qh5+ [01:28], forking the King and the Knight (or Bishop).

Key Concepts

• Psychological Pressure: You are betting that your opponent will panic while trying to defend their exposed King in the center of the board.

• Targeting the King: White's goal is to keep the Black King dancing. Common continuations for Black like 6... Kf6 or 6... Ke7 often lead to immediate trouble after 7. Qf5+ [01:50].

• Risk: If Black knows the theory or defends calmly (often by returning some material to stabilize), White will simply be down a massive amount of material with no compensation.

---

A Sample Miniature Game

The following is a classic example of how quickly the Jerome Gambit can lead to a checkmate if Black isn't careful (often seen in variations against the "Blackburne Shilling" trap) [14:13]:

Move	White	Black	Notes
1	e4	e5
2	Nf3	Nc6
3	Bc4	Nd4?	Black tries the Blackburne Shilling Trap.
4	Bxf7+!	Kxf7	White ignores the trap and enters Jerome-style play [14:33].
5	Nxe5+	Ke6	Black brings the King forward to protect the Knight.
6	Qh5	g6?	Black tries to kick the Queen, but it's a blunder [15:10].
7	Nxg6	Nf6
8	Qe5+	Kf7
9	Nxh8+	Kg8
10	Qxd4	...	White has recovered material and left the Black King stranded.

Summary of the Gambit:

• Best for: Blitz, bullet, and casual "coffee house" chess.

• Weakness: Objectively losing against accurate defense [00:39].

• Goal: Checkmate before the material deficit matters.

For a full breakdown of the lines and how to handle different Black responses, you can watch the detailed guide here:

Chess Openings - Jerome Gambit

This is an improvement.

(I especially liked the comment about Psychological Pressure, "You are betting that your opponent will panic" - reminds me of an earlier post, "Jerome Gambit: Fun": The first to panic loses the game)

Blackburne Shilling Jerome Gambit: Surprise, Surprise

When playing the Jerome Gambit, or any of its relatives, as White, it is not only helpful to be aware of the tactical tricks and traps at your disposal - it is necessary to be aware of similar surprises placed in front of you by your opponent.

EDOrel31 - shns12

15 0 rapid, lichess.org, 2026

1.e4 e5 2.Nf3 Nc6 3.Bc4 Nd4 

The Blackburne Shilling Gambit. This sometimes comes as a surprise for White.

4.Bxf7+

The Blackburne Shilling Jeorme Gambit. This often comes as a surprise for Black.

EDOrel31 has played it in 31 games, scoring 63%. Still, it is necessary to be careful.

4...Ke7 

Capturing the piece leads to an advantage (declining it leads to a disadvantage), but Black's psychology is If you want me to take the Bishop - I won't.

5.Nxe5 

He has also played 5.Bxg8 Rxg8 6.Nxe5 Ke8 7.Qh5+ Black resigned, EDOrel31 - BeauneCrusher, 15 0 rapid, lichess.org, 2023; and 

5.Nxd4 exd4 6.Bxg8 Rxg8 7.O-O Ke8 8.Qh5+ g6 9.Qxh7 Rg7 10.Qh3 d5 11.Qf3 Rf7 12.Qd3 dxe4 13.Qxe4+ Be7 14.Qxg6 Bf5 15.Qg8+ Rf8 16.Qc4 Qd7 17.d3? (17.Re1) 17...Be6 18.Bh6 Rg8 19.Na3 Bxc4 20.Nxc4 Qh3 21.Bg5 Rxg5 22.g3 Rh5 White resigned, EDOrel31 - StankoM2021, 15 0 rapid, lichess.org, 2024. 

5...d6 6.Bxg8 dxe5 7.Bd5 c6 8.Bb3 

White is ahead two pawns, with a safer King.

8...g6 

Black has a nefarious trick in mind.

9.O-O 

This move is fine, but White needs to watch out.

9...Bh6 10.d3

The trick is not deep, but the first player misses it. Better is 10.Re1. 

10...Bxc1 11.Qxc1 Ne2+ 

Ouch!

12.Kh1 Nxc1 

Now the game plays out.

13.Rxc1 Be6 14.Bxe6 Kxe6 15.Nc3 Qb6 16.Na4 Qxf2 17.Nc5+ 

A final slip.

17...Qxc5 White resigns

Blackburne Shilling Jerome Gambit: Open Lines

In the following game, White opens lines against the enemy King and Queen, and then attacks along those lines. Black's King's mis-steps increased his danger.

PedrosF1 - reykarl

10 0 rapid, lichess.org, 2025

1.e4 e5 2.Nf3 Nc6 3.Bc4 Nd4 

Blackburne Shilling Gambit

4.Bxf7+ 

Blackburne Shilling Jerome Gambit

4...Kxf7 5.Nxe5+ Kf6 

Hoping to evict the enemy Knight.

6.Nf3 Nxf3+ 7.Qxf3+ Ke7 

Black has a piece for two pawns, but his King is poorly placed.

It is interesting that Stockfish 16.1 has White meeting 7...Ke6 with  8.Qf5+ Ke7 9.Qg5+ Ke8 10.Qh5+ Ke7 11.Qg5+ with a draw in the offing. 

8.O-O Ke8 

Instead of this retreat, 8...Qe8 9.d4 Qf7 is equal. 

9.d4 d6 

10.c4 Be7 11.Nc3 Nf6 12.Bg5 h6 13.Bxf6 Bxf6 14.e5 

14...dxe5 15.dxe5 Be7 16.Rad1 Rf8 17.Qh5+ Rf7 18.Rxd8+ 

Black resigned

Blackburne Shilling Jerome Gambit: Faster Than A Speeding...

The following game shows that when fast time controls are in play, it is worth playing aggressively and taking risks - within reason.

Chicco79 - jim714

5 0 blitz, Eastern Blitz Arena, lichess.org, 2025

1.e4 e5 2.Nf3 Nc6 3.Bc4 Nd4 

The Blackburne Shilling Gambit

4.Bxf7+ 

The Blackburne Shilling Jerome Gambit

4...Kxf7 5.Nxe5+ Kf6 6.O-O 

Offering the Knight on e5. It can be taken - with care.

6...Kxe5 7.c3 Ne6 8.Qh5+ 

This is a 5-minute blitz game, and such moves are often tried - and are often successful.

Also fun, for the attacker, is 8.d4 Kxe4? 9.Qf3 checkmate.

A Jerome-ish, pawnful alternative, is 8.d4 Kf6 (best) 9.f4 h5 10.f5 Ng5 11.e5 Kf7 althought the line is, admittedly, better for Black (Stockfish 16.1, 30 ply, less than a pawn).

8...Qg5 

You can move your Queen, I can move my Queen.

Ill-advised, but, again, this is blitz.

9.f4+ 

Uh oh.

There is also 9.d4+ Kf6 10.Bxg5+ Nxg5 11.e5+ Ke7 12.Qxg5+. 

9...Nxf4

Yes, and No. In this kind of position, it hardly matters.

10.Qxg5+ Black resigned

Blackburne Shilling Jerome Gambit (Again?)

It was only after playing the following short game that I realized that I had explored the line before - the most recent time being 11 years previously.

perrypawnpusher - zgajek

"Not only the Italians plays the Italian", Chess.com, 2025

1.e4 e5 2.Nf3 Nc6 3.Bc4 Nd4 

The Blackburne Shilling Gambit.

4.Bxf7+ 

The Blackburne Shilling Jerome Gambit.

4...Kxf7 5.Nxe5+ Ke6 

Or 5...Ke7 as in perrypawnpusher - michon, blitz, FICS, 2013 (1-0, 18) 

6.c3 

Or 6.Qh5 as in perrypawnpusher - zmagic, blitz, FICS, 2007 (1/2 - 1/2, 51) 

6...Nc6 

Or 6...Kxe5 as in 

perrypawnpusher - PunisherABD, blitz, FICS, 2009 (1-0, 27);

perrypawnpusher - PunisherABD, blitz, FICS, 2009 (1-0, 14); 

perrypawnpusher - MNUNP, blitz, FICS, 2009 (1-0, 18);

perrypawnpusher - AdamRou, blitz, FICS, 2010 (1-0, 28); 

perrypawnpusher - RVLY, blitz, FICS, 2010 (1-0, 26);

perrypawnpusher - michon, blitz, FICS, 2014 (1-0, 23); 

Or 6...Qg5 as in perrypawnpusher - TheProducer, blitz, FICS, 2009 (1-0, 66); 

Or 6...d6 as in perrypawnpusher - ozypawnstar, blitz, FICS, 2010 (1-0, 29). 

7.d4 Nce7 8.d5+ 

Or 8.Qg4+ as in perrypawnpusher - Abatwa, blitz, FICS, 2010 (1-0, 8)

8...Kxe5 

Checkmate is inevitable:

8...Nxd5 9.Qxd5+ Kf6 10.Qf7+ Kxe5 11.Bf4+ Kxe4 12.Nd2+ Kd3 13.Qf5# or

8...Kd6 9.Nc4+ Kc5 10.b4+ Kb5 11.Nba3+ Ka6 12.Qa4# or

8...Kf6 9.Qf3+ Nf5 10.Bg5+ Kxg5 11.Qxf5+ Kh6 12.Nf7#. 

9.Qd4+ Kd6 10.Bf4 checkmate

BSJG: Quickly Turn the Game Into A Mess

                                               

The Blackburne Shilling Jerome Gambit (1.e4 e5 2.Nf3 Nc6 3.Bc4 Nd4 4.Bxf7+) can quickly turn the game into a mess - one that can prove deadly to the defender's King.

perrypawnpusher - zgajek

Not only the Italians play the Italian, Chess.com, 2025

1.e4 e5 2.Nf3 Nc6 3.Bc4 Nd4 4.Bxf7+ 

The Blackburne Shilling Jerome Gambit.

4...Kxf7 

Declining the piece with 4...Ke7 is insufficient, 5.Bxg8 Rxg8 (5...d6 6.d3 Be6 7.Bg5+ Kd7 8.Bxd8 Black resigned, perrypawnpusher - michon, blitz, FICS, 2013) 6.Nxe5 Kd6 7.Nf3? (7.Nc7+) c5 8.Nxd4 cxd4 as in, perrypawnpusher - michon, blitz, FICS, 2013 (1-0, 41).

5.Nxe5+ Ke6 

Not quite a strong is 5...Ke7 6.c3 d6 7.cxd4 dxe5 8.dxe5 as in perrypawnpusher - michon, blitz, FICS, 2013 (1-0, 18)

6.c3 

I have been 8 - 0 with this move. 

6...Nc6 

Black's move is Stockfish 16.1' s choice (30 ply), resulting with an advantage of about 3/4 pawn for Black. 

Grabbing material with 6...Kxe5 7.cxd4+ is a critical line, where Black does best to tip toe away with 7...Ke6 

7.d4 Nce7 

Creative.

Black's safest path was 7...Nxe5 8. dxe5 Kf7 and the King retreats toward safety. 

7...Nf6 is not sufficient, followed by 8.O-O Ke7 9.Nxc6+ dxc6 10.e5 Qd5 11.exf6+ gxf6. 

8.d5+ 

Less powerful, but still effective is 8.Qg4+ as in perrypawnpusher - Abatwa, blitz, FICS, 2010 (1-0, 8)

8...Kxe5 

Or 8...Kd6 9.Nc4+ Kc5 10.b4+ Kb5 11.Nba3+ Ka6 12.Qa4 checkmate 

Or 8...Nxd5 9.Qxd5+ Kf6 10.Qf7+ Kxe5 11.Bf4+ Kxe4 12.Nd2+ Kd3 13.Qf5 checkmate 

9.Qd4+ Kd6 10.Bf4 checkmate

BSJG: Choices, Choices, Choices

The following game shows many choices, some taken, some not - some successful, some not.

It has a real Jerome-ish feel to it.

maestro_rabbittry01 - Florence_Madarang

2 5 bullet, training Arena, lichess.org, 2023

1.e4 e5 2.Bc4 Nc6 3.Nf3 Nd4 

The Blackburne Shilling Gambit.

According to Wikipedia:

The first known mention of this line was by Steinitz, who noted it in 1895 in the Addenda to his Modern Chess Instructor, Part II. The earliest game with the opening on chessgames.com is Dunlop - Hicks, New Zealand Championship 1911. [1.e4 e5 2.Nf3 Nc6 3.Bc4 Nd4 4.O-O Nxf3+ 5.Qxf3 Qf6 6.Qg3 Bc5 7.Nc3 c6 8.Kh1 b5 9.Bb3 d6 10.f4 h5 11.d3 Be6 12.f5 h4 13.Qg4 Bxb3 14.Bg5 h6 15.Qxh4 O-O-O 16.Bxf6 gxf6 17.axb3 Nxf5 18.Qxh8 Ng3+ 19.hxg3 Rxh8#].

4.Bxf7+

The Blackburne Shilling Jerome Gambit.

Stockfish 17.1 (at 34 ply), sees the position as about 1 pawn better for Black. 

The Database has 25 games by maestro_rabbittry01 with this line. He scores 64%.

The earliest game with this line in The Database is Nater, Carl - Rogers, Ian, Begonia op 09th, Ballarat, 1975 (0-1, 46 ).

Years ago, in the post "A GM Faces the BSJG: Not Quite (Part 4)" I had written

Regarding the early Blackburne Shilling Jerome Gambit game that we have been looking at ("A GM Faces the BSJG: Not Quite, Parts 1, 2 and 3"), Nater, Carl - Rogers, Ian, Begonia op 09th, Ballarat, 1975 (0-1, 46 ), I was able to contact GM Rogers, who, in turn, was able to contact Mr. Nater.

Not surprisingly, GM Rogers said that he was shocked to see his 1.e4 e5 2.Nf3, Nc6 3.Bc4 Nd4 met by 4.Bxf7+!?. At first, he expected simply be able to refute the move, as he not had it played against him previously - and actually had not even seen it mentioned before. He settled himself down and outplayed his opponent, for the full point.

"I gave up 3...Nd4 soon after that game for multiple reasons, not least that it was a bad move!" was GM Rogers' assessment.

Mr. Nater, rather than claiming to be one of the world's foremost experts on playing the Blackburne Shilling Jerome Gambit over-the-board, modestly reported that "my openings at my prime [around ’75 probably] may have dived as deep as 4/5 moves before descending into chaos ... nowadays worse still." Not surprisingly, he did not have access to game score sheets from 45 years ago, so he could not say if he had repeated (or was repeating) his moment of chess opening inspiration.

"But there doesn't seem to be too much wrong with 4.Bxf7, more wrong with 3...Nd4" was his assessment.

4...Kxf7 

From Black's perspective, the computer evaluates 4...Kxf7 as about two pawns better than 4...Ke7, just in case you were wondering.

This is reflected in The Database, where 4...Kxf7, with 7,155 games scores  56% for White; while 4...Ke7, with 919 games, scores 71% for White.

5.Nxe5+ Ke8 

6.Qh5+ g6 

Not 6...Ke7 7.Qf7+ Kd6 8.Nc4+ Kc5 9.Qd5+ Kb4 10.a3+ Ka4 11.Qa5 checkmate

7.Nxg6

The Database has 1,820 games with this move, with White scoring 60%.

The computer is more skeptical, evaluating the position about a pawn better for Black.

7...Nxc2+ 

It would be a lot to expect Black to be familiar with the post "BSJG: From Defeat to Full Point" and the much earlier "Warning: a1 & h8", which argue that giving up the Rook first with 7...hxg6 was the better choice, i.e. 8.Qxh8 Nxc2+ 9.Kd1 Nxa1 10.Qxg8 Qg5 when Black would be better, according to Stockfish 17.1. 

maestro_rabbittry01 had actually faced 7...hxg6 ten days before the current game, with play continuing  8.Qxh8 Qg5? 9.O-O (better 9.Qxd4) 9...Nf3+ (better 9...Ne2+ 10.Kh1 Kf7) 10.Kh1 Nh4 11.g3 Nf3 12.d4 (maybe 12.Kg2) 12...Qg4 13.Nd2 Nxd2 14.Bxd2 Ne7 15.Bh6 Qf3+ 16.Kg1 d6 17.Bxf8 Qxf8 18.Qh4 g5 19.Qxg5 Bh3 $6 20.Qh5+ Kd8 21.Qxh3 Nc6 22.c3 Ke7 23.f4 Qf6 24.e5 Qg7 25.Rae1 Rh8 26.exd6+ Kxd6 27.Qg2 Ne7 28.f5 Nd5 29.Re6+ Kd7 30.Qxd5+ Black resigned, maestro_rabbittry01 - mcdhardal, lichess.org, 2023 

Later on, he saw 7...Nf6 8.Qe5+ Qe7? (better 8...Ne6 9.Nxh8 Bg7 [9...Qe7 10.d4 Ng4 11.Qh5+ Kd8 12.Qxg4 Nxd4 13.Na3 d5 14.Qf4 dxe4 15.Nf7+ Ke8 16.Ne5 Nf5 17.O-O Be6 18.Qxe4 Rd8 19.Nb5 Bd5 20.Qe1 Bxg2 21.Nxc7+ Qxc7 22.Nc6+ Be7 23.Nxd8 Qxd8 24.Kxg2 Qd5+ 25.f3 Black resigned, maestro_rabbittry01 - ako_ang_mahina_dito, lichess.org, 2023] 10.O-O Bxh8 11.d4 d6 12.Qg3 Qe7 [12...Nxe4] 13.e5 [Qd3] Ng8 [13...Nxd4] 14.Qxg8+ Kd7 15.Qxh8 b5 16.Nc3 Bb7 17.Qf6 Qe8 18.Nxb5 Qh5 19.exd6 cxd6 20.Be3 [20.Re1] Rf8 21.Qxf8? [21.Qh6] Nxf8 22.c4 Qg4 [22...Bxg2] 23.g3? [23.f3] 23...Qh3 [instead, 23...Bf3 mates] 24.f3 Ng6 25.d5 h5 26.Nxa7?! h4 27.Nc6? [27.Rf2] 27...hxg3 28.hxg3 Qxg3+ 29.Kh1 Nh4 30.Rf2 Qh3+ 31.Rh2? [31.Kg1] 31...Qxf3+ 32.Kg1 Qg3+ 33.Kf1 Qxh2 White resigned, maestro_rabbittry01 - MarcAndreBullirseTCU, lichess.org, 2023) ) 9.Nxe7 Black resigned, maestro_rabbittry01 - Olamygoodfriend, lichess.org, 2024) 

Are you still with me? The game is now even.

8.Kd1 

The Database has 223 games with this position. Black scores 32%.

Time for Black to grab the draw.

8...Nxa1 

Quite likely, the 2-minute time control affected this choice. Instead, there was 8...hxg6 9.Qxg6+ Ke7 10.Qg5+ Ke8 [10...Nf6 is interesting, but after 11.e5 Kf7 12.Kxc2 White will still get his material back, and the game looks even] 11.Qg6+ draw 

9.Nxh8+ Ke7 10.d4 

Or 10.Qf7+ Black resigned, maestro_rabbittry01 - mcdhardal, lichess.org, 2023 

There is also 10.Qe5#. 

White sees another way.

10...Qe8 11.Bg5+ Nf6 12.Bxf6+ Black resigned

BSJG: Knife's Edge

The following game shows the defender balancing on a knife's edge - and then falling off.

Uxx - Butterfli6282

30 0 lichess.org, 2024

1.e4 e5 2.Nf3 Nc6 3.Bc4 Nd4 

Blackburne Shilling Gambit.

4.Bxf7+

Blackburne Shilling Jerome Gambit. 

4...Kxf7 5.Nxe5+ Ke8 6.Qh5+ g6 

Not 6...Kf7 7.Qf7+ Kd6 8.Nc4+ Kc5 9.Qd5+ Kb4 10.a3+ Ka4 11.Nc3 checkmate

7.Nxg6 Nxc2+ 

Not wanting to give up a Rook first, with 7...hxg6 8.Qxh8 Nxc2+ 9.Kd1 Nxa1 10.Qxg8 Qg5, but that was the way to advantage.

8.Kd1 Nxa1 

The game would devolve into a draw after 8...hxg6 9.Qxg6+ Ke7 10.Qg5+ Ke8 11.Qg6+ Ke7 12.Qg5+ draw, but, again, that was for choice. 

The text makes things worse.

9.Nxh8+ Ke7 10.Qe5 checkmate

Blackburne Shilling Jerome Gambit: A Win Is A Win

It can be fun to look at a chess position and figure out the sharpest line of play. 

It is a lot easier, when the position appears in a blog post, like this one.

It is harder when the position appears on the board in a game that you are playing.

Even moreso, when the time control is 1-minute.

All that said, White survives handily in the following Jerome Gambit game.

DutchLiLi - MrJBlake

1 0 bullet, lichess.org, 2025

1.e4 e5 2.Nf3 Nc6 3.Bc4 Nd4 

The Blackburne Shilling Gambit.

4.Bxf7+ 

The Blackburne Shilling Jerome Gambit.

4...Kxf7 5.Nxe5+ Ke8 6.Qh5+ g6 7.Nxg6 Nxc2+ 

Okay...

White's Knight is threatening a Black Rook on one side of the board, and Black's Knight is threatening a White Rook on the other side of the board.

The Database has 311 games with this position. White wins 65%, even though Stockfish 16.1 evaluates the game as completely even: 0.00.

We will soon see why.

8.Kd1 Nxa1 

The logical followup, although the computer suggests that allowing White to give perpetual check is better, e.g. 8...hxg6 9.Qxg6+ Ke7 10.Qg5+ Ke8 11.Qg6+ etc, draw. 

I am not sure that either player would be happy with that outcome, but with the way things go, perhaps Black should have tried 8...hxg6.

9.Nxh8+ Ke7 

10.Qf7+ 

White has a plan to round up the enemy King. That is most important.

He did have the alternative, 10.Qe5#, but that is mostly a footnote in this chess essay. 

10...Kd6 11.Qf4+ 

With this move White still has the advantage, but he will now have to work a bit more. Instead, there was another footnote, 11.Qd5+ Ke7 12.Qe5#, which I am sure he saw just as soon as he could catch his breath.  

11... Kc6 12.Nf7 

12...Qe7 

12...Qf6 was a bit safer, with the chance to exchange Queens. 

13.Ne5+ Kb6 14.Nc4+ 

This works, although he might have added a piece to the King hunt with 14.Nc3. 

14...Ka6 15.Nba3 

The Knight keeps an eye on the c2 square, keeping its counterpart confined. Instead, 15.Nc3 was the way to keep an advantage, although White's Queen needs to be re-positioned, too.

15...Nf6 

Black is organizing his defense and thinking of eventual counter-attack.

16.d3 d6 

One or two steps? That is always the question for Black's d-pawn. In this case, the answer is two.

17.Qg5 

White wants his Queen to go Queenside.

17...h6 

Must be the clock. With 17...c5 18.Qd2 b6, Black would have limited the enemy Queen.

18.Qa5 checkmate