Update 9:26pm -- Apparently it was an April Fool's prank by ChessBase (link). They got me good, and it's times like these when I remember that it is important to say oops. I've left most of the response I wrote up intact below, to shame myself into being more vigilant in the future.
I didn't realize I'd have a chance to practice what I preach so soon. :) The edited original is below.
If you'll allow me to be pretty nerdy about chess for a second, this new chess result is absolutely unbelievable.
The King's Gambit is a very popular and very old opening move in chess. The "accepted" version of the opening (as opposed to the "declined" version) is when the Black player chooses to capture a White pawn that is seemingly free of any negative repercussion (hence the term 'gambit'; White hopes that the pawn merely looks like it is free but that really, by capturing it, Black opens up longer-term vulnerabilities).
Bobby Fischer famously claimed in the 1960s that he "proved" that the King's Gambit is strictly inferior for White -- that is, the supposedly "free" pawn offered to Black really is free; Black can capture it on the fourth move of the game and suffers no long-term weaknesses as a result of the capture.
This has been challenged by many grandmasters who actually use the King's Gambit and testify that White has a more or less equal chance of winning the game when Black captures the pawn. There are even dozens of books written on complicated 20-move-long (or longer) variations of this opening (I have a couple on the bookshelf next to me, actually).
Well, it turns out that Fischer was basically correct. Running one of today's best chess programs (Rybka) on an IBM computer cluster that is similar to the engine that powered Watson, chess researchers were able to exhaustively search the important positions that arises after the moves 1. e4 e5 2. f4 exf4 (the King's Gambit Accepted starting position).
The result: there is only one move that White can make that leads to a draw (Bishop to e2, 3. Be2). All other choices by White allow Black to literally win by force. [3]
To put this in perspective, this is a little bit like using a computer simulation to prove that kicking an onside kick to begin a football game always leads to a loss, unless you play absolutely perfectly in one particular way, and then you get a tie. Of course, football involves a lot more variables than chess; we're not yet able to describe a player's emotion levels, fatigue, motivation, etc., well enough to meaningfully simulate them on a computer... but the idea is basically the same.
