In a charming blend of man vs. machine and chess comedy, 19-year-old Grandmaster Rameshbabu Praggnanandhaa took on ChatGPT in a blindfold chess game, and won. The twist? He couldn’t see the board. And the funnier twist? His silicon opponent occasionally couldn’t see reality.
Hosted by Chess.com in a lighthearted challenge titled “Can Grandmaster Praggnanandhaa Beat ChatGPT BLINDFOLDED?!”, the game was more about entertainment than Elo. But that didn’t stop Pragg from showing just how sharp his mind is, even when the pieces are only in his head.
A Blindfold and a Bot
With nothing but his memory and mastery, Praggnanandhaa Rameshbabu sat down, literally spinning around in a swivel chair to “shake hands” with his invisible, AI opponent. From move one, he was focused:
“E4,” he declared confidently, launching into the game with the classic king’s pawn opening.
ChatGPT, armed with opening knowledge and verbose commentary, replied with 1…c5—the Sicilian Defense—and even explained it aloud. So far, so good.
Traps, Themes, and Tactical Trolling
The early middlegame saw Pragg maneuver into a familiar Alapin setup, subtly testing whether ChatGPT might fall for a known trap. It didn’t—but the real blunders were still to come.
Pragg played sharply:
- He built up pressure with quiet developing moves.
- He sacrificed material to open the position.
- He even pinned ChatGPT’s pieces—and the bot noticed! (sometimes.)
But as the game progressed, the AI started to… glitch. After a tactical flurry, Pragg cleanly won a queen. But soon after, the same queen mysteriously returned to the board.
“Do you want the computer to bring the queen back on the board?” the host asked.
“If it’s not going to bring more pieces, then I’m fine,” Pragg chuckled.
With grace, he allowed ChatGPT one illegal resurrection.
The Endgame… and the End of the AI
After a few more moves, it was clear the human was cruising. Pragg activated his king, coordinated his knights, and confidently marched into a favorable endgame.
The final moment? With a playful tone, the grandmaster delivered the killing blow:
“Yeah, I think I’ll just take the king.”
Knight takes king. Game over. The crowd (and the internet) erupted.
Verdict: A Win for the Humans (This Time)
After turning around to face the board he’d never seen, Praggnanandhaa shared his thoughts:
“Yeah, it was playing okay till some point… then it started blendering. It’ll improve. We’ll see.”
His verdict was fair. ChatGPT impressed with decent opening play and some awareness of positional ideas—but collapsed under pressure. Then again, it’s hard to win when your opponent is a world-class grandmaster playing blindfolded.
Takeaway
This wasn’t about proving if AI can beat humans in chess (we already know supercomputers can). This was about joy, creativity, and watching one of the brightest young stars in chess have fun while flexing his brilliance.
Final result: Praggnanandhaa 1 – ChatGPT 0 (with one queen resurrection).
The future of AI in chess? Promising. But for now, the humans still have the upper hand—and the sense of humor.

I’m Xuan Binh, the founder of Attacking Chess, and the Deputy Head of Communications at the Vietnam Chess Federation (VCF). My chess.com and lichess rating is above 2300, in both blitz and bullet. Follow me on Twitter (X).
1 thought on “Blindfolded Brilliance: Praggnanandhaa Outsmarts ChatGPT in a Hilarious Showdown”
Comments are closed.