1. What went wrong in the Messi ChatGPT ad?
A Messi ChatGPT ad published on Instagram on Wednesday, October 22, 2026, shows Lionel Messi receiving a chess lesson from OpenAI's chatbot on a board where each side's king and queen have been swapped. The assistant then recommends an opening that, from that faulty setup, ends in checkmate after four moves. The post drew about 1.9 million likes in a day.
The error is visible from the first frame for anyone who knows the game. In a standard setup, each queen starts on the square that matches her own color, with the king placed beside her. The advertisement reverses that arrangement for both White and Black, which means every move the chatbot suggests is being played from a position that could never occur in a real game.
Online chess commentators noticed almost immediately, and the reaction spread well beyond the chess community. The mistake is easy to describe and easy to check, which helped it travel quickly across social platforms and discussion forums.
2. How the opening collapses in four moves
In the clip, Messi explains that he asked ChatGPT for a chess opening that suited his style. The assistant answered with a line it called "GOAT Horns," a pun on the nickname that fans have used for Messi for years, where GOAT stands for "greatest of all time." The joke lands as marketing, but the chess behind it does not hold up.
The suggested sequence runs 1.e4 a6 2.d4 c6. On the swapped board shown in the ad, those moves open a diagonal for White's queen and allow a forced checkmate on the fourth move, a pace comparable to the well-known Scholar's mate, one of the fastest ways a game can end. On a correctly arranged board, the same four moves do not produce a forced mate.
The outcome carries a certain irony. Messi built his reputation by winning matches, not by losing them in a handful of turns, yet the lesson the advertisement presents would have a beginner on the wrong side of a very quick defeat if the board were set up as shown.
3. What is an opening, and why does this one matter?
An opening is the first handful of moves in a chess game, a phase in which players try to develop their pieces and claim space before the middle game begins. Beginners are usually taught a few sound principles and a small set of reliable lines, because early mistakes are the most common way to lose quickly.
The line in the ad has some real-world precedent. Magnus Carlsen, the top-ranked player in the world, has played this opening. He has also shown a taste for playful choices: in 2021 he used the Bongcloud, a joke opening that moves the king on the second move, against Hikaru Nakamura.
That context cuts both ways. A world champion can afford casual experiments because his skill covers the risk. A newcomer receiving a first lesson cannot, and the commentary around the ad pointed out that copying a top player's relaxed approach is not what a beginner's curriculum is made of.
4. Why are people reading this as more than a prop error?
The reaction goes beyond a prop department slipping up, because the product being promoted is presented as a knowledgeable tutor. OpenAI says more than 1.2 billion people use ChatGPT every week, and the advertisement features the second most-followed person on Instagram, whose account has more than 517 million followers.
A famous athlete being taught by an assistant that cannot arrange the board correctly invites an obvious question. If a system sold as a patient and capable teacher fails at the very first step of the lesson, readers may wonder how much confidence to place in its answers on subjects they cannot check as easily as a chessboard.
The scene is also a reminder of how visible errors can be in consumer advertising. A written answer containing a mistake might go unnoticed. A picture of a board with its pieces in the wrong places can be judged by anyone with a basic knowledge of the game in a matter of seconds.
5. Why does a setup mistake change the chess?
The way the pieces are arranged determines which lines are open and which squares each piece can reach. Swapping the king and queen on both sides changes the geometry of the opening moves, so a sequence that is harmless in a normal game can become a direct attack in the altered one. That is the reason the same four moves lead to a forced mate in one case and to nothing decisive in the other.
It also explains why chess players reacted so strongly. For them the board is not decoration. A single misplaced piece is the kind of detail that is seen before anything else, in the same way that a misspelled word on a printed sign draws the eye of a proofreader before the message registers.
For a casual viewer, the same scene may seem perfectly ordinary, and that gap is part of the story. The advertisement was aimed at a general audience of Messi's followers, yet the audience that responded most loudly was the one that knows the game best.
6. How did viewers react online?
Much of the conversation unfolded on Reddit. One user, writing as Cannonheim, asked how nobody involved noticed the problem before the video went live, pointing out that spotting a swapped king and queen takes very little time for anyone who has played.
Another commenter suggested the person responsible for sending the video out was either careless or far too clever, which in either case explained why people were discussing it. A third user proposed that a random intern might be to blame for the oversight.
Some users argued that this was not the first time. They said an earlier OpenAI advertisement had also misplaced a piece, with Black's queen shown on e8 rather than d8. One commenter used that example to mock a company valued at close to a trillion dollars for being unable to put a queen on the right square. These claims come from online discussion and are not independently documented here.
7. Where does the ad sit in OpenAI's recent marketing?
The campaign arrived about two weeks after OpenAI released GPT-6 with a feature called Intelligent UI. The feature lets ChatGPT respond with interactive elements such as charts, buttons and forms instead of plain text, and the company has been promoting it as a step toward a more practical assistant.
Messi's marketing relationship with OpenAI is more recent still. It began in June, when he posted an image created with ChatGPT Images ahead of the World Cup. The chess spot is part of the same effort to link a globally recognized athlete with an everyday use of the chatbot.
The official ChatGPT account also posted the advertisement on X on October 22. The post repeated the framing used in the video, in which Messi says he asked for an opening and the assistant came back with Goat Horns.
8. Is chess a known weak spot for ChatGPT?
Chess has been an awkward subject for general-purpose chatbots for some time. ChatGPT is not a dedicated engine in the way that Stockfish, a purpose-built chess program, is, and the two should not be compared as equals. Dedicated engines are designed to search positions exhaustively, while a chatbot generates text that resembles chess commentary.
The gap showed up in public in 2025. ChatGPT lost the final of a chatbot chess tournament hosted by International Master Levy Rozman, and during play it made illegal moves, including promoting pawns that did not exist. The advertisement therefore echoes a limitation that was already on record.
None of this means that language models cannot discuss chess usefully, since they can explain ideas, named openings and general strategy. It does suggest that tasks requiring an exact, rule-bound understanding of a board are where mistakes become easy to see and hard to defend.
9. What does the episode say about checking AI-generated material?
Advertising built around an AI model carries a particular burden, because the product is the subject of the message. Any visible flaw is taken as evidence about the product itself, whether or not the error came from the assistant or from the people who staged the scene.
The commentary around the clip largely treated it that way. Viewers did not separate the question of who set up the board from the question of whether the tutor should have caught it, and the second point is the one that gave the story its momentum.
For companies using AI in public-facing campaigns, the practical lesson is a familiar one: an ordinary human review of the final footage, by someone who understands the subject shown, is inexpensive compared with the attention that a visible error can attract.
10. What has OpenAI done since, and what comes next?
OpenAI has reportedly fixed the board in a separate version of the advertisement made for X, changing the placement of the pieces so that the setup is now correct. Correcting the arrangement does not resolve the other issue, however, since the opening itself remains a poor choice for a beginner like Messi.
The original Instagram video, which carries the swapped pieces, remains online. As long as it does, it is likely to keep circulating as an example of how a small production detail can overshadow a major celebrity campaign.
No further statement from the company was cited about the mistake. The next signal to watch is whether OpenAI replaces or removes the original post, and whether later advertisements in the Messi series pass a more careful review before they are published.