Unfortunately for ChatGPT, the experiment did not yield favorable results. Caruso detailed the match in a LinkedIn post, revealing that the chatbot struggled significantly during gameplay. He noted that ChatGPT frequently confused rooks for bishops and failed to recognize critical tactics such as pawn forks. The AI also lost track of the board, initially attributing its difficulties to the Atari icons being too abstract, yet even after switching to standard chess notation, its performance did not improve.

Caruso humorously remarked, “It made enough blunders to get laughed out of a 3rd-grade chess club,” highlighting just how far from a competent chess player the chatbot proved to be. He further commented that ChatGPT got "absolutely wrecked at the beginner level," with the 90-minute match continuing to deteriorate as the AI repeatedly requested to restart the game.

The historical context of computer chess matches adds another layer to this narrative. For decades, the ability of computers to outplay human opponents in chess has been a barometer of artificial intelligence capability. A landmark moment occurred in 1997 when IBM's Deep Blue triumphed over world chess champion Garry Kasparov in a highly publicized series of matches, cementing the notion that machines could surpass human intellect in this strategic game.

Despite the disheartening results of Caruso's experiment, it is crucial to note that this does not render ChatGPT ineffective for chess-related conversations. Its design as a language model, rather than a sophisticated chess-playing supercomputer, inherently limits its effectiveness in this domain. In fact, a few years ago, a developer created a plugin known as ChessGPT for ChatGPT, specifically to enhance its chess-related interactions. This suggests that while playing chess might not be ChatGPT's strong suit, discussing the game is likely more fruitful.

As of now, OpenAI has not responded to requests for comments regarding this unusual experiment. This lack of immediate communication from the company leaves room for speculation about the implications of such findings for ChatGPT's development and its future applications in strategic games.

It is also worth noting that Ziff Davis, the parent company of CNET, has recently filed a lawsuit against OpenAI, claiming infringement of its copyrights during the training and operation of its AI systems. This ongoing legal issue adds yet another layer of complexity to the unfolding narrative of AI technologies.