AI Open AI and Elon Musk's Grok faced off in a chess tournament

OpenAI's artificial intelligence o3 achieved a convincing victory over Elon Musk's xAI-developed Grok 4 model in the final of the Kaggle AI Exhibition chess tournament. According to The Independent, the final match score was 4:0 in favor of o3, adding extra tension to the long-standing rivalry between OpenAI co-founder Sam Altman and Musk, who previously attempted to acquire OpenAI after his departure.

In the decisive match, the o3 model demonstrated flawless play, winning all four games. Both systems reached the final after confidently defeating competitors in previous stages of the competition. In response to his model's defeat, Elon Musk stated that Grok's gaming ability is a "side effect" and was not a priority during its development.

The Kaggle AI Exhibition tournament brought together several leading AI models from companies such as Anthropic, DeepSeek, Google, and Moonshot AI. Although computers have been able to defeat the world's best chess players since 1997, when IBM's Deep Blue supercomputer beat Garry Kasparov (recognized as a foreign agent in Russia), and later Google DeepMind created self-learning systems for chess and go, the current competition became the first major test for large language models (LLMs) like ChatGPT in strategic games.

Mats Andre Christiansen, CEO and co-founder of Take Take Take, emphasized that unlike the Kasparov versus Deep Blue match in the 1990s, the key value of this tournament lies not so much in the spectacle, but in exploring the thought processes and decision-making of modern AI models.

Interestingly, in July, the current world chess champion Magnus Carlsen himself demonstrated superiority over ChatGPT in an online match, without losing a single piece. Commenting on the outcome of the final between Grok and o3, Carlsen estimated Grok's chess rating at about 800 points, and o3's at 1200. For comparison, his own peak rating is 2882 points.