OpenAI and Google's DeepMind Achieve High Scores in International Math Olympiad: A Showcase of AI's Promising Future and Intensifying Corporate Rivalry

In an unprecedented display of artificial intelligence (AI) prowess, AI models from Google DeepMind and OpenAI took home gold medals at the daunting International Math Olympiad (IMO). This event, one of the most difficult high school-level competitions worldwide, saw these AI models outshine human competitors, adding another intriguing chapter to their inventors’ corporate rivalry.

Benchmarks like the IMO carry significant weight among AI researchers, many of whom hail from backgrounds in competitive math. Google and OpenAI’s stunning feats in the competition bring a new spin to AI usage in solving complex problems. Just last year, Google bagged a silver medal at the IMO using a system that required human assistance in translating problems into a machine-readable format. This year, however, both companies made headway by using ‘informal’ AI models that directly ingested the questions and provided proof-based answers in natural language.

Yet, amidst the celebrations, there were disagreements. Google raised questions about the premature announcement of OpenAI’s gold-medal win. Google DeepMind’s CEO and researchers struck out at OpenAI for declaring their success before the official IMO evaluation and immediately after the high school winners’ announcement. Amid the contention, it’s clear the AI competition is heating up, mirroring the competitive intrigue seen amongst the human mathematicians these AI systems so accurately emulate.