Zoom's AI Scores Highest Ever on Humanity's Last Exam, But at What Cost?

The company Zoom Video Communications announced last week that it had achieved the highest score ever recorded on the Humanity's Last Exam, a benchmark designed by subject-matter experts worldwide to stump even the most advanced AI models. According to the company, its AI system scored 48.1 percent on the exam, edging out Google's Gemini 3 Pro, which held the previous record at 45.8 percent. Zoom's chief technology officer, Xuedong Huang, wrote in a blog post that the company had achieved a new state-of-the-art result on the challenging Humanity's Last Exam full-set benchmark, scoring 48.1, which represents a substantial 2.3 improvement over the previous state-of-the-art result. Huang attributed the success to the company's focus on developing large language models and its commitment to advancing the field of artificial intelligence. The announcement has raised a provocative question that has consumed AI watchers for days: How did a video conferencing company with no public history of training large language models suddenly vault past Google, OpenAI, and other leading AI research organizations? Critics have pointed out that Zoom's achievement may have come at the expense of copying off its neighbors, with some suggesting that the company may have leveraged the work of other researchers or organizations to achieve its success. The Humanity's Last Exam is a benchmark designed to test the ability of AI systems to reason, understand, and apply human knowledge in a variety of domains. The exam consists of a series of questions that require AI systems to demonstrate their ability to think critically and make informed decisions. The exam is considered one of the most challenging AI benchmarks in existence, and achieving a high score on it is seen as a major milestone in the development of artificial intelligence. Experts in the field of artificial intelligence have expressed a range of opinions on Zoom's achievement. Some have praised the company for its innovative approach to developing large language models, while others have raised concerns about the potential implications of its success. "This achievement is a testament to the power of collaboration and innovation in the field of artificial intelligence," said Dr. Fei-Fei Li, director of the Stanford Artificial Intelligence Lab. "However, it also raises important questions about the ethics of AI development and the need for greater transparency and accountability in the field." The success of Zoom's AI system has significant implications for society, particularly in the areas of education, healthcare, and customer service. As AI systems become increasingly capable of performing complex tasks, they are likely to play a major role in shaping the future of these industries. However, the use of AI also raises important questions about the potential risks and consequences of relying on machines to make decisions and provide services. As the AI community continues to grapple with the implications of Zoom's achievement, the company is already looking to the future. In a statement, Zoom's CEO, Eric Yuan, said that the company is committed to continuing to advance the field of artificial intelligence and to exploring new applications for its technology. "We believe that AI has the potential to transform the way we live and work, and we are excited to be at the forefront of this revolution," Yuan said.