Researchers at Stanford University’s HAI Institute are calling for a redesign of tests to challenge artificial intelligence (AI) to become smarter and more advanced. The institute has found that AI abilities have reached a level comparable to an ordinary person in various tasks, but there is still room for improvement.
One area where AI has made significant progress is in image classification and text understanding in English. The level of AI has seen a dramatic increase in recent years, as evidenced by its improved performance in various tests, such as the Stanford MATH test, where Open AI’s AI system was able to solve a high percentage of math problems. However, challenges remain, like issues with popular large language models still prone to errors like “hallucinations.”
To reach their full potential, AI systems need to further improve in areas like mathematics and design. Stanford University is developing new tests that can accurately compare AI systems and identify areas where AI capabilities differ from human abilities. The introduction of new language models like GPT-5 is expected to shape future tests and further push the boundaries of AI capabilities.
Despite these advancements, researchers are optimistic about the potential for AI to excel in various tasks while acknowledging the importance of identifying and leveraging the unique strengths of human intelligence. As AI continues to evolve and improve, it will be crucial for researchers to develop new testing methods that can accurately assess its capabilities and limitations.
