Artificial intelligence (AI) systems are becoming so advanced that even the smartest humans are struggling to create tests that these systems can't pass.
Standardized benchmark tests that were originally used to measure AI progress are now being aced by AI systems, prompting the need for harder tests.
A new evaluation called 'Humanity's Last Exam' has been introduced as the hardest test ever administered to AI systems.
The test aims to determine if AI systems have become too intelligent for humans to measure accurately.