A new artificial intelligence (AI) model has just achieved human-level results on a test designed to measure “general intelligence”. On December 20, OpenAI’s o3 system scored 85% on the ARC-AGI ...

Many people object to intelligence tests. Some say IQ test scores are too often abused. They says it’s unfair that when children “fail” these tests it can mean they receive a worse secondary education ...

OpenAI’s o3 system scored 85% on the ARC-AGI benchmark, well above the previous AI best score of 55% and on par with the average human score. Reading time 4 minutes A new artificial intelligence (AI) ...

Many online IQ tests come across as little more than digital parlor tricks. Users often end up with vague, unverified results. In other cases, the platforms tell people what they want to hear. Either ...

American IQ test scores have dropped during a recent 13-year period, a remarkable finding that runs counter to the well-established trend of increasing IQ scores throughout much of the 20th century.

This 1937 version of the Stanford-Binet Intelligence Scales was produced in the United States. It used objects because this IQ test was aimed at children. Credit: SSPL/ Getty Images The Hechinger ...