OpenAI's o3 system scores 85% on general intelligence test matching human performance

OpenAI's o3 system has achieved a significant milestone by scoring 85% on the ARC-AGI benchmark, up from the previous best of 55%. This score matches the average human performance on the test, indicating progress toward artificial general intelligence (AGI). The ARC-AGI benchmark measures an AI's ability to adapt to new situations with minimal examples. OpenAI's o3 model demonstrated high adaptability, suggesting it can generalize rules from limited data effectively, a key aspect of intelligence. While the results are promising, many details about the o3 system remain unclear. Further evaluations are needed to fully understand its capabilities and potential impact on the field of AI and AGI development.