OpenAI's new AI model o3 scores 85% on general intelligence test matching human performance

OpenAI's new AI model, o3, has achieved a score of 85% on the ARC-AGI benchmark, surpassing the previous best of 55% and matching the average human score. This marks a significant advancement in AI's ability to demonstrate general intelligence. The ARC-AGI test measures how well an AI can adapt to new situations using limited examples. Unlike previous models, o3 shows a high capacity for generalization, suggesting it can learn and apply rules from fewer data points. Details about o3's inner workings remain unclear, and further evaluation is needed to understand its true capabilities. If it proves to be as adaptable as humans, it could lead to major changes in AI development and application.