Researchers at DeepMind have published a study on how AI systems perceive and organize visual information. Unlike humans, AI systems tend to focus on specific objects or features, rather than understanding the relationships between them. This difference in perception can lead to biases in AI decision-making. The study aims to understand these differences and develop more human-like vision capabilities for AI. --- Why it matters: This research matters because it could improve the accuracy and fairness of AI systems that rely on visual information, such as self-driving cars or medical image analysis tools. Source: https://deepmind.google/blog/teaching-ai-to-see-the-world-more-like-we-do/