AI Sees Your Location, But With A Bias Toward The Wealthy World
This work addresses fairness and privacy issues in AI for users worldwide, particularly highlighting biases that disadvantage less developed regions, and is incremental as it systematically evaluates existing VLMs on a new benchmark.
The paper tackles the problem of regional biases in Visual-Language Models (VLMs) for geographic recognition from images, finding that these models achieve up to 53.8% accuracy in city prediction but perform significantly worse in less developed (-12.5%) and sparsely populated (-17.0%) areas.
Visual-Language Models (VLMs) have shown remarkable performance across various tasks, particularly in recognizing geographic information from images. However, VLMs still show regional biases in this task. To systematically evaluate these issues, we introduce a benchmark consisting of 1,200 images paired with detailed geographic metadata. Evaluating four VLMs, we find that while these models demonstrate the ability to recognize geographic information from images, achieving up to 53.8% accuracy in city prediction, they exhibit significant biases. Specifically, performance is substantially higher for economically developed and densely populated regions compared to less developed (-12.5%) and sparsely populated (-17.0%) areas. Moreover, regional biases of frequently over-predicting certain locations remain. For instance, they consistently predict Sydney for images taken in Australia, shown by the low entropy scores for these countries. The strong performance of VLMs also raises privacy concerns, particularly for users who share images online without the intent of being identified. Our code and dataset are publicly available at https://github.com/uscnlp-lime/FairLocator.