Conversation

We all know AI models can now create realistic images and videos but how do they fare at identifying where a real image was taken? Bellingcat researchers have put large language models to the test; we ran 500 geolocation tests, with 20 models each analysing the same set of 25 images. Find out which LLM performed best and which one failed at identifying our geolocation challenges more than others here: https://www.bellingcat.com/resources/how-tos/2025/06/06/have-llms-finally-mastered-geolocation/?utm_source=mastodon