Pokémon Go: 30 Billion Images for a Secret Map
Le brief IA que les pros lisent chaque soir
Les 7 actus IA du jour, décryptées en 5 min. Gratuit.
Inclus dès l'inscription : notre sélection des meilleurs guides & comparatifs IA.
Choisis ton rythme
Gratuit · Pas de spam · Désabonnement en 1 clic
Pokémon Go: A Global Phenomenon with Unexpected Implications
The summer of 2016 marked the arrival of Pokémon Go, a game that quickly captivated the attention of the global public. Within days, crowds of players gathered in parks and public squares, smartphones in hand, to capture virtual creatures like Jigglypuff and Gengar through augmented reality. What seemed like a simple gaming adventure was, in fact, a unique social experience, encouraging people to step outside, interact, and rediscover their urban environment.
However, behind this collective enthusiasm lay a far more ambitious project orchestrated by Niantic, the game's developer. By using Pokémon Go as bait, Niantic collected vast amounts of geolocation data and images taken by players. In less than ten years, this collection has enabled the creation of one of the most comprehensive mapping databases ever assembled. The game, free and ad-free, thus served as a discreet means to digitize the real world for the benefit of the company's future clients.
The Creation of a Colossal Visual Database
Today, this initiative is driven by Niantic Spatial, a branch dedicated to artificial intelligence and real-world mapping. This entity was born after the sale of Niantic's gaming division to Scopely, a company owned by the Saudi fund Savvy Games. Niantic Spatial recently revealed that Pokémon Go was merely a facade for a large-scale digitization project of the public domain.
The idea may seem daunting, but it reflects a certain ingenuity. Players, in order to progress in the game, took photos of the same locations from various angles and at different times of the day. Thus, millions of players unknowingly contributed to creating an impressive visual corpus: over thirty billion images, primarily of urban areas. These snapshots have allowed Niantic Spatial to develop a positioning system far more precise than traditional GPS.
The Chief Technology Officer of Niantic Spatial, Brian McClendon, recently commented on this advancement. He stated that the company could locate users with an accuracy of a few centimeters and, more importantly, determine what they were looking at. This statement highlights the scale of the project and raises questions about the transparency of the game's terms of service and privacy policy.
A Game That Hides an Ambitious Commercial Project
It is now difficult to view Pokémon Go as a mere innocent game. Niantic Spatial has transformed the collected images into a valuable resource for robotics companies like Coco Robotics, which use this data to guide their delivery robots in complex urban environments. Thanks to this positioning system, the company has created a unique spatial reference for clients requiring precise localization in the real world.
What sets this system apart from previous technologies is its ability to replace abstract GPS coordinates with instant visual identification of environments. Niantic Spatial's artificial intelligence model functions like a human brain, recognizing places it has seen thousands of times before. This recognition is made possible by the countless photos taken by players, encouraged by the game to explore and photograph their surroundings.
For John Hanke, CEO of Niantic Spatial, this project is just the beginning. He aims to create a complete virtual simulation of the real world, constantly updated to reflect changes in the physical world. This ambitious project, dubbed the company's "Pokédex," has been made possible through the involuntary participation of millions of passionate Pokémon players.
While players focused on capturing rare Pokémon and engaging in collective game activities, Niantic quietly conducted the largest field study in history, fueled by players' enthusiasm for virtual achievement. Nintendo, a partner of Niantic, has yet to comment on these revelations, leaving uncertainty about its role in this massive digitization project.
Brief IA — L'actualité IA en français
L'essentiel de l'actualité de l'intelligence artificielle, décrypté et expliqué chaque jour.