Niantic, the company behind the extremely popular augmented reality mobile games Pokémon Go and Ingress, announced that it is using data collected by its millions of players to create an AI model that can navigate the physical world.
In a blog post published last week, first spotted by Garbage Day, Niantic says it is building a “Large Geospatial Model.” This name, the company explains, is a direct reference to Large Language Models (LLMs) Like OpenAI’s GPT, which are trained on vast quantities of text scraped from the internet in order to process and produce natural language. Niantic explains that a Large Geospatial Model, or LGM, aims to do the same for the physical world, a technology it says “will enable computers not only to perceive and understand physical spaces, but also to interact with them in new ways, forming a critical component of AR glasses and fields beyond, including robotics, content creation and autonomous systems. As we move from phones to wearable technology linked to the real world, spatial intelligence will become the world’s future operating system.”
By training an AI model on millions of geolocated images from around the world, the model will be able to predict its immediate environment in the same way an LLM is able to produce coherent and convincing sentences by statistically determining what word is likely to follow another.
Thing is, consider Google maps. It’s been harvesting data secretly and openly for a long time. I vaguely remember a time when Street View cars were found to be harvesting WiFi information in Australia and their response was, “oops, our engineers made a mistake.” Yeah, right.
But, Google maps is an amazing tool. All that traffic info? All those time estimates? Maybe it’s worth it. Maybe if people knew what they were providing, and the result they’d get, they’d still be happy to give all that “free” data to Google.
Putting aside the ethics of a company taking (stealing? or shall we call it, pirating?) all the ownership of that knowledge asset, if they make a really useful tool from it perhaps Pokémon players will be glad to have been part of such an epic achievement.
The traffic data is not as good as it appears. It is completely closed, only given to police and goverment agencies. No API, no numerical values for speed (only 5 ‘color codes’ that are relative to location, so are almost useles) and numerical data is not given even to academics. I spent almost a whole month trying to get actual useful data for academic purposes, but Google really went out in their path to make it impossible.
It has the potential to be an excellent tool: crowsourced real-time data, access to historical data and it is incredibly fine-grained, improving over goverment data (at least in my city) by a 10 or 100x factor. But no, it had to be yet another Google’s tool for spying on people, not giving it away and sell it to police.
I worked for a company contracted by government agencies (city/county/state/fed) to gather traffic statistics. We were used because they were not able to use Google traffic data as a blanket rule.
But for ordinary drivers, it’s great.