When using AI, it's really important that we get good at prompt engineering, which is a fancy word for “giving instructions in a clear way”. The big issues I see with this are two things. One, we're actually pretty terrible at communicating. And two, we don't actually know what we want the agent to do. The first one is self-explanatory. We use big, vague words that don't clearly state what we want or what makes an output good versus bad. I'm guilty of this all the time, but honestly, it's a lot of work to write a detailed prompt (especially when all you want is to find the cheapest place to buy groceries). The second one is more important though. I've caught myself wanting AI to write good copy, but then I realize I've never actually defined what "good copy" means. I don't have enough experience to even understand the nuances of what makes a strong foundation for a good piece versus a bad one. I'm using copywriting as the example here because marketing and lead gen is where I'm struggling in business right now, but this applies to anything. If you're having trouble with writing and you try to get AI to write something, of course it's not going to understand what makes it sound human versus not human. You haven't taken the time to explain it. Same thing with AI-generated images and video. We know it can produce good results, but it also produces bad results a lot of the time because the defaults are things like waxy skin and unnatural body compositions. The issue is that we haven't told it the specifics (things like "use a light skin tone with some blemishes and natural imperfections"). These models are exceptional at what they can do. But a lot of the time when we don't get the output we want, it's user error, not a model problem. TLDR: Bad AI outputs is a skill issue.

📝 TL;DR Years of Pokémon Go and Ingress players scanning streets, parks, and landmarks have quietly helped build a huge 3D map of the world that is now being used to guide delivery robots. The same tech that helped you find a PokéStop is now helping robots find the right curb, building entrance, and pizza drop-off spot. 🧠 Overview This is one of those stories that sounds like a joke until you realize it makes perfect sense. Niantic’s games got millions of people to point their phones at the real world, capturing billions of ground-level images and location signals that no satellite map could easily collect. Now that spatial data is being repurposed for robotics. Niantic Spatial is using it to help Coco Robotics improve navigation for delivery bots, especially in dense city areas where GPS alone is too sloppy for precise drop-offs. 📜 The Announcement The big reveal is that the crowdsourced mapping behind Niantic’s AR games has become a practical infrastructure layer for the real world. Niantic Spatial says its visual positioning system has been trained on more than 30 billion ground-level images and scans collected over the years through products like Pokémon Go and Ingress. That system is now being used in partnership with Coco Robotics, a delivery robot company, to help its robots navigate urban routes with much finer accuracy. Instead of just knowing the general block, the robots can get much closer to understanding the exact curb, storefront, or entrance they need. ⚙️ How It Works • Ground level world map - Players scanning PokéStops and landmarks created a rich street-level dataset that captures buildings, signs, benches, paths, and other real-world details. • Visual positioning system - Instead of relying only on GPS, robots compare what their cameras see to this mapped world so they can localize themselves much more precisely. • Better last meter navigation - The hardest part of delivery is often not the route, it is the final approach, finding the correct side of the street, entrance, or drop-off location.