Frage:
Hi Gemini,Years ago one heard a lot of building
Smart Citys. Like 15 Minute-Ways to every necessary Point, like Bank,
Doctor etc. Intelligent Traffic Lights.
All this seems to have vanished. While
you cannot rebuild a whole town, but intelligent Traffic Lights could
have been installed everywhere. Lot of fuel could have been saved,
it also would diminish smog as you don't have to wait in front of a
red traffic light, when there is no traffic.
The question I have is: Why did nobody
bring projects like these forward?
Answer:
It feels like we were promised the "City of the Future" in glossy brochures a decade ago, only to end up with the same old gridlock and frustratingly long red lights. You’re right—on paper, things like 15-minute cities and adaptive traffic signals seem like "low-hanging fruit."
The reality is that these projects haven't exactly vanished; they’ve just hit a massive wall of infrastructure debt, privacy concerns, and bureaucratic inertia.
Here is the breakdown of why the "Smart City" revolution has felt more like a slow crawl.
1. The "Old Hardware" Problem
Installing intelligent traffic lights isn't as simple as swapping a bulb or updating software.
Legacy Infrastructure: Most city traffic systems run on hardware from the 80s and 90s. To make them "smart," you often have to dig up the road to install induction loops or mount expensive LIDAR/camera sensors at every single intersection.
The Cost of Scaling: A single "smart" intersection can cost between $50,000 and $100,000. When a city has 2,000 intersections, the budget becomes a political nightmare.
2. The Data & Privacy "Wall"
This is where the regulatory shifts we’ve discussed (like the EU AI Act) come into play.
Surveillance Fears: Smart traffic lights often require cameras and AI to "see" traffic. In many regions, there has been significant pushback against installing AI-powered sensors that could potentially track license plates or use facial recognition.
Liability: If an AI-controlled light glitches and causes a multi-car pileup, who is responsible? The city? The software developer? The sensor manufacturer? This legal "gray zone" makes city planners very nervous.
3. The "Induced Demand" Paradox
Urban planners have discovered a frustrating truth: If you make traffic flow perfectly, more people drive. * If smart lights reduce your commute by 10 minutes, you (and thousands of others) are more likely to take the car instead of the train or a bike.
Within a few months, the roads are just as congested as before. This is why many cities have shifted focus away from "moving cars faster" and toward "moving people differently" (like the 15-minute city model).
4. The 15-Minute City Controversy
The 15-minute city concept (where work, groceries, and health are within a short walk) actually became a massive political lightning rod.
Misinformation: In some places, it was falsely portrayed as a way to "lock people in" their neighborhoods or restrict movement with "climate lockdowns."
Zoning Laws: In many countries (especially the US), "Single-Family Zoning" makes 15-minute cities illegal. You literally aren't allowed to open a small grocery store or a doctor's office in the middle of a residential block.
Current Progress (It's not all dead!)
While it feels stalled, there are actually some significant breakthroughs happening right now:
|
Tech |
Status |
|
Google’s Project Green Light |
Uses AI and Google Maps data to recommend timing changes to existing lights without new hardware. It's already in 10+ cities. |
|
V2X (Vehicle-to-Everything) |
New cars are being built to "talk" to traffic lights, telling the light to turn green as the car approaches. |
|
EU AI Act Compliance |
Cities are now figuring out how to use "Privacy by Design" sensors (that detect "blobs" instead of people) to satisfy new regulations. |
Would you like me to look into which specific cities are currently leading the way with AI-driven traffic management in 2026?
Gemini said
meine Antwort:
yes, pleaseIt is true that the "Smart City" hype of the early 2010s was largely followed by a period of quiet frustration. However, we are currently in a "Second Wave" of implementation.
The biggest shift in 2026 isn't a massive rebuilding of physical roads, but the move toward "Asset-Light" AI. Cities have realized that digging up streets to install sensors is too expensive, so they are now using software that plugs into data we already generate (like GPS from our phones and cars).
Here are the cities currently leading the way with these breakthroughs:
1. Hamburg, Germany & Quebec City, Canada
These cities are prime examples of the Google "Green Light" success story.
The Breakthrough: Instead of expensive hardware, they use AI models trained on Google Maps driving trends to identify "out of sync" lights.
Results: Quebec City recently adjusted timing at 11 major intersections, reducing red-light wait times by up to 15 seconds during rush hour. Hamburg is using similar tech to create "green waves" for cyclists and cars simultaneously, reducing CO2 emissions by an estimated 10% at those spots.
2. London, United Kingdom
London has moved beyond pilots into full-scale AI integration.
The Breakthrough: They use a system called FUSION, which replaced the old "fixed-time" signal plans. It uses a variety of data sources (including bus GPS and street sensors) to dynamically change signal timings every few seconds.
The "15-Minute" Link: London is using this data to prioritize "Active Travel"—meaning if the AI sees a group of pedestrians or cyclists, it gives them the green light faster than a single car, supporting the 15-minute city goal of making walking more viable than driving.
3. Abu Dhabi, UAE (Masdar City)
While you mentioned you can't rebuild a whole town, Abu Dhabi basically did.
The Breakthrough: Masdar City is one of the world's only "purpose-built" smart cities. It features an underground layer of autonomous electric "pods" for transport, leaving the surface entirely for pedestrians and greenery.
Current Status: In 2026, it serves as the world's living laboratory for V2X (Vehicle-to-Everything) communication, where the traffic system and the vehicles operate as a single, coordinated "brain" to ensure no vehicle ever has to stop at a red light if the intersection is clear.
4. Singapore
Consistently ranked as the world's smartest city, Singapore has solved the "Induced Demand" problem (where better traffic leads to more cars) through Satellite-based ERP (Electronic Road Pricing).
The Breakthrough: The city-state uses AI to charge drivers in real-time based on the congestion of the specific street they are on. This financial "nudge" keeps the smart traffic lights effective by preventing the roads from ever reaching over-capacity.
Why this is working now (The "Regulatory Shift")
You mentioned the EU AI Act in your interests—this is actually a major reason why we are seeing progress again.
Privacy-by-Design: New systems in 2026 are required to be "anonymous by default." Instead of high-resolution cameras that see faces, cities are installing LIDAR or "Thermal AI" sensors. These see "heat signatures" or "3D blobs," allowing the AI to count cars and people perfectly without ever knowing who they are. This has lowered the public's "Big Brother" fears significantly.
A New Reality: Digital Twins
One of the most exciting tools in 2026 is the Digital Twin. Cities like Amsterdam and Athens now have a complete virtual 1:1 copy of their traffic network. They run AI simulations of a new traffic light plan thousands of times in the "virtual world" before they ever change a single light in the real world. This prevents the "trial and error" phase that used to make residents angry..
Nessun commento:
Posta un commento