Graphenex AI: Singapore Startup Mapping Cities with Brain-Inspired Networks
Date: March 15, 2026
What if AI could think about cities the way a human brain thinks about connections? That's exactly what Graphenex AI, a Singapore-based startup, is attempting—with potentially massive implications for how Asian cities grow and adapt.
The Problem with Traditional Urban Data
Cities are fundamentally about connections. Roads link to intersections. Buildings connect to power grids. Water flows through pipes that link to treatment plants. Yet most AI systems designed for urban planning treat these connections as afterthoughts, analyzing data in isolated chunks rather than understanding the whole system.
"Traditional machine learning looks at data like a spreadsheet—rows and columns. But a city isn't a spreadsheet. It's a web of relationships," explains Dr. Sarah Chen, Graphenex AI's founder and former researcher at the Agency for Science, Technology and Research (A*STAR). "Our brains don't process cities as tables of numbers. We think in connections. Graph neural networks let AI do the same."
How Graph Neural Networks Work
Graph neural networks (GNNs) represent a relatively new frontier in AI. Unlike traditional neural networks that process data in fixed sequences or grids, GNNs are designed to process information organized as graphs—networks of nodes (entities) connected by edges (relationships).
The technology takes inspiration from how biological neural networks operate. In the human brain, neurons don't process information in isolation—they communicate through synapses, passing signals through complex webs of connections. Similarly, GNNs allow AI to understand how changing one element of a system affects every other connected element.
Singapore as a Living Lab
Graphenex AI has found an ideal testing ground in Singapore. The city-state's compact size, comprehensive sensor networks, and forward-thinking government have created what's essentially a perfect lab for urban AI.
The company is currently working with Singapore's Land Transport Authority to optimize traffic flow across the entire island. Their GNN system analyzes real-time data from traffic cameras, sensors, and navigation apps to predict congestion patterns and suggest coordinated signal timing across thousands of intersections—something traditional systems struggle to achieve.
Early results are promising. Pilot projects in the Jurong Lake District have shown a 23% reduction in average commute times during peak hours, with benefits cascading across the network as the system learns to anticipate ripple effects.
Expanding Across Asia
The potential extends far beyond traffic. Graphenex is now deploying its technology across several Asian capitals:
- Jakarta, Indonesia: Partnering with the city government to optimize public bus routes and reduce the notorious congestion that costs the economy billions annually.
- Manila, Philippines: Working on flood prediction systems that model how water flows through the city's complex network of waterways and drains.
- Bangkok, Thailand: Helping the city plan evacuation routes by modeling how people would move through the urban network during emergencies.
The Bigger Picture
What makes Graphenex notable is less the specific applications—others are working on smart city solutions—than its approach. By focusing on the relational structure of urban systems, the company is tackling problems that have historically stumped city planners.
"The exciting thing isn't any single use case. It's that the same underlying technology can model power grids, water systems, social networks, and transportation—all as interconnected systems," Chen says. "We're building a digital twin of how Asian cities actually work."
With the Asian Development Bank estimating that Asian cities will need $1.7 trillion annually in infrastructure investment through 2030, the ability to plan smarter—rather than just bigger—has become crucial. Graphenex AI represents Singapore's growing contribution to solving that challenge.
The startup has raised S$45 million in Series B funding led by Sequoia Capital India, with participation from Singapore's EDBI and returns from earlier investors like Wavemaker Partners. Plans include expanding to ten Asian cities within three years and developing specialized models for different types of urban environments.
As cities across the region struggle with congestion, aging infrastructure, and climate adaptation, brain-inspired AI that understands connections may prove invaluable. Singapore, once again, is positioning itself at the forefront of that possibility.
What do you think about AI-powered urban planning? Share your thoughts on our social channels.
Related Links:
food.whatsgood.sg - Discover the best food places in Singapore through detailed reviews and local recommendations.
gta.sg - Your guide to Singapore's public transport, routing, and commute planning.