SewerAI uses AI to spot defects in sewer pipes

Climate change is increasing the risk, frequency and cost of sewer failures. Floods are becoming more frequent, causing backups that often overwhelm wastewater treatment systems. To make matters worse, America's infrastructure is seriously outdated. The EPA estimates that nearly $700 billion in investments will be needed over the next 20 years just to maintain existing sewer, stormwater and other drinking water systems.

Matthew Rosenthal and Billy Gilmartin, both of whom come from the wastewater treatment industry, saw an opportunity to use technology to solve the problem—at least on a small scale. Five years ago, the pair co-founded SewerAI, which uses AI to automate the data collection and defect flagging required during a sewer inspection.

“Most infrastructure was built after World War II and is now reaching the end of its useful life, resulting in more frequent failures and higher costs,” Rosenthal told TechCrunch. “SewerAI is revolutionizing underground infrastructure inspection and management with its AI-driven software-as-a-service platform.”

SewerAI began as a side project for Rosenthal. After co-founding two wastewater analysis and services companies, he began taking online courses on AI. While experimenting with AI models to predict sewer damage in inspection videos, Rosenthal enlisted the help of Gilmartin, who was working at a wastewater inspection company at the time.

Photo credits: ChannelAI

Today, SewerAI – whose customers include municipalities, utilities and private contractors – sells cloud-based, AI-powered subscription products designed to simplify on-site inspections and data management of sewer infrastructure.

One of these products, Pioneer, allows field inspectors to upload inspection data to the cloud and flag issues—data that project managers can then use to plan repairs to pipes. Another tool, AutoCode, automatically flags inspections of pipes and manholes and creates 3D models of infrastructure from video captured with a GoPro or other camera.

“The incumbents offer on-premise or on-truck software that has seen little innovation in the last 20 years,” said Rosenthal. “SewerAI's technology increases revenue and profits by enabling more inspections per day at a lower cost.”

SewerAI is not alone in the emerging market for AI-powered pipe inspections. The company's competitors include Subterra, which maps, analyzes and predicts problems with pipelines; ClearObject, which offers software that analyzes pipe inspection footage for damage; and Pallon, which develops algorithms to detect potential problems in sewers from still images.

ChannelAIPhoto credits: ChannelAI

What sets SewerAI apart, Rosenthal says, is the quality of its data — particularly the quality of its model training data. Rosenthal says SewerAI has footage of inspections of 135 million feet of pipe from municipalities and independent contractors. While it's only a fraction of the 6.8 billion feet of sewer pipe in the U.S., it's a large enough dataset to train a competitive defect-detection AI, Rosenthal says.

“Our products streamline field inspections and data management, enabling customers to proactively manage infrastructure rather than react to emergencies,” Rosenthal said.

SewerAI's selling point convinced investors like Innovius Capital, which, along with others, invested $15 million in SewerAI's latest funding round, bringing SewerAI's total raised to $25 million. The money will be invested in launching, training AI models, hiring employees, and expanding SewerAI's product portfolio beyond inspection tools.

“SewerAI continues to grow and we are seeing increasing demand for our platform as we enable people to achieve more with their existing budgets. This has led to us closing our first seven-figure contracts,” said Rosenthal.

You might also like

Comments are closed.