Originally a manufacturer of consumer dash-cams, Waylens has emerged as a leader in machine learning-enabled 360-degree cameras designed for commercial trucks. Riding the wave of the growing telematics market, whose size was estimated at about $50 billion in 2022,^[3] Waylens has deployed its IoT platform in over 170,000 vehicles across the country, providing it through resellers who package it along with traditional telematics devices that track everything from fuel consumption to engine faults.

“We’re looking for pedestrians, we’re looking for stop sign violations, we’re looking for tailgating,” said Verhaeghe. “It’s very safety focused.”

“A standard IoT device could be sending 50 kilobytes of data per day, which is like a single text document. We’re sending gigabytes — the equivalent of full-length movies, streaming in 4K.”

While traditional dash-cams did little more than record road incidents for later review, Waylens’ platform can detect dangerous behavior inside and outside the car and provide immediate coaching to correct it. Every image captured by the platform is analyzed by a sophisticated machine learning algorithm that runs on board the device, searching for risky behavior such as distracted driving.

Although Waylens’ algorithm runs on board the platform, connecting to the cloud using the high-speed nationwide network powered by AT&T Business allows these cameras to receive updates, improve their machine learning model and, most importantly, report on possible incidents. When the platform spots dangerous behavior, it issues an alert to the driver, allowing them to correct it immediately. If the behavior persists, a short clip is uploaded to the cloud, allowing the driver’s employer to review the incident and, if necessary, take action to correct it.

“Our infrastructure allows companies to deploy machine learning models closer to the data source, whether in vehicles, factories, or field devices, while maintaining secure, scalable connectivity,” said Lee Seto, Senior Business Development Manager, IoT, AT&T Business. “This helps enable low-latency, high-throughput data processing, streamlining machine learning workflows and making it easier for companies like Waylens to scale their safety-first solution across America’s roads.”

Waylens’ machine learning model relies on the processing of millions of images from hundreds of thousands of trucks. This massive quantity of data has allowed the model to improve by leaps and bounds, from identifying dangerous driving with 85 percent accuracy when it was first introduced to 98 percent in less than two years. Sending and receiving all that data provided a unique challenge that AT&T Business helped Waylens solve.

“Robust digital infrastructure lays the groundwork for machine learning, ensuring data accessibility, computational power, scalability and security.”

“A standard IoT device could be sending 50 kilobytes of data per day, which is like a single text document,” said Verhaeghe. “We’re sending gigabytes — the equivalent of full-length movies, streaming in 4K. AT&T Business helps us do that in a cost-effective way.”

AT&T Business’ support of Waylens is a tangible illustration of how telecom makes AI-driven innovation possible. Since the inception of Waylens’ commercial dash-cam platform, AT&T Business has been its sole data provider. This reliable connectivity is a selling point for Waylens, which has been able to simplify its pricing model by including data in the cost of a subscription to its dash-cam platform. Verhaeghe said AT&T Business made it easy to adjust its plan as the business scaled, its needs grew and its cameras began looking for more subtle events, which required more data to detect.

“Our infrastructure allows companies to deploy machine learning models closer to the data source, whether in vehicles, factories, or field devices. This helps enable low-latency, high-throughput data processing, streamlining machine learning workflows.”

“We’ve had significant opportunities come up,” he explained, “and we went to AT&T and said, ‘Hey, can you help us scale this? Do we have the bandwidth?’ They were always there to assist us in securing that business.”

Waylens uses AT&T Business’ Control Center tool to interpret data, managing hundreds of thousands of SIM cards from a single screen. Seto described Control Center as “smart, simple IoT connectivity management, ideal for any company deploying connected devices at scale.” He cited healthcare, utilities, manufacturing, transportation and retail as other IoT-reliant industries where it has helped businesses improve operational efficiency, increase security and allow for greater automation.

Training, licensing requirements and professional experience mean that truck drivers are often the safest drivers on the road. Despite this, according to Joseph Young, media director of the Insurance Institute of Highway Safety (IIHS), collisions involving large trucks “have been on the rise pretty significantly in the United States,” with the number of serious crashes increasing by 38 percent from 2009 to 2023.^[4] When a large truck collides with a passenger car, it’s much more dangerous for the people in the smaller vehicle.

“We’re looking for pedestrians, we’re looking for stop sign violations, we’re looking for tailgating. It’s very safety focused.”

“A large truck carries a lot of mass,” Young said. “It could be 80,000 pounds striking the back of a passenger vehicle. Anything that reduces collisions involving large trucks offers tremendous benefits.”

Even though those collisions “are not necessarily the fault of the drivers,” Young believes that technological solutions, such as speed limiters, improved underride guards and next-generation dash-cams like Waylens’ could make the American highway a safer place to drive. And he suggested that driver monitoring has a future not just in large trucks, but in ordinary passenger cars as well.

“Driver attention systems are something we’ve been looking into a lot,” he said. “Having these systems in large trucks is helpful and we’re hoping to see them in more passenger vehicles soon.”

Verhaeghe believes machine learning has a major part to play in improving American road safety, saying that “its precision is getting better by the day — it’s limitless.” He suggested it could help change everything from how drivers are trained and assessed when applying for licenses to how municipalities design roads or adjust dangerous intersections. The data is there — all it takes is the right technology to harness it.

“Robust digital infrastructure keeps your data accessible and ensures computational power, scalability and security,” said Scott Manwaring, Application Sales, AT&T Connected Solutions. “It lays the groundwork for machine learning.”