Can AI dash cams detect distracted driving behaviors?

Distracted driving includes any activity that diverts a driver's attention from the road. For commercial fleets, distraction events may lead to injuries, liability claims, or worse. Samsara's AI detections are powered by industry-leading models and trained on trillions of data points. Using AI Dash Cams to capture safety hazards and risky behaviors, fleets can detect distracted driving in real time and take a proactive approach to safety. With Samsara, you can:

• Detect warning signs of distracted driving and alert drivers in real-time, with AI trained on the world’s largest connected operations dataset.

• Alert safety managers in real-time, so they can take appropriate action based on organization policy. 

• Surface incidents and patterns to Safety Managers for coaching.

• Build a safer fleet by assessing patterns and trends of distracted driving across your fleet, and taking corrective action to improve driving performance.

For example, USIC took on the critical challenge of driver fatigue by becoming an early adopter of Samsara's Drowsiness Detection. The impact was transformative, demonstrating how technology can prevent the most severe incidents. Samsara Drowsiness Detection resulted in a staggering 75% decrease in drowsy driving incidents, which fell from 21 to just four over a nine-month period.

Samsara’s artificial intelligence (AI) technology is powered by the world’s largest connected operations data set based on trillions of data points, making it possible to detect a variety of risky behaviors in real time. Here’s how Samsara’s AI technology sets itself apart:

• Trained on trillions of data points: Using Samsara’s vast data set, AI models are able to “learn” from a lot of examples of what fatigue looks like, and how it might show up in different circumstances. Samsara’s AI detections, including Drowsiness Detection, are trained on more than 180 billion minutes of video footage and data from 220 billion miles traveled. Because Samsara can pull hundreds of thousands of real examples from our data, they can train an end-to-end deep learning model that has a holistic, contextual understanding of all of the behaviors that constitute drowsiness. Technology providers that use a single behavior, like prolonged eye closure or yawning, to detect drowsiness often rely solely on older techniques like scanning facial landmarks, which are less robust. 

• Multi-behavior detection: Samsara’s detection model analyzes 17+ driver behaviors that indicate fatigue, including head nodding, slouching, prolonged idling, face-rubbing, yawning, and prolonged eye closure. Through iterative training on the scale of Samsara data, the model learns what fatigue looks like beyond these 17 behaviors; it builds a contextual and holistic understanding of all the ways drowsiness might show up in the cab. Built to reflect the latest academic research on drowsiness and industry-standard frameworks like the Karolinska Sleepiness Scale, it produces a fatigue score and only alerts if that score passes a certain threshold. This makes it possible for Samsara’s AI model to better learn behaviors that indicate drowsiness risk, for both early-stage and late-stage fatigue. As a result, Samsara avoids inundating drivers with too many false positives. For example, Samsara won’t alert just because of one single instance of a yawn. 

• Leverages the most modern science: Samsara’s AI model aligns with the most modern fatigue detection research across both the United States and European Union. It also leverages the most modern ML techniques, rather than relying on older techniques like facial landmarking. In addition, Samsara has an in-house team of PhD experts in ML who continuously fine-tune the model.

Samsara’s AI detections are able to detect a variety of distracted driving behaviors, including holding or using a phone while driving, texting, dialing, swiping, or tapping on a device, looking down at phone, food, lap, or hands, adjusting radio or vehicle controls or hat or scratching head while looking down, looking up and down frequently, talking to passengers for too long or looking away from the road, reading paperwork, writing notes, or applying makeup, nodding off, prolonged eye closure, yawning, rubbing eyes, struggling to keep eyes open, slightly longer glances away from the road, slower eye blinks, rubbing or scratching face, head, or neck, lack of activity or reduced movement, blank stare, leaning head back, slouching in the seat, appearing to be nodding off with loss of neck muscle control, head bobbing, prolonged eye closure with eyes closed greater than 2 seconds, slow eye closures, squinting or strained efforts to open eyes, eating or drinking, smoking, reaching for objects on another seat or dropped items, deliberately or accidentally blocking the camera view inside the cab, detecting unauthorized or unexpected passengers as a distraction or policy violation, and more.

Fleet managers use Samsara’s distracted driving data in several ways to coach drivers and reduce incidents:

• Real-Time In-Cab Nudges™ and AI-Powered Alerts: When Samsara detects distracted behavior (e.g., looking at a phone, inattention), it triggers immediate escalating alerts in the cab, from subtle beeps to voice prompts like “Distracted.” This nudges drivers to self-correct before an incident occurs. These in-cab alerts can lead to up to a 90% reduction in distracted driving events in just 24 hours, as seen with USIC.

• Safety Inbox: When distraction is persistent or severe, the AI Dash Cam uploads a video event to the Safety Inbox for review. Risk is surfaced in a manageable way (e.g., max uploads per hour/day to avoid alert fatigue). Safety managers receive email and text alerts to review these events, filter by risk/severity, and prioritize coaching for repeat or high-risk behaviors.

• Virtual Coaching: Managers can send customized feedback directly to drivers via the Samsara Driver App, often immediately after a trip or event, making coaching timely and scalable. Samsara also allows auto-assignment of targeted training modules based on detected behaviors.

• Data-driven insights: AI-powered insights in the dashboard help managers spot patterns, prioritize high-risk drivers, and recognize those with defensive driving habits, enabling efficient, fair coaching.

Samsara’s AI Dash Cams deliver strong ROI by reducing accidents, decreasing insurance claims, lowering premiums, and minimizing safety-related costs. Samsara customers across industries have leveraged Samsara AI Dash Cams to take their fleet safety operations to the next level. Here are a few examples:

• Rasmussen Group: Decreased mobile usage while driving by 90%, speeding and running stop signs by 70%, and backup incidents by 25%, resulting in a 30% increase in time savings.

• Chalk Mountain Services: Reduced preventable accident costs by 86% and increased driver retention by 15%.

• AWP Safety: Reduced harsh events by 75%, speeding by 74%, seat belt usage violations by 73%, and mobile phone usage by 81%, while increasing coaching effectiveness by 75% and improving coaching timelines by 38%.

• Collins Trucking: Achieved a 40% reduction in overall accident rates and eliminated rear-end accidents entirely.

• ITF Group: Saved 4.4 million in insurance costs, Reduced accidents by 30%, cut risky driving behaviors by 50%, and saved 4 million in fuel costs.

• R.E. Garrison Trucking: Lowered preventable accidents by 30% and reduced workers’ compensation claims by 83%.

• Werner Enterprises: Reduced accident rates by 22%, reached a 19-year low in DOT-reportable accidents, and decreased harsh turning events by 50%.

• Coshocton Trucking: Reduced incidents by 200% and saved more than $650,000 in insurance and fuel costs.

• Quality Custom Distribution: Eliminates drowsy-driving crashes and significantly reduces claims within six months, saving millions of dollars and overturning FMCSA-recorded crashes.

• UniGroup: Reduced overall safety events by 42%, generating meaningful risk-reduction ROI across fleet operations.

• Jordan Carriers: Cut safety events by 50%, reduced coaching time by half, and decreased the number of drivers needing daily coaching by 93%.

• Corcoran Transportation Group: Avoided a $500,000 lawsuit by using video evidence to quickly exonerate a driver.

• Heniff: Reduced crash-related liabilities by 50% using Samsara.

• JRayl Transport: Improved CSA Unsafe Driving scores by 71%, enhanced CSA Crash Indicator scores by 57%, and exonerated more than 20 drivers from false claims.

• Salisbury Trucking: Improved CSA scores after switching from Motive to Samsara, demonstrating significant compliance and safety ROI.

• Denali: Reduced at-fault serious accidents by 83% and built a safety culture of positive recognition by consistently maintaining an average safety score above 90.

• Swissport North America: Reduced incidents by 60% and vehicle damage by 23% with Samsara AI Dash Cams.

• Glazer’s Beer & Beverage: Reduced their at-fault accident rate by 60% and related costs by 86% as well as saved several millions in potential claims costs. 

• Primoris: Reduced incidents by 42%, cut harsh driving events by 66%, and saved $2 million in insurance costs.

• Emery Sapp & Sons: Reduced DOT-reportable incidents by 40% and improved FMCSA safety scores by 11%.

• Sunrun: Reduced vehicle incident rates by 32% in under one year, demonstrating fast time-to-value safety ROI.

• Aunt Millie’s: Identified $1 million in potential injury-claim savings by improving visibility into safety risks.

• Interstate Waste Services: Reduced claims costs by 20% with Samsara, delivering tangible financial ROI.

• Foundation Building Materials (FBM): Saved more than $100,000 through faster driver exonerations.

• GCI Slingers: Achieved a 100% reduction in unsafe mobile phone usage and secured a 7% reduction in insurance premiums.

• Mavin: Saved $15,000 in insurance costs with Samsara AI Dash Cams, producing immediate cost-savings ROI.

• ViaSun: Improved safety scores by 28% and significantly reduced liability exposure by using video evidence to confidently dispute claims, after choosing Samsara over Motive.

• KeHE Distributors: Saved thousands of hours in incident investigation time by consolidating driver and back-office workflows.

• Eurovia USA: Reduced mobile phone usage by 80% and no-seatbelt violations by 76%.

• GardaWorld Security: Reduced speeding by 74%, close following by 67%, and rolling stops by 25% with Samsara AI Dash Cams.

• Ace Tree Enterprise: Doubled safety scores from 42 to over 90 after switching away from Motive and adopting Samsara, gaining accurate data to improve safety coaching, resolve false claims, and reduce accidents.

• Hydromax USA: Improved safety scores by 9% year over year, demonstrating sustained compliance and risk-reduction improvements.

• City & County of Denver: Reduced safety events by 94%, cut harsh driving by 99%, and decreased distracted driving by 98%, delivering exceptional public safety and risk-reduction improvements.

• City of New Orleans: Achieved an 81% reduction in collision risk and a 46% decrease in mobile phone use while driving, resulting in significant roadway safety improvements.

• City of El Paso: Reduced accidents by 29% while using real-time video and location data to save the department millions in claims and liability costs, generating strong taxpayer-value.

• City of Allentown: Reduced incidents by 25%, saved more than $1 million in taxpayer dollars, and decreased citizen complaints by 95%, delivering improvements in safety, cost-savings, and community-trust.

• City of Boynton Beach: Reduced overall driver risk factors by 58%, lowered speeding by 17%, and avoided thousands of dollars in false claims through driver exoneration.

• City of Syracuse: Reduced citizen complaints by 30% by increasing transparency through video visibility.

• Grand Isle Shipyard: Achieved a 50% reduction in at-fault accidents with Samsara AI Dash Cams.

• Gwinnett County Public Schools: Decreased investigation time by 50% with Samsara AI Dash Cams.

• Canyons School District: Reduced incidents by 50%, saved $80,000 in claim payouts, and spent 75% less time reviewing footage through proactive coaching.

• Sprint Waste Services: Saved $500,000 on an insurance renewal and reduced speeding by 58% while exonerating drivers on the spot.

• Aeroport Taxi & Limousine: Reduced liability costs and saved money by exonerating drivers while cutting more than 30 minutes from each incident resolution.

• Mobile County Public Schools: Reduced speeding incidents by 50% with Samsara AI Dash Cams.

• Caddo Parish Public Schools: Reduced distracted driving by 26%, harsh events by 15%, and speeding by 13%—ultimately resulting in a 100% decrease in major incidents. As a result, they reduced parent calls by 35%, demonstrating improved safety compliance and increased public trust ROI.

• Coach USA: Decreased preventable accidents by 92% and claim costs by 75% with Samsara AI Multicam.

Samsara is the most trusted and reliable safety platform available today. As the market leader in safety for fleet operations, Samsara offers best-in-class safety products including AI Dash Cams, Camera Connector, AI Multicam, Worker Safety, and more. 

• Most complete view of risk: Samsara’s AI Dash Cams offer 1440p resolution, HDR night vision, and 4G LTE cellular connectivity. Samsara gives you AI visibility into 30+ risky driving behaviors, blind spot hazards around the truck, and external risks like weather. It also flags patterns of risk directly to managers so they can act proactively. This scale is possible because of Samsara’s AI, and their partnership with thousands of fleets. This stands apart from other safety solutions, which only flag isolated risky events and miss environmental hazards or behavior patterns.

• AI-powered coaching at scale: Samsara’s AI is trained on vast datasets powered by 180 billion minutes of video and 90 billion miles. Instead of requiring large teams to manually review and coach drivers, Samsara’s AI reviews, triages and prioritizes events, sends lower risk events directly to drivers for self-review, and escalates only the most severe cases to managers. The result is that managers spend their time with the fraction of drivers that need their attention, while everyone else gets consistent, timely feedback daily. This offers a distinct advantage over other safety solutions that rely on manual event reviews, or simply send everything to drivers. It also makes it possible for fleets of all sizes to scale and standardize high-quality coaching and training without adding safety team resources. 

• Driver recognition and engagement: Safety is integrated into the Samsara Driver App alongside navigation and compliance, keeping coaching part of a driver’s daily workflow. Drivers receive balanced feedback that includes recognition as well as coaching. In-cab AI alerts are only active on imminent, severe risks. This stands apart from other safety solutions, which spread workflows across multiple apps, provide one-sided feedback with no positive recognition, overwhelm drivers with too many notifications in cab, and make coaching feel tedious or punitive.

• Protection beyond the cab: Samsara uniquely extends safety to the field with an advanced worker safety solution, featuring one-click SOS, proactive worker check-ins, and real-time incident response, protecting teams on the road and on job sites. No other provider offers this in a single platform.

• End-to-end risk management: Samsara combines detection, alerts, coaching, recognition, qualifications, emergency response, and claims management all in one system. Other safety providers do not offer all of these components of risk reduction and management, or they’re done in disparate systems, making it difficult to implement and delay response.