Samsara has launched Agent Studio, a platform that lets fleets build AI agents to handle tasks such as flagging hours-of-service risks, analyzing fuel use, and answering drivers’ questions about company policies.

Available to all Samsara customers, the platform includes prebuilt artificial intelligence agents as well as tools for creating custom chat and voice agents without writing code.

The agents draw on fleet data such as vehicle location, diagnostics, video and driver behavior, along with company-specific information, to identify issues and take action.

“AI is quickly adopted when it’s useful, and in physical operations that means an agent has to understand the operation and meet people where the work happens,” said Johan Land, Samsara’s chief product officer. “In the back office, that might happen through chat. On the road, it happens through a phone call.”

Agent Studio works alongside Samsara Assistant, which answers questions and supports tasks within the company’s dashboard, and prebuilt agents for safety, maintenance, and compliance.

How Fleets Are Using AI Agents

Samsara said customers are expanding their use of agents beyond report generation to tasks such as identifying fuel savings, anticipating compliance problems, and providing driver support.

Among organizations with agents deployed, 34% already run two or more.

Among organizations using voice agents, the company said, 74% have increased weekly usage since their first month; more than half have at least doubled it.

Examples cited by the company include:

• Flagging hours-of-service risks. S. Bertram Foods uses an agent to track remaining stops, driving time, and shift time. Samsara said it flags potential HOS violations three hours in advance and has reduced weekly manual review time by 84%.
• Finding unidentified drivers. StreamTech Industrial uses an agent to identify trips where a driver has not logged in. According to Samsara, the company reduced unassigned driving time by 500 hours over two months.
• Analyzing idling. A North American passenger transit provider uses an agent to distinguish idling related to driver behavior from regeneration cycles, maintenance, and weather, helping managers identify where to focus fuel-saving efforts.
• Answering driver questions. A food distributor built a “Driver Wingman” agent that uses company policies to answer questions about parking, personal conveyance, and fuel stops.
• Automating reports. Grand Isle Shipyard uses an agent to compile forms, inspection reports, and asset data into monthly reports. Samsara said the process now takes about five minutes per division, a 99% reduction in report generation time.

Bringing AI Support Into the Cab

Agent Studio also lets customers build agents that make and receive phone calls, giving drivers operational information and company guidance outside normal office hours.

For example, Samsara said a truck entering a geofence around a distribution center could trigger a call with instructions about entrances, parking, yard speed limits, and required safety equipment. Drivers could ask follow-up questions, and the agent could collect information about the visit after the truck leaves.

Managers could then use Samsara Assistant to examine patterns, such as which locations have the longest dwell times.

Samsara said voice interactions surpassed chat interactions within four weeks of its voice-agent launch and have remained higher each week since. Among organizations using voice agents, about 75% have increased their weekly use, with more than half at least doubling it.

The company also reported that AI agent adoption on its platform has increased sevenfold since June.  The first 500,000 interactions took 112 days; the most recent 500,000 took 16 days.