Samsara has made its Agent Studio platform generally available to all customers, and the usage curve is steepening fast: the telematics and fleet software vendor now records AI-agent interactions about seven times faster than when the tools launched in June. The most recent 500,000 interactions accumulated in 16 days, down from 112 days initially, according to the company.

Samsara counts adoption by interactions — a complete agent session through chat or voice — and voice is doing the heavy lifting. Voice agents overtook chat in interaction volume within four weeks of launch and have led every week since, the company said. More than 1,000 organizations now use Samsara agents weekly, though Samsara did not break out how many of those use voice specifically, how many customers run agents in production, or how many have built custom agents.

What can fleets actually build?

Agent Studio, launched in June and opened to the full customer base on Tuesday, is a no-code builder that lets fleets and other customers create custom AI agents drawing on their operational data. Typical tasks include compiling reports, flagging potential hours-of-service violations and answering drivers' policy questions. The platform ships inside existing Samsara subscriptions, meaning no separate line item for carriers already on the software.

Samsara's chief product officer framed the adoption numbers as evidence that usefulness, not novelty, drives uptake. "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," Johan Land said.

The company cited three customer deployments, all with figures it supplied and did not fully document:

• Grand Isle Shipyard compiles forms, inspection reports and asset data into monthly reports via an agent, cutting report-generation time by 99% — about five minutes per division — and freeing roughly $100,000 in annual specialist capacity, by the company's own estimate.
• S. Bertram Foods tracks remaining stops, driving time and shift time to flag potential HOS violations three hours in advance, reducing weekly manual review time by 84%.
• StreamTech Industrial identifies trips where drivers failed to log into their vehicles, eliminating 500 hours of unassigned driving over two months.

Samsara did not provide detailed measurement methods for all of the reported outcomes, a gap worth noting when the numbers are this large.

Why is adoption outpacing the industry?

The announcement lands against a backdrop of logistics AI stuck largely in pilot purgatory. A March 2026 Boston Consulting Group survey of more than 180 logistics providers and shippers found about 40% of logistics service providers had pushed AI beyond pilots, but only about 10% had scaled it across core operations. Just 13% reported measurable value from AI, with unclear returns on investment and internal capability gaps cited as the leading barriers.

A separate September 2026 survey by Descartes found 19% of shippers and 15% of logistics service providers using AI at scale. Samsara's interaction figures suggest fleet operations may be an exception to that sluggish pattern — at least where agents are bundled into software fleets already run.

Voice agents bring AI into the cab

The voice agents deploy through Samsara's dash cams and operate hands-free — a design choice with regulatory weight, since federal rules prohibit commercial drivers from holding a phone while driving but allow hands-free use. Most voice interactions to date are calls initiated by agents and configured by the organizations using them.

The use case is concrete: when a truck arrives at an unfamiliar facility, an agent can call the driver with instructions on entrances, parking, yard speed limits and required safety equipment. Drivers can ask follow-up questions, and managers can use operational data to investigate issues such as extended facility stops. Organizations can configure agents to trigger on conditions like a vehicle entering or leaving a geofence.

Agent Studio rounds out an AI stack that already includes Samsara Assistant, which lets users query operational data through the dashboard, and Samsara MCP, which connects authorized Samsara data to external tools such as ChatGPT, Claude and Microsoft Copilot.

What remains unproven?

For freight and logistics operators, the pitch is less time spent gathering information, monitoring compliance and answering routine questions. But the announcement offers limited evidence on breadth of deployment — no customer counts for production or custom agents — and no validation that reported time savings convert into sustained financial returns. As logistics companies weigh further AI investment against the weak scaling numbers in industry surveys, those are the metrics that will decide whether Samsara's sevenfold acceleration in adoption reflects durable value or early curiosity.