NewTide joined this year’s U.S. Bank Voyager Roundtable in Nashville. CEO and founder Vlad Collak spoke about where AI actually earns its keep in fleet management, and why the industry needs to move past general enthusiasm and start measuring what these tools return.
AI Adoption in Fleet Management
AI use in fleet management keeps climbing. 70% of fleet managers now report using some form of AI, up from 53% a year earlier. But using AI and getting value from it are two different things. Collak’s point was simple: the managers seeing real returns started with a specific problem to solve and made sure the tool fit how their operation already runs. The ones who plan past the pilot are the ones who get somewhere.
“The question isn’t whether to engage with AI — it’s whether you’re leading it or chasing it.”
From AI Tools to Operational Agents
The bigger shift Collak described is the move from AI tools to AI agents. A tool answers a question or offers a suggestion. An agent does the work itself. It can carry a process through several steps, keep track of where a job stands instead of starting cold each time, and take action inside the systems a fleet already runs on. A person stays in the loop and can step in whenever judgment is needed.
NewTide’s platform enables this capability through five integrated components:
- The Brain — Agentic AI : A decision-making engine that supports workflow execution and coordinates intelligent actions across operations.
- The Eyes — Computer Vision : Continuous monitoring of camera-based inputs to identify events, risks, and operational activity in real time.
- The Ears — Conversational AI Voice Assistants: Voice-enabled assistants that support driver communication, field operations, and hands-free interaction.
- The Hands — API Integrations : System integrations across FMS, TMS, ERP, and fuel card platforms, enabling AI agents to take action within existing workflows.
- The Nervous System — Data Helm : A unified data layer that connects operational information across systems and provides the foundation for coordinated AI-driven decisions.
Industry-Specific AI Agents
Collak presented several examples of AI agents designed specifically for fuel and fleet operations.
These included:
- A document ingestion agent that reads, extracts, and matches bills of lading and delivery tickets within seconds
- A spend control agent that compares fuel card transactions against truck location and tank-level data to help identify potential fraud
- A maintenance agent that detects early indicators of equipment issues and initiates work orders before failures lead to operational disruption
Each agent is structured to manage a defined operational function while escalating exceptions to human teams when judgment or intervention is required.
Key Implementation Considerations
The session also highlighted several considerations for organizations evaluating or expanding AI initiatives:
- AI adoption should be guided by defined business objectives.
- Use cases should be selected based on strategic relevance and data availability.
- Successful implementation requires internal support from both business and IT stakeholders.
- AI agents should be tested, monitored, and governed with the same discipline applied to other enterprise software systems.
- Organizations should focus on measurable operational outcomes rather than AI adoption alone.
Closing
The Voyager Roundtable pulled together people running fleets across the industry, and the conversation stayed where it’s useful: what AI is doing in day-to-day operations right now. That’s the discussion Newtide wants to be part of, and it’s the kind of problem our agents are built to handle in fuel and fleet.