Masabi has formalized more than a year of artificial intelligence research into a dedicated incubator called Masabi AI Labs, currently working with four transit agencies in North America and Europe alongside unnamed payment and technology partners on six defined use cases spanning data analytics, revenue protection and emerging conversational interfaces.

The fare-payment vendor said the incubator consolidates discovery sessions conducted with transit agencies over the past twelve months into a formal programme. The six use cases are:

• Making transit data easier to interrogate and act on
• Improving customer service
• Strengthening revenue protection
• Augmenting repetitive operational and back-office processes
• Helping teams manage increasingly sophisticated fare systems
• Exploring how passengers could discover and purchase public transit through emerging AI interfaces

How is Greater Dayton RTA using the tool?

Greater Dayton Regional Transit Authority (RTA), an existing Masabi customer, has begun piloting an AI-powered natural language data tool built inside the lab. The interface allows technical and non-technical staff to query ridership and fare-validation data in plain language and receive structured analytics in response, removing the need to route routine questions through specialist reporting teams.

The Ohio-based operator has applied the tool to three operational workstreams: route-level validation data, transfer patterns feeding into an ongoing system redesign, and fare-media usage. Masabi said the agency was using the outputs to shorten the time required to move from a question to an actionable insight, while retaining traceability to the source datasets.

What does the revenue-protection track involve?

In a separate workstream, Masabi AI Labs has partnered with a large U.S. transit agency — which the company did not name — and a machine-learning specialist to analyse ticket purchase and validation records. The joint work has examined distinct categories of revenue leakage and tested whether pattern-recognition models can identify fare-evasion behaviours that rule-based systems miss.

The longer-term objective, according to Masabi, is to give agency finance and enforcement teams a clearer picture of where revenue is being lost, and to direct investigation and inspection effort toward the routes, modes or fare products where it produces the greatest return.

How is Masabi positioning the programme?

Masabi has framed the launch as a deliberate step back from speculative AI projects. CEO Brian Zanghi said the company's approach to artificial intelligence mirrors its broader engineering ethos. "At Masabi we have never been about technology for technology's sake. The company was founded with the ethos of solving everyday hassle for people using technology and our approach to AI has been the same," Zanghi said. "The question we are interested in is where it can genuinely make public transit better. That means starting with the problems our customers and their passengers experience. By working directly with agencies and technology partners, we can explore ideas against real-world requirements, understand where they add value and establish the safeguards needed to use them responsibly."

The company argues the use cases address a structural problem. Transit agencies generate large volumes of ticketing, fare-validation, ridership, payment and revenue data, yet extracting answers can require specialist analysts, dedicated reporting tools and manual aggregation. Masabi AI Labs is positioning its tools as a means of compressing that workflow.

Masabi has not disclosed a target date for graduating any of the pilots into general availability. The company expects to add additional agency partners and to report on measurable outcomes of the revenue-protection and customer-service pilots as the lab's work extends into its next phase.