A beta project Geotab launched last year that introduced generative AI models into its platform has now evolved into an AI copilot designed for fleets to reduce time to insight into their connected vehicle data.
Project G, an analytics digital assistant using natural language models to transform the process of analyzing fleet data, tested a chat interface linked to Geotab Data Connector, where customers could get data insights on their fleet based on a specific question. The beta project worked similar to Chat GPT. During Geotab Connect, the company’s annual user conference that was hosted last week in Las Vegas, Geotab Vice President of Data and Analytics Mike Branch revealed two barriers Project G posed: it had no memory and had limited data access.
The company worked with its fleet customers to solve these problems, resulting in Geotab Ace. Ace will launch with an early access mode in the second quarter of this year.
“The two problems that I just spoke of, we solved those,” Branch said during his keynote. “One is that no longer is it only Geotab’s Data Connector data; now you have access to zones, rules, groups, almost everything in MyGeotab, which is really transformational if you think about it. And it's no longer a goldfish; it has memory, so that same conversation that you're used to having with Chat GPT, you can have that with Geotab Ace as well.”
Ace distills billions of data points daily and converts that data into intelligent insights via conversational AI directly in the Geotab platform, enabling fleets to more easily interact with their data and enhance their ability to make data-driven decisions. Fleets have access to data like predictive safety analytics, predictive maintenance, trip data, zone activity, electric vehicle statistics, exception events, GPS tracking, and more.
Fleets can ask Ace a broad range of complex questions and get customized answers tailored to their business. The platform now remembers past interactions to improve future responses, which are given with a thorough explanation of its interpretation of the question.
Branch said those responses are immediate and instrumental to a fleet’s success.
“Since kind of time immemorial, I think one of the things that customers are really interested in doing is driving down time to insight,” he said. “We've got 50,000 customers across so many different segments. There's no one-size-fits-all dashboard for everything. So our hypothesis is that generative AI can really help reduce that time to insight.”
