For field service, the promise of revolutionary technology lies with the data that powers it.
TL;DR: Techs, AI is not coming for your job. Technicians are a vital part of a continuous cycle that feeds the AI machine.
There has been a lot of excitement surrounding the potential of generative* artificial intelligence (AI) to revolutionize, well … everything … the field service industry included. (*Generative AI refers to algorithms and models that can create new, original content based on a given set of input data, such as images, text, or audio.)
Change is coming whether we like it or not. But, AI is not something to be feared. AI allows us to empower technicians in ways we’ve always dreamed possible. The danger of AI is not the elimination of the human element, but the AI failing humans with insufficient results. Once technicians begin relying on AI for answers in the field, the results they get will only be as valuable as the data powering the AI systems.
Above is a real-time look at the answer ChatGPT gives for a maintenance question. At first glance, this is extremely impressive, taking just over a minute to generate a comprehensive answer. However, because GPT is trained off public data, the answer is relatively simple, and does not account for brand specifics, advanced issues beyond surface troubleshooting, or variables such as model year, weather, time of year, or past maintenance history.
So, what constitutes a successful AI tool for field service? We believe success relies on tapping into highly advanced and robust data sets of collected jobsite data. Think about this through the lens of prompt engineering—the better the prompt you feed ChatGPT, the more nuanced response you’ll receive. Technician engagement and integration with a software like XOi continuously feeds an AI system to generate thousands of specific data points. Run that data through AI models with fine-tuned, field service specific prompts, and you get powerful, predictive answers related directly to the specific equipment and environment you are servicing.
The basis of the XOi software is built off data collection in the field. For 10+ years, technicians using XOi have been collecting video, photos, and notes from the jobsite — all content that can be transcribed to fuel the prompts that generate highly specific AI suggestions. Having historical data is just the beginning. The way we see it, truly successful AI tools will rely on a continuous collection cycle of engagement and reengagement from technicians in the field.
AI technology is moving at lightspeed. (So rapidly, will this blog even be relevant when you read it?) Our guiding light through this revolution remains steadfast — the technician is the hero. AI should be built to enhance an expert’s job with highly relevant suggestions, not replacements for the nuance that a tech brings to the job.
Field service AI technology needs the data collected by humans on a jobsite, and the function of AI should exist to make technicians better, faster, and smarter, not eliminate the need for the human element.
