The Oil and Gas Chat Bots Are Coming

Abstract

Meet Sandy, Nesh, and Ralphie. They are newborn chat bot programs that have been designed specifically to seek out the answers to oil and gas professionals’ tough questions. These bots, also termed virtual assistants, stem from a branch of artificial intelligence (AI) known as natural language processing (NLP), which has quickly entered the mainstream thanks to the efforts of tech giants Amazon, Apple, and Google. Their software innovations have enabled many millions of people to engage in dialogue with laptops, smartphones, and speakers. Consumers are using NLP mostly inside their homes for simple requests; music and weather reports round out the top two uses in one recent survey. But there is now a push to get this technology into the world’s offices where it has the potential to increase worker efficiency. This market test is just barely under way in the oil and gas business where adoption will hinge on a virtual assistant’s ability to quickly generate reliable assessments of complex issues involving reservoirs, seismic data, and well logs. And like many nascent developments in the digital arena, the technology’s capabilities will also depend on the amount of training it receives from its earliest users. These first movers are among those vying for the chance to make chat bots an essential part of the upstream sector’s future. Nesh Houston-based startup Nesh has created a virtual assistant by the same name to help industry analysts and engineering techs build intelligence reports. The firm is in the middle of pilot projects with a European operator and one of the largest US shale producers. Sidd Gupta is the chief executive officer and started the company last year after spending most of his career with Schlumberger, including a stint on one of its software teams. Gupta acknowledged that chat bots have been criticized in the past for being “dumb things” that can only complete a rigid set of pre-programmed steps. Next-generation chat bots like Nesh aim to reverse this perception with more advanced programming that fetches data for inquisitive users from multiple, usually disconnected, sources. Gupta drew inspiration for the new venture from a few sources, but one in particular exposed the need for a chat bot technology that could make using complex oil and gas software a more intuitive experience. After a friend was laid off during the recent downturn from his role as a reservoir engineer, Gupta said he was turned down for a job that seemed like a perfect fit, with just one exception: the friend was trained on a different reservoir simulator than what was used by the prospective firm. “I thought that was a very weird reason to reject someone for a technical role,” he said. “I realized that this was a very systemic issue in the oil and gas industry—we judge people’s technical capabilities based on what software they can use.”