This article is the first of a three-part series on how artificial intelligence is already shaping the global energy system and could revolutionize it in the future.
Of all the intriguing new technologies in the works today, artificial intelligence (AI) may become the largest disrupter of the global energy system. AI’s ability to aggregate, sift through, and analyze data is already changing the business model for energy producers and could revolutionize how energy is delivered and consumed in the future. In 2017, Bill Gates advised college graduates to enter one of three fields: AI, energy, or biosciences.
More than any area, energy will affect the future of local communities, states, and the planet at large. AI will make energy markets smarter, cheaper, and more efficient, but the question remains: will AI simply reinforce old energy habits or nurture new ones?
At this stage, AI appears to be reinforcing the old habits, namely continued reliance on fossil fuels for over 80% of global consumption. Predictive analysis and machine learning are making it cheaper and more efficient to explore for, produce, manage, and transport oil, gas, and even coal, though the real gains are in oil. Last year, McKinsey estimated that the oil and gas industry could save up to $50 billion in upstream activities through improved use of data aided by AI.
It is unclear at the moment whether supply-side cost reductions will translate to advantages on the demand side. They will certainly make oil and gas more competitive compared to other energies, and consumers have long chosen fossil fuels because of their greater thermal content per cost. Forecasting from what we know, AI seems perfectly aided to extend the status quo.
Undeniable upside upstream
The exploration and production (E&P) of oil and gas, also known as the upstream, is one of the most capital-intensive activities humankind has ever undertaken. With such high up-front investment costs, companies always seek to cut costs and, more importantly, make the right decisions about where and when to drill.
By looking at reams of geological data, forecasts for future demand both globally and locally, and even political and legal frameworks, AI can make better decisions about field selection, decisions that have often relied on foresight or blind luck in the past. Test or pilot wells will still be necessary to determine the specific pressure, volume, temperature, and permeability characteristics, but AI will allow companies to better understand what they are dealing with. British Petroleum (BP), for instance, is using marine autonomous systems (or drones) to better understand its underwater operating environment in its offshore fields in Brazil.
Fuzzy logic machine learning can also increase productivity once a field is operational. Remote sensors connected to wireless networks at oilfields can collect and transmit data that AI can use to understand a particular reservoir’s characteristics. Since drilling is highly repetitive, data is accumulated quickly. AI can detect variations across numerous engineering-control applications in the oil production phase, including enhanced recovery techniques, well stimulation, and infill drilling.
Finally, AI will replace human jobs and cut costs for producers. One might opine that these job losses would prompt a new generation of Luddites, but the labor force in oil and gas has always been small, itinerant, and politically weak. Moreover, AI’s ability to separate variables among mountains of data will prevent mistakes, errors, and accidents and create a safer, albeit smaller, workplace. Labor poses no obstacle to the adoption of AI.
A more manageable midstream
Agglomerating and deciphering data could have an equally profound effect on oil and gas storage and transportation, whose lack of transparency has long created market inefficiencies between global supplies and demand and produces price volatility.
AI is, in fact, already starting to aggregate global oil inventories, forecast demand and supply shortages, and streamline supply chains. U.S. geospatial analytics company Orbital Insight has already begun using a form of AI – convolutional neural networks (CNN) – to analyze satellite images and quantify the levels of crude oil in storage tanks. Since storage tanks have floating roofs, the technology perceives the level of shadow in the tanks to calculate the volume inside.
Transportation needs will also be easier to forecast with AI. In recent years, a group of energy analysts under the Twitter hashtag OOTT, the Organization of Oil Trading Tweeters, has been using satellite imagery to chart the movement of oil tankers. AI can easily replace these admirable but unwieldy efforts, and these midstream improvements will invariably trickle downstream. Greater automation, lower labor costs, and improved safety and efficiency of delivery will be seen in lower prices at the pump.
The shadow of secrecy
Private oil and gas companies are notoriously secretive, intensely guarding data and company intelligence from their competitors. Since AI is more powerful with more data, its application will favor large international oil companies (IOCs) who work across multiple jurisdictions and national oil companies (NOCs) who know their own fields well from years of experience. Large IOCs and NOCs are also the most capitalized, and therefore the best positioned to invest in AI.
It is possible that strategic secrecy might give way under the stress of lower prices, declining demand, and greater competition from renewable energies. Many major oil companies have forecast the world will reach peak oil demand in the next five to fifteen years. Large producers, moreover, have a long history of collusion – see the Seven Sisters from the 1930s to 1970s and OPEC countries since the 1970s. In the early 2000s, several majors joined forces to become the supermajors, including Exxon-Mobil, BP-Amoco, Total-Elf, Chevron-Texaco, and Conoco-Philips.
The threat from renewables and alternative transport fuels could prompt greater consolidation and cooperation. AI has the potential to spur additional consolidation in the industry in the coming decades.
Innovate or die
The oil and gas industry was built on adaptation to technological breakthroughs. From liquefied natural gas (LNG) to offshore drilling techniques to the more recent shale technologies, including horizontal drilling, hydraulic fracturing, and 3d seismic imaging, the industry often pivots on a dime. Innovation is part of its DNA. Exxon Mobil helped found MIT’s Energy Initiative, committing $25 million over five years to support research on the subject. Total and Google signed an agreement to jointly develop AI solutions for subsurface data analysis earlier this year.
The challenge facing the industry is not running out of oil or natural gas – it is whether its product will remain more desirable than newer, cleaner sources. Most of us hope renewables win out, and in my next piece, I will explore how AI’s potential on the demand side for renewables exceeds that of hydrocarbons. That said, it has always been unwise to underestimate the ability of the oil and gas industry to adapt and even collude when it faces existential threats.