HOUSTON: Oil companies are reconsidering sites once deemed too costly or challenging to develop as artificial intelligence speeds up oil and gas drilling, executives said at the CERAWeek conference in Houston.
With AI cutting time and costs, US oil producer Devon Energy’s chief technology officer Trey Lowe warned that companies failing to adopt it “will get left behind at this point”.
Reuters reported that AI was a key focus in many sessions at the world’s largest energy gathering, as producers look for ways to stay profitable amid falling oil prices and concerns that US President Donald Trump’s tariffs could weaken global energy demand.
Major UK oil firm BP’s senior vice president of wells, Ann Davies, said the company has been using AI to guide drilling and detect issues in wells before they occur, allowing more wells to be drilled each year while improving capital allocation.
Last month, BP announced plans to increase spending on oil and gas production to boost investor confidence.
US oil producer Devon Energy also used AI to drill in areas that were previously “unfeasible”, said Mr Lowe, adding that AI helped the company gather information on underground faults and adjust drilling locations.
In Texas and Colorado, Chevron introduced AI-powered drones for monitoring shale operations, extracting oil and natural gas from shale rock formations, remotely detecting emissions leaks, and alerting field workers.
After three months of testing drones with autonomous drone company Percepto, Chevron reduced downtime for repairs and maintenance, said Russell Robinson, deputy programme manager of facilities and operations. He added that the drones also cut inspection time, allowing more assets to run longer and boosting oil and gas production.
The company is now considering expanding drone use to monitor its refineries.
Mr Lowe said Devon Energy used machine learning models to monitor its oil rigs in the US, resulting in a 25 per cent improvement in the productive life of its oil and gas wells.
In offshore drilling, BP sped up seismic data processing in the Gulf of Mexico from six to 12 months to eight to 12 weeks with AI, helping geoscientists determine where to drill and predict difficulties, a spokesperson said.
The oil and gas industry has used AI for years, but new advancements like large-language models are transforming the sector, said Chicheng Xu, founder of OpenPetro AI and a former petrophysicist at Aramco.
He explained that AI can quickly analyse data and create 3D visualisations of deep-sea features, a process that would take much longer for humans. /TISG
Read also: Gen Z drives GenAI adoption in Southeast Asia and Hong Kong, outpacing global average
Featured image by Freepik (for illustration purposes only)