The Future of AI in Exploration

AI is quickly becoming a valuable and game-changing tool for the data science and geoscience industries. These next generation technologies will increase efficiency giving scientists the time to focus on interpreting that data. Currently, these professions spend about 80% of their time loading and organizing data. AI can facilitate labeling and organizing of data, however, the data needs to be properly collected and stored first. By establishing better methods of data collection, storage, and organization, AI can be more effective. Here are just a few of the ways that AI is set to change exploration.

Document Automation & AI

There are millions, if not billions, of legacy and exploration documents that exploration companies own and acquire to do their business. AI can automate and unlock the searchable content based on domain centric search of all facets of information in these documents. This means geoscientists no longer need to look for content manually — instead they will have a Google-style search for subsurface information at their fingertips.

Well Log Automation

There is immense data from newly drilled and historical wells, but we only evaluate a handful of wells to understand the subsurface. By building an automated, end-to-end rock physics and petrophysical suite, geoscientists can pre-process, merge, and classify the world’s well databases. This use of AI, when coupled with three-dimensional volume data, will unlock new capabilities and depth of understanding.

Seismic Segmentation

When we segment different layers of the Earth, we get the opportunity to see and interpret what is invisible to the eye. Automated seismic interpretation is a cornerstone of subsurface feature extraction. Using machine learning to infer system depositional and tectonic features is driving exploration work into a new generation.

Studio X & AI

Studio X is working to advance AI in exploration. Through Xeek, our machine learning crowdsourcing community, we provide challenges where contestants build efficient solutions for the industry, including AI-based solutions. Xeek’s initial toolset, to be launched to a Beta audience in late 2021, will include tools such as a tool based on the  “Go With The Flow” Challenge, an agent-based model that shows fluid movement through the subsurface in a visual format, and a  tool based on the MinesFault Challenge, which creates 3D visualizations of faults on a number of different parameters. Xeek is continuing to develop and release tools throughout the year, and geoscientists can also suggest tools they would like to see developed.

We are always looking for partners and participants to join us as we carve new pathways for how the energy industry works. By providing technology, community, and resources, we can take bold ideas further, faster.