ChatGPT’s performance on Oil & Gas problems

“All models are wrong, but some are useful.” -- George Box, British statistician

ChatGPT has burst upon the scene like an oversized balloon. It is even being used in the oil and gas industry, as described in the open access article “Industrial Engineering with Large Language Models-A case study of ChatGPT’s performance on Oil & Gas problems” (https://arxiv.org/abs/2304.14354 )
Written by Oluwatosin Ogundare, Srinath Madasu, Nathanial Wiggins, of California State University, San Bernardino, IBM Technology, and University of Houston, respectively, it discusses the potential and the limitations of ChatGPT in the oil and gas industry.

To give you a taste, here is the article’s conclusion …

///////
Conclusion
Large Language Models (LLMs) have the potential to be useful in industrial engineering, particularly oil and gas engineering. This paper identifies some of the limitations of current LLM approaches, particularly ChatGPT, in solving problems in the oil and gas industry. The potential applications of LLMs in solving problems in various areas of oil and gas engineering is demonstrated with examples from rock physics, but generally includes other important aspects like Full Waveform Inversion (FWI), acoustic reflectometry, hydrodynamic pressure pulse reflection, and well intervention. Areas for improvement are suggested, including improving the nature of the data used to train LLMs, and enriching the output of LLMs with domain-specific knowledge which in many cases involves imposing physical constraints on the output.
///////
Google® Better!
Jean Steinhardt served as Librarian, Aramco Americas (https://americas.aramco.com/ ), Engineering Division, for 13 years. He now heads Jean Steinhardt Consulting LLC, producing the same high quality research that he performed for Aramco.

Follow Jean’s blog at: http://desulf.blogspot.com/ for continuing tips on effective online research
Email Jean at jstoneheart@gmail.com with questions on research, training, or anything else