First Break           EarthDoc           Learning Geoscience

EAGE AI Committee - Newsletter on AI, ML and all things Digitalization #1/2023

January 2023

Happy 2023! For their first newsletter of the year, the EAGE A.I. Committee is sharing their highlights of 2022 and wishes you a fantastic year ahead!


As a group of EAGE members and volunteers, the EAGE A.I. Committee is dedicated to helping you navigate the digital world and finding the bits that are most relevant to geoscientists.

You are welcome to join EAGE or renew your membership to support the work of the EAGE A.I. Community and access all the benefits offered by the Association.

EAGE Membership Benefits: Join or Renew


Curious to know all EAGE is doing for the digital transformation? 

Visit the EAGE Digitalization Hub



Data Science Highlights in 2022

Contributed by: Ashley Russell

Outside of the energy industry I was quite impressed by some of the improvements in natural language processing applications which are getting closer and closer to “human” conversation. Humor is one of the most human things that of course a machine should not be able to grasp – and no they still can’t tell jokes, but Google this year did release PaLM which does a quite good job of explaining jokes. Read more here.

In the same vein, we all got psyched out this year when a Google machine learning engineer thought he had created sentient life from his chatbot. Regardless, the technology on understanding and replicating human communication is advancing quickly.

Within the energy industry, it is clear to me that a lot of focus is on industry-wide data efforts – from OSDU and the Industry 4.0. There are has been major advancement in both this year, including implementation of OSDU into working oil and gas assets. Although this is not a data science story – the capability to do data science across our businesses is only enabled by quality labelled standardized data.



A Journey in Machine Learning from 2022 and a Path Forward for 2023

Contributed by: Jan van der Mortel

Highlight is in short, starting to build a better understanding of probability, both in concept and some of its’ many applications in Machine Learning.

Without going in too much detail, the plan forward is to build out this knowledge the coming year and beyond, with particular attention to recent developments in normalizing flows, and adding graph type networks into the mix.

Although (or because of) many of the basics being a bit harder to fully grasp, I believe this can be an underpinning of many future developments / advancements in Machine Learning.




Highlights in AI in 2022

Contributed by: Lukas Mosser

2022 has been a transformative year for machine learning, and AI.

My personal highlights from this year have been diffusion models, which have completely redefined what is possible with generative models, and how they can be integrated with other modeling approaches such as natural language processing.

Here you can find the in my opinion three most important advances in diffusion models this year:


My personal second highlight came very late in 2022, which is ChatGPT. When I first started interacting with the system I was truly speechless. Of course, this is a language model, of course this is still predicting the most likely word given the context, but the fact that a system can be created that interacts in such a natural manner, with many reasonable and potentially useful insights to be gained, really changed my personal outlook on the whole field. In a positive way, seeing the potential benefits, and also in a negative, seeing the potential unexpected harm that can be achieved.
If you haven’t tried it, I highly suggest you give ChatGPT a spin. Ask it about some subsurface topics; you’d be surprised to see just how deep the rabbit hole goes.



Discover EAGE Learning Resources on A.I. and machine learning

EAGE Digitalization Conference 2023


This newsletter is edited by the EAGE A.I. Committee.

NameCompany / InstitutionCountry
Ashley RussellEquinorNorway
Jan H. van de MortelIndependentNetherlands
George GhonCapgeminiNorway
Cédric M. JohnQueen Mary University of LondonUnited Kingdom
Lukas Mosser Earth Science AnalyticsAustria
Oleg OvcharenkoNVIDIAUnited Arab Emirates
Nicole GrobysWintershall DeaGermany
Robert FergusonUniversity of CalgaryCanada
Ruslan MiftakhovGeoplatUnited Kingdom
Surender ManralSchlumbergerNorway

Leave a Reply

Your email address will not be published. Required fields are marked *


Siddharth Misra


Prof Dr Siddharth Misra’s research focuses on improving subsurface characterization and prospect evaluation for the exploration of hydrocarbons, minerals and water resources.

His major contribution is in the theory of electromagnetic responses of geological formations to various charge polarization phenomena. The theory has enabled him to introduce a multi-frequency electromagnetic log-inversion technique to remove dielectric effects for improved estimation of hydrocarbon pore volume.


           First Break           EarthDoc