We use cookies to personalise content, to provide social media features and to analyse our traffic. We also share information about your use on our site with our socal media and analytics partners who may combine it with other information that you've provided to them or that they've collected from your use of their services.
Accept all
Customize
Please check an answer for every question.
We use cookies to personalise content, to provide social media features and to analyse our traffic. We also share information about your use on our site with our socal media and analytics partners who may combine it with other information that you've provided to them or that they've collected from your use of their services.

04/03/2019 News

Artificial Intelligence And The Energy Sector: Huge Potential, Tough Questions

Artificial Intelligence And The Energy Sector: Huge Potential, Tough Questions

A new industrial revolution

When did you start thinking about artificial intelligence (AI) in your daily life? The arrival of Apple’s virtual assistant Siri, back in 2011, may come straight to mind, followed a couple of years later by its counterparts, Amazon’s Alexa and Google Home.

Those working for the energy sector have long been aware of the power of artificial intelligence, with applications that extend well beyond knowing when to turn your lights off. From the 1980s on, our industry has tapped into what can be considered a form of AI, with its “expert systems” helping geologists find oil and gas.

Artificial intelligence has clearly seen several years of exponential development, thanks to the abundance of data and the massive reduction in the cost of computing power — helped along, of course, by the internet giants.

It's important to remember that artificial intelligence is a mathematical program of analysis and optimization of data. It is undoubtedly very powerful, but ultimately cannot compare with how the human brain operates. A simple way to explain this is that in order to create a powerful program (deep learning or neural network) capable of recognizing cats, the AI algorithm must be provided between 100,000 and 1 million images of cats. In comparison, a human baby is able to recognize a cat after having seen just two or three images.

Furthermore, artificial intelligence programs have no notion of good and evil, and are unable to set their own goals or to be inherently innovative or creative. In order for artificial intelligence to be able to produce a painting in the style of Van Gogh or a sonata that sounds like Mozart, the algorithm would first need to be nourished by the artist’s works in order to copy their creative scheme. And yet none of this is true creation.

Still, there is no denying that there are applications of artificial intelligence at all levels of our economy and society. This was a key topic of my discussions and curiosity during my visit in January at the Consumer Electronic Show (CES) in Las Vegas, along with members of Total’s executive committee as part as of our annual innovation week. Among the many applications of artificial intelligence, I highlighted self-driving cars, AI in health care, and even real-time audio translation. A new industrial revolution is underway, and everyone wants to be part of it.

The energy sector, an ideal field of study

It’s also key that the AI experts work together with those who will apply their tools in the real world. The energy sector is an ideal case study to see how this can happen. It's important that a company such as Total investigates the possible applications of AI to determine in which domains this technology corresponds with our future strategy. Due to its activities requiring the use of huge amounts of geophysical data processed by supercomputers, Total has a culture conducive to boarding the AI train quickly.

At Total, we have recently reached an agreement with Google that allows our reserve geologists, geophysicists and engineers to work with members of Google Cloud’s machine-learning team to refine the process for interpreting images from below the ground and the automation of analysis of technical documents. Eventually, our geoscience engineers will have an AI personal assistant that frees them to focus on the other critical tasks of their profession.

Total has also recently signed a partnership agreement with Tata Consultancy Services to create a digital innovation center in India to focus on refining processes. Artificial intelligence will be used to improve industrial efficiency, energy performance and availability rates. With this digital innovation center, we will help pioneer the development of a smart, connected refinery that will allow us to strengthen our industrial competitiveness. Make way for Refinery 4.0!

Training for good cooperation

Like any technology, AI has both positive aspects (health, safety, taking care of repetitive or tedious tasks…) and more worrying aspects (risks of infringement of individual liberties in particular). The changes emerging are potentially destabilizing and require us to approach them with serenity and pragmatism, rather than alarmism or cynicism. We must take into account the reality of the development of artificial intelligence with a clear mind.

Every leader must pose questions about the impact of artificial intelligence on their sector of activity, open opportunities for the company to improve efficiency, create new services, and simplify the lives of its employees and customers. They must also consider whether the company has the right skills and adequate training programs. At Total, for example, we have set up a successful program to raise awareness about artificial intelligence among managers.

Both inside and outside the energy sector, the nature of this particular technological disruption requires extra attention to what’s coming next. What are the ethical questions bound to arise? How can we support the development and large-scale application of AI without the risk of losing control?

My belief is that both governments and private companies should continue to invest in artificial intelligence, but do so in a way that is fully conscious of the human implications. One way would be to apply some of our mathematical fire power to modeling the possible consequences of AI on our economic and social future.

Ultimately, it's about clearly defining the roles that AI will bring to current and future jobs and lives. We shall let the machines do what they do better than us: manage and analyze the data, propose options, and devise solutions. If done this way, women and men stay at the heart of the production process, ready to make the decisions that cannot be automated. AI may become history’s most powerful tool, a precious accessory to the human mind — never its rival.