Repsol: "We want to lead generative AI to guarantee its benefits and avoid risks"

In 2022, Repsol completed its first wave of digitalization. In five years, the Spanish energy company had carried out more than 500 initiatives with an impact on its income statement of 800 million euros of annual free cash flow. During this period, Repsol launched programs such as Pan Caliente, which predicts the demand for bread in station stores; In-Well, a global platform that generates real-time recommendations of the best drilling model for a specific subsurface region, or Vivit, a solution for managing energy services in homes.

Once this first digital wave was completed, Repsol decided to immerse itself in a new stage of transformation, a second stage that spans from 2023 to 2027 and which has become a fundamental lever of the company's strategic plan for this period of time.

"The idea in this second wave is to be able to achieve the same impact that we had in the first, those 800 million euros per year of free cash flow. To achieve this, we are going to make an investment of approximately 130 million euros per year in digitalization and one of the main protagonists will be everything related to generative AI", explains Juan José Casado, chief digital

Until 2027, the person responsible for digitalization of the Spanish company assures that in addition to looking for new initiatives and use cases, the focus will be on people, turning them into a digitizing agent.

"The great challenge at this moment is to provide workers with the necessary skills and training so that they can digitalize all the processes they carry out on a daily basis. And there we are making a very strong investment in training our teams," says Casado. "But if you really want to scale digitalization through people and with the focus on them, you have to give them tools, self-service solutions so that they can exercise that digitalization themselves."

Among these tools and solutions to which the manager refers, Repsol has focused on data, no-code solutions and generative artificial intelligence.

"We are offering a great data portal so that everyone has quick and direct access to the data they need for decision-making. We are also highly encouraging the development of do-it-yourself solutions, that is, non-tools. code so that people can make their own applications of the things they need on a daily basis. And the third tool that all employees will have access to and that will help us scale digitalization and, above all, will. generate a paradigm shift in the productivity of companies is generative AI," he adds.

Approximately 60% of the 500 initiatives that the company carried out in the first wave of digitalization were based on data, algorithms and artificial intelligence.

"We are convinced that artificial intelligence is one of the keys to the competitiveness of the Spanish industrial fabric, and that is how we want to promote it. If you think about the industrial world, the investments made are very large, so any process that you are capable of efficiency through algorithms and AI will have a very high impact on efficiencies, and therefore, on the income statement directly," says Juan José Casado.

Proof that artificial intelligence is very present at Repsol are the loaves of bread that are dispensed at its service stations, based on a stock prediction algorithm that allows the company to sell more and be more efficient.

But beyond traditional AI, Casado has seen a great opportunity in generative AI. "Generative AI is going to be one of the key levers for us, and that is why we were pioneers in the energy sector by creating the first Generative AI Competence Center. We are trying to lead that artificial intelligence to understand very well how we can guarantee the great benefits that it will bring and be sure to minimize any risk," explains Casado.

Repsol's so-called Generative Artificial Intelligence Competence Center has 50 professionals dedicated to identifying how this technology can create value in business, promote new ways of working that facilitate employees' daily lives, and capture productivity improvements.

Likewise, the energy company has carried out a pilot by installing Copilot for Microsoft 365 to 300 employees and intends to extend its use to more workers after measuring the improvements in productivity resulting from the technology giant's generative AI.

Likewise, work groups have been organized in which employees have proposed more than 250 use cases in which this technology could make their daily lives in the office easier. "This is an example of how we are starting to work with generative AI focusing on people, with ideation sessions in which employees from different areas participate and they are the ones who tell us how they think it can help them in their day. "We have more than 200 ideas collected and there are already more than 50 use cases that are being developed in all business areas," explains the head of digitalization at Repsol.

A practical example already in use of generative AI in the company is the launch of Harvey, a program implemented in the legal area of ​​Repsol capable of analyzing legal documents, comparing contracts or regulations, and proposals for improving clauses. , translation of texts into legal language, providing answers to legal queries or summary of documentation. Its use plans to make work more efficient in the legal area, where they already calculate that they are saving about three hours of work per week.

"At the moment, we are focusing a lot on training and equipping our employees with generative AI, although we also use it to improve our communications with clients. We are learning from this technology and seeing how it can serve our employees, but it is only the beginning," says Juan José Casado.