Artificial intelligence for energy leaves experiments: trust requirement imposes itself on infrastructure
Listen to this article
Read by Anchor
When energy companies move from testing a smart tool to embedding it in daily operations, the question is no longer just about the model’s capability; the harder question is who owns the data, where it is processed, and who bears the consequences of error in an infrastructure that cannot tolerate downtime. This was the thread that linked the discussions on the third day of the Adibk 2026 exhibition and conference, where artificial intelligence (AI) applications sat alongside digital infrastructure and energy flexibility.
The message that emerged in the sessions was not a celebration of a new experiment but a shift from trial to large-scale deployment. Speakers tied the success of this shift to reliable data, digital infrastructure, effective governance, cybersecurity and investment, conditions that are less flashy than a robot or a new model but that determine whether the technology remains a limited showcase or becomes part of an industrial plant’s operation.
The value lies in the process, not the façade.Helmut von Stroh, chief executive of Siemens in the UAE and the Middle East, said the greatest value of artificial intelligence appears when it improves core industries and services by raising productivity, reliability and resource-use efficiency. In the energy sector, this vision drives applications linked to reservoir modelling, autonomous operations, network management and predictive maintenance, rather than adding a separate digital assistant to operational decision-making.
Other sessions linked transparency, digital technologies and data sharing on one side with stronger industrial partnerships on the other. But data sharing is not a call for unrestricted openness. John Gilly, chief executive of Kint, described the intersection of data demand and energy demand, noting that every country wants control over its data and its energy. At this intersection, data sovereignty shifts from a broad political term to a design issue: who determines access to operational data, and how critical infrastructure remains protected when linked to platforms and models.
Trust is the next operating layer.In a discussion dedicated to energy data sovereignty, Denis Joule, chief executive of AIQ, stressed that trust will dictate the speed of AI adoption. He presented the UAE system as layers of energy, data, infrastructure, sovereign cloud and models governed by security controls, emphasizing that data should remain the client’s property. This is not a belated legal add-on but a prerequisite for operating a system that relies on sensitive data about fields, refineries and networks.
At the same time, the debate did not portray automation as a direct substitute for humans. Ashish Koshi, chief executive of Inception42, said the aim is human expertise augmented by technology. This distinction matters in sectors that make decisions based on field data: relying on intelligent systems expands teams’ analytical capacity, but it does not eliminate the need for someone to review context and assume responsibility.
Investment is inseparable from broad acceptance.The sessions also addressed rising energy demand alongside expanding AI and digital computing, and the need to win community support for new infrastructure projects. From Egypt, Tarek Al-Malla said attracting private capital to these projects requires more supportive policies and incentives. Therefore, buying computing capacity or launching a data platform is insufficient; the underlying infrastructure needs financing, flexibility and regulation that open the door to investment.
For workers in the Gulf and Egypt, the practical change is that AI skills in energy will be measured by their ability to connect data, security, engineering and governance within a single workflow. Demonstrations that included reservoir simulation, process analytics, pipe-inspection robots and power-plant stations indicate that competition is no longer about an isolated model; it is about a trusted system that can operate near critical assets without the owner losing control of their data or decisions.