AleaSoft Energy Forecasting, June 16, 2026. Artificial Intelligence is now commonly associated with generative models, conversational assistants and the growth in electricity demand linked to data centres. However, several decades before these technologies emerged, Artificial Intelligence had already begun to demonstrate its ability to solve complex problems with a real economic impact. One of the most notable examples was PROSPECTOR, an expert system developed in the 1970s to support mineral exploration. Its success made it one of the first documented cases in which Artificial Intelligence contributed to decision‑making under uncertainty and generated tangible economic value.
The first expert systems and the emergence of a new way of analysing information
During the 1970s, Artificial Intelligence was at a very different stage from where it is today. The available computing capacity was limited, and computing resources were modest by current standards. It was in this context that expert systems emerged: programs designed to reproduce part of the reasoning of human specialists through logical rules and probabilistic models.
One of the most representative developments of that period was PROSPECTOR, a system created to assist geologists in mineral resource exploration. Its purpose was to analyse geological, geochemical and structural information in order to estimate the probability of finding particular mineral deposits, combining multiple sources of information and partial evidence.
Unlike today’s models based on Machine Learning or neural networks, PROSPECTOR operated on the basis of expert knowledge manually incorporated into the system. Its ability to combine incomplete information and assess different scenarios made it one of the first practical examples of probabilistic reasoning applied to real‑world problems with a significant economic impact.
The case that made PROSPECTOR a historical reference
PROSPECTOR’s significance extended beyond the academic sphere when its analyses identified a high probability of finding a molybdenum deposit in the Mount Tolman area of the United States. Subsequent drilling campaigns confirmed the existence of the deposit.
This result attracted considerable attention because it represented one of the first documented examples of an Artificial Intelligence‑based system contributing directly to a decision with a tangible economic impact. The technology was no longer confined to experimental environments and had demonstrated its usefulness in real‑world applications involving resource management and business decision‑making.
The PROSPECTOR experience showed that combining specialist knowledge with computational models could generate highly valuable information for addressing problems in which data were incomplete and uncertainty was a determining factor. This principle remains the basis of many modern forecasting systems, which are now used to analyse complex scenarios and support strategic decisions across multiple sectors.
From mineral exploration to advanced Artificial Intelligence models
Following the early successes of expert systems, Artificial Intelligence began to spread gradually into fields such as medicine, finance, industry and energy. Although methodologies have evolved significantly over recent decades, many of the fundamental principles remain the same.
Today, Artificial Intelligence models are capable of processing vast volumes of information from satellite imagery, sensors, meteorological data and historical records. Machine Learning and Deep Learning techniques can identify complex relationships and patterns that would be difficult to detect using traditional methods.
However, the objective remains similar to that which drove the first expert systems: to reduce uncertainty and provide decision‑support tools for complex environments. The difference lies in today’s capacity to process large volumes of information and continuously update models as new data become available.
Managing uncertainty as a common element
The history of PROSPECTOR illustrates a characteristic that remains essential in modern Artificial Intelligence applications. In many sectors, the main challenge is not simply having access to more information, but correctly interpreting incomplete data, interrelated variables and uncertain scenarios.
This need is particularly relevant in areas where decisions depend on multiple simultaneous factors and where small variations can have significant economic consequences. The ability to integrate expert knowledge, historical information and advanced analytical models remains one of the main drivers behind the development of Artificial Intelligence.
More than fifty years after the first expert systems, technological progress has multiplied computing and analytical capabilities. Nevertheless, the fundamental principle remains unchanged: using computational models to improve the understanding of complex systems, identify relevant patterns and support strategic decisions under conditions of uncertainty. This approach is also present in hybrid models that combine fundamental models, statistical techniques and Artificial Intelligence tools to analyse complex systems and anticipate their development.
AleaSoft Energy Forecasting’s analysis and forecasts on Artificial Intelligence and electricity markets
On June 18, 2026, AleaSoft Energy Forecasting will hold its 67th webinar, entitled ‘Energy markets in Europe in the second half of 2026: evolution, prospects and opportunities for renewable energy, PPA, batteries and hybridisation’. The session will feature Daniel Fernández Alonso, Strategy, Regulatory Affairs and Communications Director at ENGIE Spain, and Oriol Saltó i Bauzà, Associate Partner at AleaSoft Energy Forecasting. The discussion panel of the webinar in Spanish will also bring together Lola López Serrano, Head of Strategy and BESS RES at ENGIE Spain, and Juan Noguera, Head of Acquisitions, Investments & Financial Advisory (AIFA) at ENGIE Spain. The webinar and discussion panel in Spanish will be moderated by Antonio Delgado Rigal, CEO and founder of AleaSoft Energy Forecasting.
The webinar will analyse recent developments in European energy markets and the outlook for the second half of 2026, with particular attention to the factors that will continue to influence market prices, fuel prices and CO₂ emission allowance prices. It will also address the opportunities and challenges facing the renewable energy sector, the current situation and trends in Spain’s PPA market, and the outlook for battery energy storage and hybridisation with solar photovoltaic energy.
In an environment characterised by growing complexity and the need to interpret large volumes of information, Artificial Intelligence, statistical models and hybrid models are playing an increasingly important role in electricity market analysis, scenario modelling and investment assessment. AleaSoft Energy Forecasting’s accumulated experience in developing advanced models enables it to provide market forecasts, scenario analyses and decision-support tools for renewable energy projects, energy storage, PPA contracts and risk management strategies in European electricity markets.
Source: AleaSoft Energy Forecasting.