O&G companies are being compelled to maximize their assets’ value and optimize investments, in order to enhance process efficiency and wisely decide when and where invest to reach their business goals. In consequence of such a scenario, companies need to apply new technical resources to accelerate the development of O&G exploration processes and adapt to making decisions in situations characterized by uncertainty. In this sense, Artificial Intelligence (AI) and Data Science in general have demonstrated to be a mighty technology in supporting complex and complicated tasks. This paper describes some AI-based applications tailored to specific needs of O&G industry processes, highlighting some research and development efforts that the authors’ institution has made in the field of predictive intelligence and which might meet those needs. For instance, machine learning and deep learning applications to identify engineering assets in degraded conditions, from images captured by drones; machine learning or PCA (Principal Components Analysis) techniques to make predictive models more easily treatable in order to identify failure patterns of key equipment subjected to stress and unforeseen conditions of use; and approaches to cope with uncertainties in asset operation and maintenance, including data elicitation with experts and probabilistics models. The computational intelligence thus employed for predictive purposes can bring insights in situations of uncertainty – or when data is unavailable – and consequently support organizational decisions, since the recent complex processes impose hard challenges and huge costs to O&G companies in exploring efficiently the resources and maintain the production goals.