Non-technical losses (NLT) constitute a significant prob-lem for developing countries and electric companies. The machine learning community has offered numerous countermeasures to mitigate the problem. Yet, one of the main bottlenecks consists of collecting and aecessing labeled data to evaluate and compare the validity of proposed solutions. In collaboration with the Uruguayan power generation and distribution company UTE, we collected data and inspected 311k costumers, creating one of the world's largest fully labeled datasets. In the present paper, we use this massive amount of information in two ways. First, we revisit previous work, compare, and validate e,arUer: fi.Jldings tested in Jiluch smaller and less diverse databases. Second, we compare and analyze novel deep neural network algorithm, which have been more recently adopted for preventhig NLT. Our main discoveries are: (i) that above 80k training examples, the performance gain of adding more training data is marginal; (ii) if modern classifiers are adopted, handcrafting features from the consumption signal is unnecessary; (iii) comple- mentary customer information as well as the geo-localization are relevant features, and complement the consumption signal; and (iv) adversarialattack ideas can be exploited to understand which are the main patterns that characterize fraudulent activities and typical consumption profiles.