The paper will present several existent uses of big data in the criminal justice system, for example, for the prevention of payment card fraud by means of skimming; for the prediction of crime with predictive software; the use of algorithms to predict the recidivism of parolees. Such knowledge, built on a large amount of seemingly unrelated data, whose credibility is based on complex mathematical algorithms, may legitimise increased social control, limit privacy and undermine the basic principles of criminal procedure. Taking into consideration the benefits of algorithmic service, the paper will claim that our society will have to find a balance between the benefits of using big data and the disturbing effects it may pose for society (e.g. abuse through data fishing, loss and theft of personal data, etc.). It will present the pitfalls of reliance on big datapredictions used by law enforcement and criminal justice agencies and the risks big data carries as regards encroachment on fundamental liberties