Venoms of various animals, including snakes, scorpions, spiders, etc., are complex mixtures of predominantly protein and peptide toxins that have multiple, mostly adverse effects on living organisms. However, venoms or their components find applications in the pharmaceutical, agricultural and cosmeceutical industries. Due to limited sources and difficulty in obtaining, some venoms are very valuable. Smuggling of wildlife products, including animal venoms, is a global problem today and their illegal traffic is significant. Often it is not possible to prove that transported material is animal venom that leads to smugglers escaping responsibility. The development of method for quick and easy animal venom identification for distinguishing venom from a substance of different nature (not-venom) and for determining the venomous animal taxa is relevant task. Raman spectroscopy, as applied to mixtures of substances of protein nature, in particular, venoms, has such advantages as high information content and direct (no-preparation) measurement of dry samples. We have shown that Raman spectroscopy in combination with principal component analysis can provide discrimination between venom and some not-venom samples and also clustering of different animal venoms based on their similarity. The use of machine learning algorithms allows for the automatic identification of venoms and not-venoms among the substances studied. Proposed method looks promising from the point of view of on-site identification of animal venoms, for example, at airports and at border control.