All these cases have in common that the benefit that human beings can receive thanks to these practices is irrelevant or non-existent. Despite this, a large number of non-human animals are subjected to serious damage. Now, reasoning ethically requires rejecting all forms of discrimination based on the characteristics of individuals that have nothing to do with their ability to be harmed or benefited. Not taking into account the damage that animals suffer in these experiments, or giving them a minor importance, simply because they do not belong to the human species, is a type of arbitrary discrimination, speciesism. Just as not respecting someone for their skin color or gender is unjustified, so is not respecting him for his species. From an impartial point of view, the suffering and death of these animals outweighs the trivial benefit obtained by human beings.Experimentation with animals for environmental, cosmetic or military purposes should, therefore, be rejected. Fortunately, part of these practices are already being banned in some legal systems, as is the case of experimentation for cosmetic products in the European Union or India.Second, a minority of experiments with animals does have a biomedical character. This is compatible, however, with not all of them seeking to alleviate or cure serious ailments. Given the damage suffered by the animals with which it is experienced, neither would be justified practices. It must be admitted, however, that a part of biomedical research does have the purpose of eliminating serious ailments and increasing the quality and duration of human life. In these cases we must compare the suffering and death that is caused to non-human animals with the great benefits that some human beings would obtain in case of success in the investigation. The fact that the benefits to humans are not trivial can lead one to believe that in these cases, unlike the previous ones, animal experimentation is justified. This would be, however, an errorExperimentation with animals was only extended and standardized in the modern biomedical community during the 30s and 40s of the last century. For ethical reasons, it was intended to avoid clinical trials that subjected human beings to a risk of damage that was too high, while at the same time preventing the clinical use of treatments that were not duly tested. Given the state of scientific knowledge, it was believed that the similarities between nonhuman and human organisms, in spite of their differences, were sufficient.Thus, it was thought that it was possible to predict the effect in human patients of, for example, a drug, from its effect observed in clinical trials with other animals. Under this assumption, legal systems usually require trials with non-humans before making them in humans, and as a requirement for those who investigate receive public assistance. These are some of the factors that explain the current predominance of this model. However, there are strong reasons, based on evidence that we did not have in the past, to question the scientific value of animal experimentation, particularly in comparison with other methods.We now know that the similarities between nonhuman and human organisms are much smaller than previously thought. Those who advocate experimentation with animals indicate that, although methods such as tissue culture tests are useful, it is necessary at some point to also pass tests on non-human animals because these are models that show how an organism as a whole works . But the fact is that the organisms of human beings do not react like those of other animals to different medications. Even small genetic differences between individuals of different species can cause large differences in how they process chemicals.Therefore, there is no guarantee that test results in non-human animals can be extrapolated successfully 1. They do not allow us to reliably anticipate what the reaction will be in humans. This is verified when it is observed empirically that the predictions made by this method are correct in a very low percentage 2. In the US alone, 96% of drugs that successfully tested animals failed in human trials, because they were ineffective, harmful or both 3. Also, a large number of drugs marketed have been toxic in humans after passing tests with other animals, as was the case of thalidomide.The scientific value of biomedical experimentation with animals is, at least, considerably lower than many people assume 4. This has different implications. In the first place, it assumes that the human beings who participate in the trials of these drugs, and who consume them once commercialized, are exposed to receive damages not detected in the previous phase of experimentation with animals. Secondly, it prevents the development of treatments that would be beneficial for humans but for which some harmful effect has been detected in non-human animals. Well, indeed, there are potential medicines (such as aspirin) that, although seriously harmful, or even lethal, for animals, are not for humans.Currently there is a large number of research methods that do not use animals, such as the use of cell and tissue cultures, organ models or computational models. Despite the dubious scientific value of animal experimentation, however, the resources used to further develop these alternative methods are, by comparison, minimal. For every euro invested in them, several thousand are used in the promotion of animal experimentation. This includes the propaganda and lobbying expenses of the companies that benefit from it. Thus, for example, the pharmaceutical industry allocated only in its political relations with the EU in 2014 more than four times the total amount invested the previous year across the EU in non-animal methods 5. There is no justification for using all these resources to continue with these investigations when they could be used in potentially more efficient ways to improve human (and non-human) healthFinally, there is an additional implication of defending animal experimentation against methods that do not use them. If the only thing that matters to us is to obtain the greatest advances for human health, over ethical objections, then the method to be followed would not be the experimentation with nonhuman animals, but we should be willing to subject other human beings to such experiments, even against their will. At the end of the day, from a methodological point of view, there is no better alternative for biomedical research. Of course, this is ethically rejectable. Causing serious damage, against your will, to other human beings is not justified simply because it prolongs or improves the lives of others. Specifically, we would consider it unacceptable to cause such damage to certain human beings simply because their cognitive abilities are similar to those of nonhumans currently used in experiments.Now, if we reject speciesism, we can not believe that this type of research justifies causing serious harm to nonhuman individuals. As it has been explained, the mere belonging to a species is an irrelevant factor. Nor can intelligence be, since we also reject experimentation with humans with intellectual functional diversity. To ethically evaluate a practice, such as that of experimentation with non-humans, it is necessary to impartially consider the interests of all individuals affected by it, regardless of the species to which they belong. This requires rejecting all discrimination, including speciesismIn summary, the dilemma that must be confronted is the following: either we accept experimenting with sentient individuals, or we reject it and opt for other research methods. Under a criterion of mere efficiency, we should choose the first alternative, which would justify employing human beings in the experiments. Deciding ethically, however, forces us to choose the second one, abandoning experiments with non-human animals and investing in the development of other methods. Persisting in the situation, therefore, is ethically unjustified.