many users to comment with hatred: the individuals who show violent tendencies and start from discriminatory beliefs (by race, sex, ideology or others) feel free on the Internet to express themselves, thanks to the anonymity the network offers them.

One of the main reasons for the spread of hatred speech is the performance (or lack of it) of the platforms. Several strata accuse social networks of passivity and believe that they should be ultimately responsible for filtering the content, and not allow messages inciting hatred to be published.

In this regard, it should be noted that, according to the PRISM study, only 9% of hatred messages on Facebook are removed by the social network (ReasonWhy, 2017) and, according to the European Commission report made after the creation of the code of conduct agreed by the major information technology firms last May to combat incitement to hatred on the network in Europe (European Commission, 2016), the percentage of content withdrawn by companies with respect to the total of alerts on social networks about those that incited hatred, it had a dangerously low rate: Facebook received 270 alerts and 28% of improper content was removed, Twitter received 163 alerts and 19% of content was withdrawn, and YouTube eliminated almost half of the content of the 123 notifications received.

The seriousness of this situation is best understood if we consider some of the effects of the hatred messages disseminated on the networks:

1. Suicide is the third cause of death from 15 to 29 years of age (INE, 2013), which is a figure of more than three hundred under thirty years of age each year. Among the reasons why this happens, the most indicated one is bullying, and especially, electronic harassment.

2. In 2014, the National Police registered in Spain 1,285 hatred crimes, 102 out of which were in the Community of Madrid. However, only one out of every 10 victims of hatred crimes comes to file a complaint (EFE, 2015).

3. Of the hatred crimes reported in Spain in 2014, 40% were related to insults derived from the sexual orientation of the victim, followed closely by those related to racism and xenophobia (Observatory against LGTBfobia, 2017).

4. According to the Report of the European Commission (El Mundo, 2016), out of the messages denounced for incitement to hatred, 23.7% were for their anti-Semitic nature, 21% for criticism of the ‘national origin’ of the victim, and 20.2% due to Islamophobic hatred. Passivity of users is the most common response to hatred messages related to Islam. Each week, more than 25,000 people in Spain receive messages of misogynistic content

5. 38% of tweets are written with the intention of annoying, insulting or threatening someone, according to a recent study by the University of Texas.

The fight against hatred speech has been fully established in various media for decades; however, the Internet has not developed the same way. Thus, it is estimated that around 11,000 websites contain content that incites irrational hatred (Delgado & Stefancic, 2014), without establishing a culture of universal response to hatred messages on social networks.

Cyberhatred, in addition, adds a series of particularities that make it difficult to fight it easily:

a. Communicative overabundance or overinformation.