 So, it's called Chromatocracy, the Pantone of Mexican Social Mobility. A few words of me, I do work in the world of technology, but all the speech about figuring out how to simply monetize projects with AI or data science absolutely bores me to tears. You know, I'm reading a really cool article on Wired Magazine about a new washing machine that predicts with AI when the subcontainer is going to be empty. It's like watching paint dry for me. What I do care about, however, is how to translate and use this technology with AI or data science to improve the knowledge of our society. So another thing I'm just learning, I started learning English five months ago. So maybe something, my English, I'm going to do my best. So the question that I want to answer is, is there a mathematical formula for succeed in Mexican social mobility? Is there a model or a formula? And the first part is, since before it became an independent nation state in 1821 in Mexico, those populations have been troubled by issues of race and class. Anything from Spain here? No? Okay. Because when Europeans arrived to colonize the Americans more than 500 years ago, they introduced a hierarchy based on skin, color, and race that persists to this today. This is called casta pintura, casta pinting that represents all the hierarchies based on skin, tone, and race in Mexico. If you are a Spanish and a Creole, you have less social mobility on the ancient Mexico. And also, in contemporary Mexican advertising and films, lower class people are almost always shown with darker skin. And richer class people are shown with white skin. The lead character in the Oscar-winning film Roma, for example, is a darker skin, representing somebody considered poor in Mexico. Her boss, on the other hand, is rich and white. Some Mexican companies choose to feature white skin models to represent their Mexican brands. This so-called aspirational advertising has become common practice for most advertising companies in Mexico. This is, we are going to see a little video about, this is a recent campaign creating the last year, 2018, for the luxury department store chain called Palacio de Hierro. They tried to make the brand appear diverse by feature an androgynals model, a freckle model, and a disabled person. Maybe this sounds Greek, except they are all still like skinned people. So if you want, this is a commercial branded from Mexico, but you don't recognize Mexican people in the advertising, it's maybe a US people. No, this is no, this is Mexico, but the people, it's not Mexican. So let me stop that, I lost my, here. So well, the perception within the Mexican advertising industry is that if a white person use a specific brand, the product is considered to be desirable. Mexican marketers believe that if a consumer buys an aspirational product that is used by a skinned, a lighter skinned person, the consumer thinks they are participating in a desirable, wide lifestyle. So what happened next? The National Institute of Statistics and Geography in Mexico, each by year produce a study about social mobility in Mexico. What happened if I live better than my parents, if my parents' lifestyle is better than my grandparents, but since last year, they include another feature, a new feature in the study that was the skin tone. They found that whiter people have a better social mobility than darker skinned people. This was really strange in Mexico, because for our culture, we think that we don't have racins or, yes, racins in our culture, we think, ah, this is for other countries, not happens in Mexico. But when the institute publishes their work, all the journalists start to scream in, it's mostly because they don't have, they don't know statistics. But the other journalists start thinking that, ah, this is not happening. The racins based on color skin in Mexico doesn't exist. So to prove that, I made an algorithm first. I made a web scrapping of the Congress website in Mexico. I allowed to have the photo of the congress people, the name and the political party. Then I have all the data of the congress women and congressmen in Mexico. Then with the photo, I trained a model for the automatic Google image search. So I found, I know, ah, this is the congresswoman, I know one photo, I'm searching in the Google image database, Google tell me, ah, all these pictures are for the congress woman, and I'm training the model to recognize her on all images. The next step was a face recognition algorithm to tell the model, ah, this is the person that you are looking for. So that was, in this step was a funny thing for me because I'm from science, but also I'm from the art. I am an artist and a scientist, and I collaborate for the face recognition algorithm. I'm re-use another algorithm that I use for an art project. Let me play the video. This piece I made it with a Mexican artist called Rafael Osano-Hemmer, is the most famous electronic artist in Mexico. I made the algorithm to this piece. The piece is called Level of Confidence, it's an art project to commemorate the mass kidnapping of 43 students from Ayotzinapa Normalist School in Iguala. The project consists of a face recognition camera that has been trained to tirelessly look for the faces of the disappear students. As you stand in front of the camera, the system uses an algorithm to find which student facial features looks most like yours and gives a level of confidence of how accurate the match is in a percent. The biometric surveillance algorithm used are typically used by military and police forces to look for suspicious individuals, whereas in this project, they are used to search for victims instead. This is an open source project for face recognition. You can download it on GitHub. It's a little ... I'm going to share the video also. But when I use this algorithm that I developed with Rafael Osano for this project for face recognition, so the first step, I have all the photos and the data of the congresspeople. I use an automatic Google search image. Then from the image, I use this algorithm that I take for the art piece that I collaborate with Lozano Hemmer to identify the face of each person, and then I make a skin tone subtraction and calculation. I made another model to recognize the skin, and I calculate a really, really accurate tone scale based on the average of all the photographs. With that, we realize something really, really strange in Mexico. The right-wing political parties in Mexico are wider than the left-wing parties. The left-wing parties are darker than the right-wing parties. For Mexico, maybe we think that, but based on data, confirm that this is happening. In Mexico, this was a shocker discover. So then I have these, I have the results of INEGI, the National Institute of Statistics. I have the results by set of our congress. The next step is to define who is more successful in Mexico. For that, I made a web scrapping of LinkedIn, searching all the people that presumed to be CEOs in Mexico. So that happened. I found below, you can find the six first skin tones based on the Perla skin color palette. You can find the three firsts of them, 45 plus, 35 plus, nine is 80-something. 80-something percent of all the CEOs in Mexico have white skin tone, and only less than 10% of the CEOs in Mexico have brown, have darker skin. The next is only for geographics. This is where the CEOs are located in Mexico, in the main cities, principle in Mexico City, the state of mix in Mexico, Monterey and Guadalajara, that are currently the most important cities in Mexico, they are located in the CEOs. The next one is where the CEOs study. You can see this, this is the 14%, it's the National Autonomous University of Mexico, but is the only one that is free to enter. The other universities are private universities. The first one, it is both private, it's the only one that is free, there are all private schools. So the 80-85% of all the CEOs came from private schools, universities in Mexico. But there is a difference. This is, for example, where the students of the CEOs came from items, one of the private schools. This is how it's separated by an area of a student. The area one is the engineering and science, the two areas, it's chemical and biological, the three, it's all the stuff about, how do you say? Management and all this stuff, and the four areas, it's humanities. You can see the CEOs that came from private schools have a training in area three, it's management, all this stuff of the area three. The CEOs that came from the national university, they studied something related to engineering or science. And this is really strange because the way that the private universities and the national universities formed their students, they have maybe a bias, and that was funny. The other thing, the gender of the 93% are men and only the 7% in Mexico are CEOs. And this is the model from the mathematical formula for succeeding in Mexican social mobility. It's statistics, it's not a law, but if you are white and you come from private university, the probability in Mexico to become a CEO, it's higher than the other options. We present this study and became really famous in Mexico, in the TV, and the Robista Chilango is one of the most famous reviews, newspapers. I don't know how to say Robista. A publication, it's a really famous publication in Mexico City, it's called Chilango. They dedicate the number of their February issue for my study, they dedicated all the number for thinking about the raisins based on the skin tone in Mexico. Thank you. When the editor came to write a review or something like that, I told you, I don't know how to write because I'm a scientist. I know how to write, but not for a journalist level. So both I make stuff with computer, with programming. So we decide to make this interactive cabin. It's like a photo booth cabin. You can stay, the same algorithm recognizes you, take your skin tone, and actuate the skin tone with TensorFlow technology. And you can, the algorithm measures all you, the algorithm and the photo booth gives the people, this thing is a photograph of her, skin tone, and data related with your skin tone, your degree, where you're from, something like, I don't know, the 2% with your same skin tone and your level of study in Mexico have, I don't know, something access to water or no, I don't know, it's only for, it's a project to take a look about the skin tone discrimination in Mexico. So for this project, we won the national human rights prize from media. And we are really, really excited about it. So for me, code and data is a creative tool. Not only primarily for unravel the hidden dynamics on Mexico. And for me, this works very, very, very well. So you can download all the code. It's on GitHub. You can use the face tracking, the skin color tracking for whatever you want. So thank you for being with me.