Las empresas deben enseñar a ver la IA como herramienta y no como amenaza

Companies must be taught to see AI as a tool and not as a threat

The catastrophist discourse on Artificial Intelligence, which predicts the massive destruction of jobs, does not convince the innovation director of the Data Science and Artificial Intelligence Institute (DATAI) of the University of Navarra (UNAV) and academic partner of IA+Igual, Iván Cordón. 

Madrid, May 21. The AI expert considers it crucial that companies that adopt these tools make their employees see that they can benefit from them instead of seeing them as a threat that comes to replace them. Iván Cordón is clear that people are not going to be replaced by artificial intelligence, but they may be replaced by other people who know how to use this technology.

IA+Igual.– Given your strong background in AI and your experience in helping companies make the most of their data, how do you think AI is currently transforming HR processes?

Iván Cordón.- It is clear that AI is a cross-cutting technology or discipline, and to the extent that data exists, it has a transformative capacity for practically any sector. Speaking of Human Resources, there is a transformation about how AI is helping to improve their processes. AI through data makes it possible to understand how to help employees improve through training or perform better and be happier. There are already platforms that do this and recommend which skills or candidates in your company are better for certain positions. This benefits both the company and the individual. There is a company-employee and employee-company relationship that in my opinion needs a bidirectional projection.

AI enables two things. On the one hand, it streamlines processes and makes them more agile and faster, as could be the case of hiring new employees. But it also makes it possible to improve these processes, making them more effective and generating value for the company. Both could be applied to talent management.

This is from a process point of view, but we also have to think at a more generic level. We have to keep in mind that there has been a shift in how AI itself is approached in companies. We have gone from a society where HR valued someone with experience or who had solid knowledge, to a point where, thanks mainly to generative AI, we are matching profiles by skills. In other words, those people who are able to use this technology are the ones who are also adding value to companies. This is fundamental, because when it comes to hiring new people we can consider that a person with skills and knowledge to handle AI tools is much more interesting than a person with 10 years of experience but who does not have the ability to handle them. Hence these two ways, one more of processes and this way of transforming how the companies themselves are operating.

Ia+Igual.- What do you think is the biggest challenge and the biggest advantage companies face when implementing AI systems in their human resources processes?

I.C.- The biggest challenge facing all companies in implementing AI is people. We have to think that AI is here to stay and it is going to help us, it is not here to destroy anything, without forgetting that success and failure are made by people.

This is the key. We must have the organization fully aligned so that people embrace artificial intelligence and see it as a tool they can take advantage of, not as a threat coming to replace them.

There is a phrase that I like a lot in this aspect and that talks about how those people who work (not only in human resources), are not going to be replaced by artificial intelligence, but they will be replaced by other people who know how to use this technology. This is the reason why, when it comes to implementing it within the organization, it is very important to align everything and make this adoption and integration process effective.

Transforming role

IA+Igual.- What role does transparency play in the development and application of AI algorithms in HR processes?

I.C.- The role that transparency should play, although sometimes it does not, is fundamental and should be mandatory. It is true that there is already a small effort, which is already being worked on through the new law on Artificial Intelligence recently approved by the European Union.

In Human Resources, the impact of algorithms falls directly on people, so it is important that the final decisions are also made by a person. We must be able to understand why an algorithm recommends me to fire a person, or tells me that he/she is not well or that he/she is going to leave the company. Because if you do not understand the cause or the variables that make the algorithm say this, you will not be able to make an appropriate decision.

It is necessary to be able to give an explanation to the person who will be impacted by the decision. Not only from an ethical point of view, but also from a regulatory point of view, because we must be able to explain to the regulatory entity in this case, the European Union, why this decision is taken. In fact, many of the processes that have an impact on people are going to be prohibited, and the use of certain types of Artificial Intelligence is going to be prohibited. This is the case of surveillance, restricted to entities such as the police or intelligence services.

In this regard, it is necessary to be able, as far as possible, to explain and understand why that decision or that recommendation comes to us, so that we can make the decision in the most informed and transparent way possible.

Avoiding new forms of discrimination

IA+Igual.- What should companies do to prevent the use of AI from generating new forms of discrimination in employee-related processes?

I.C.- We have to be aware and very critical with the processes and with the incorporation of AI to the human resources processes in the company. When we are going to incorporate technology we have to understand very well what are its limitations, its pros and cons, and make an exhaustive analysis of the suppliers. We have to make sure that those companies that have developed Artificial Intelligence software comply with current regulations and that they have a focus when developing their models that puts the person at the center. We must follow "Human centered AI", that is, an AI that is fair and non-discriminatory, that is unbiased in its decisions and that allows for an explainability, as far as possible, of how the final decisions or recommendations are carried out. This is fundamental; we must have processes in companies that allow us to carry out this incorporation in a safe and reliable way.

IA+Igual.- Having worked with companies on several continents, have you noticed significant differences in the adoption and application of AI in HR between different regions or company cultures?

I.C.- Of course. To explain it, let's talk about biases. Biases are not associated with a model, an AI model is not biased, the bias comes from the data, and the data are a representation of the society in which we live, and each country has its biases.
There are cultures in which there is a much stricter care of the person, as is the case in Europe, where we already have a first regulation and we focus on making AI increasingly safe, reliable, explainable and ethical. On the other hand, there are cultures like the United States where these values do not transcend and there is more freedom. On the one hand, there is a way to try to regulate, which is what we do here, and there is a way to try to delegate and have the companies themselves self-regulate the use of AI and determine whether this use is ethical. Each of these ways has its pros and cons, but obviously there are differences in Latin America, in the United States, Europe or Asia in the way companies adopt artificial intelligence. This has to do with the different societies and of course with the people and culture in which these companies work. 

Emerging AI trends in HR

 IA+Igual.- With the rapid advancement of technology, what are the emerging trends in AI that you think will have the biggest impact on talent management in the near future?

I.C.- AI is undoubtedly already having a major impact on improving talent management. The integration of generative AI will facilitate this process, allowing a quick and efficient interpretation of employee information. This will allow us to know the employee better and will allow us to manage the talent we have in a more effective way

The company needs to think about how it can help employees be better so that the benefit is mutual. If employees feel fulfilled and are more comfortable in their work it will inherently improve productivity and the value they create for the company.

IA+Igual.- DATAI is part of the IA+Igual project as an academic partner. How did you structure this working group from UNAV?

I.C.- When I received the opportunity to apply as a partner, it was clear to me that, although we are experts in the technical side, artificial intelligence and a project like IA+Igual needed to have a 360-degree approach. Therefore, the structure of our proposal always had a multidisciplinary focus, mainly on three pillars: technical pillar, fundamental for the algorithms and how the data is built; an ethical pillar, about how it is managed and how the development of AI should be and a regulatory pillar, in which we enter into how society and entities such as the EU will regulate artificial intelligence. This fits with DATAI's multidisciplinary approach, and so we believe we can apply AI in HR, focusing on three aspects that help us understand the process and design a regulation to analyze these tools. In this way, we can determine whether they are well developed and worth certifying as secure.
IA+Igual.- What impact do you think the success of the AI+Equal project could have on the future of AI applied to Human Resources?

I.C.- The impact can be tremendous if we manage to develop a methodology that is understood as valid for auditing human resources algorithms. And I stay on the human resources side because it is not the same as auditing other types of algorithms. HR fundamentally impacts people, so these types of projects must be approached from a multidisciplinary and vertical perspective. This means focusing on the specific topic and analyzing in detail the intrinsic characteristics of each aspect in order to be able to contribute effectively.
To the extent that we can provide a clear and practical methodology that can be used in companies to improve, complement and validate the development of AI in a reliable, ethical and humane manner, we will be able to say that the AI+Equal project has been a resounding success.
Many projects talk about ethics in a generic way, but we want to make it concrete. We are going to create a roadmap for auditing algorithms in an empirical and concrete way. Thus, each company will be able to use this guide to audit and diagnose their tools, determining whether they are fine or need improvement.
A relevant role

The Data Science and Artificial Intelligence Institute (DATAI) of the University of Navarra (UNAV) plays a transcendental role in the investigation of the scope and impact of this revolutionary technology.

  • It is a research institute founded in 2019, whose function is to research in the latest artificial intelligence technologies with a strong focus on the transfer in society and to companies.
  • Team: 25 full-time staff and more than 100 collaborators including philosophers, neuroscientists, lawyers, computer scientists, physicists, mathematicians... A multidisciplinary group that shows its vision of AI and its vision of the world.
  • Pillars:
  1. Basic research: which allows scientifically generating and producing new advances for the evolution of society and science in the context of data science, AI and advanced statistics.
  2. Training: through a master's degree in Big Data and artificial intelligence and various courses.
  3. Transfer: which tries to bring together all the results of the research part and transfer them to society. To do this, they work with companies, in projects and we collaborate in initiatives such as IA+Igual as academic partners.
  • Objective: It is born with a vocation of explainability and its objective is to bring its expertise and knowledge from the research point of view to a real environment to help, for example, to audit AI tools applied to HR. This is where we want to work, making people understand that through AI we can improve companies, people or society. They understand AI as a tool and a transversal technology that allows us to help on different fronts and positively impact society.
  • Scientific areas:
  1. Digital and personalized medicine (helping to improve diagnostics or patient care)
  2. Advanced statistics through the optimal design of experiments.
  3. Natural language processing with the advent of generative AI.
Related Information
Social dialogue
Webinars Campus IA+Igual

Leave a Comment

Your email address will not be published. Required fields are marked *

Scroll al inicio