Artificial intelligence is making more and more decisions

Summary:

Artificial intelligence is making more and more decisions It is the first global framework to deal with the issue in a way that is all-encompassing, fair, and respectful of basic rights. As of November 2021, all 193 Unesco member states unanimously agreed to the AI ethics recommendation. Broad application of artificial intelligence using the Chinese company's platform offers technological advantages such as the use of specific processors for artificial intelligence and the use of applications in remote areas through small devices.

AI makes more decisions

Artificial intelligence (AI) is today a reality that goes beyond the technology sector; its presence stands out at different levels and sectors; even its application is increasing in the area of health, said Eleonora Lamm, Regional Advisor in Bioethics and Ethics of Sciences of the United Nations Educational, Scientific and Cultural Organization (Unesco) in Montevideo.

She participated in the "Artificial Intelligence Forum. First results", organized by UNAM and the company Huawei, and referred to the recommendations of the international organization on the ethics of AI, which are because more and more public and private decisions are made through artificial intelligence, whose results "are not free of biases or ill-defined frameworks". So, it's important to keep an eye on the information and its privacy, even though there are also issues with human rights and ethics.

As of November 2021, all 193 of Unesco's member states unanimously agreed to the AI ethics recommendation. This made it the first global framework to deal with the issue in a way that is all-encompassing, fair, and respectful of basic rights.

It is an ambitious and comprehensive instrument, the product of two years of negotiation, which covers the following areas: respect, protection, and promotion of human rights and fundamental freedoms; the need to ensure diversity and inclusion; the well-being of the ecosystem and the environment; and living in peaceful, just, and inclusive societies, values that are transformed into principles that today govern the fields of AI.

These translate to "proportionality" and "no harm," "safety and security," "equity and non-discrimination," "sustainability," "privacy and data protection," "oversight" and "human determination," and "transparency and enforceability," "responsibility and accountability," "awareness and literacy," and "multi-sector governance and collaboration."

Unesco's recommendations on AI ethics "have several specific areas of action such as data policy, gender, development and international cooperation, an environment and ecosystems policy action area, health and well-being, communication and information, education and research, and culture."

All of these policy action lines include concrete steps for putting artificial intelligence to work for humanity. Furthermore, unlike other instruments, this suggestion has an important gender perspective, as approximately 300 million more women than men lack access to mobile internet, and women hold only 25% of jobs in the field of technology.

In this context, she said that one of Unesco's two global principles is its concern to ensure that gender biases are not reproduced in this instrument, "but we are also concerned with those populations that are traditionally discriminated against, such as people with disabilities. However, the development of AI-based technologies aimed at this sector does not take into account the diversity implied by disability, which can also generate bias and lack of access.

Work is also being done with indigenous peoples in this area. In Latin America and the Caribbean, they represent eight percent of the population. Thirty percent of them live in extreme poverty and experience the digital divide, so it is essential to work on the one that considers them and gives importance to the sovereignty of their information, such as languages and cultural heritage.

It also seeks to generate knowledge in populations, mainly discriminated against, so that there is literature in Spanish and from the region to show what is happening; there is a Mooc (open, massive, and online course) on AI "so that no one is left behind to socialize and democratize access to information".

Broad application of artificial intelligence

On the other hand, the 10 AI-related projects, approved in 2021 as part of the Alliance to Promote the Development of Digital Capacities in Mexico, focused on addressing problems in health, agriculture, video surveillance, and the environment.

This was explained by researchers who obtained this support through which they had access to spaces in Huawei's servers, received training from the German Cooperation Agency for Sustainable Development, interact with other research groups, and got closer to the experience of linkage and knowledge transfer.

The experiences were shared in the panel "Artificial Intelligence for the Solution of Social Problems", moderated by the researcher of the Institute of Social Research (IIS) of the UNAM, María Josefa Santos Corral, which was part of the Forum "Artificial Intelligence First Results", organized by the Institute for Research in Applied Mathematics and Systems (IIMAS).

The professor and researcher of the Faculty of Engineering, Architecture, and Design of the Autonomous University of Baja California (UABC), Dora Luz Flores Gutiérrez, explained that her project consisted of determining the drug resistance of pulmonary tuberculosis in patients in Baja California, which she carried out with her colleague Raquel Muñiz and three graduate students.

Accessing Huawei's servers allowed them to process complete tuberculosis genomes in a matter of hours, while conventional servers can take up to five months. This time, he stressed, it means diagnosing and treating patients.

In turn, Juan Pablo García Vázquez, professor at the UABC School of Engineering, explained that he and six colleagues carried out a precision agriculture project, in which they used AI to recognize the level of ripeness of fruits such as dates.

Using the Chinese company's platform offers technological advantages such as the use of specific processors for artificial intelligence and the use of applications in remote areas through small devices. In addition, they developed a robotic vision system to select the fruits, thus reducing the time people spend on this task.

Meanwhile, Fernando Pech May, professor, and researcher at the Computer Systems Engineering Academy of the Instituto Tecnológico Superior de los Ríos, in Tabasco, explained that at that university they developed a plan for monitoring floods in southeastern Mexico to prevent the severe damage they usually cause in the region. They were able to use AI on the Huawei platform to look at a lot of data, in this case, satellite images.

Their project is important because they detected that many municipalities lack current risk maps or make them manually. In addition, they were able to integrate authorities, civil society, the government, and academia.

Other research focused on the detection of strawberry maturity; assistance in environmental monitoring; the development of a prototype of a web-based medical assistance system for the classification of mammographic studies; etc.