Building on its 75-year history, Unesco has the responsibility and the ability to help solve the challenges of this century. The digital revolution is one of them. From the beginning of my term in 2018, he led us to launch an ambitious project: to give the world an ethical framework for the use of artificial intelligence (AI). Many thought it would not be possible. Three years later, thanks to the mobilization of hundreds of experts from around the world and intense international negotiations, the 193 Member States of UNESCO have just officially adopted this ethical framework. It is a historic moment.
The Unesco recommendation on the ethics of artificial intelligence is a call to change the business model so that the world of AI is safe, fair and inclusive. It is a call to strengthen the capacities of governments to deal with AI through effective legal frameworks, regulations and institutions. We harness multilateralism as a mechanism that helps ensure that governments act for the good of the people. Thus, it protects and promotes existing human rights anchored in binding international law and provides an anticipatory framework to help mitigate future risks and offers concrete policy actions designed to solve emerging challenges.
It is based on four core values, which guarantee human rights, the flourishing of the environment and ecosystems, diversity and inclusion, and peaceful, just and interconnected societies. The values are broken down into ten principles.
Proportionality ensures that the use of AI systems does not exceed what is necessary and appropriate to the context, while security addresses unwanted damage and vulnerability to attacks. The principle of fairness and non-discrimination calls for the benefits of AI technologies to be shared taking into account the specific needs of different vulnerable groups and addresses discrimination, digital and knowledge gaps, and global inequalities in this regard.
Sustainability ensures the development of sustainable societies based on the achievement of a complex set of goals on a continuum of social, cultural, economic and environmental dimensions that are affected by AI technologies. Data protection and the right to privacy emphasize issues related to how data is collected, used, shared, archived and disposed of for AI systems.
The principle of transparency and explicability recognizes that people have the right to be aware when making a decision based on AI algorithms, and in those circumstances requires explanatory information from both the private and public sectors and a possibility of redress, while the principle of responsibility and accountability guarantees the attribution of responsibility to human beings.
This global ethical framework recognizes the need to ensure public awareness and understanding of AI technologies and the value of data, so that all members of society can make informed decisions about the use of AI systems and are protected from undue influences.
Beyond the “what” of values and principles, the recommendation offers clear guidance on “how” to move forward and achieve impact, proposing political actions in areas ranging from ethical governance and administration and development and international cooperation. to culture, health and social welfare.
Thus, the area of political action on gender ―one of Unesco’s global priorities― focuses on the need to increase diversity in the field of AI and ensure that women and girls are represented in the development of these tools and benefit equally from technology. It is designed to address a critical gender and AI gap and to help address biases and stereotypes in algorithms.
The recommendation underscores the importance of data governance, promotes a better understanding of the role of data in the development of safe and equitable algorithms, and establishes the rules to keep control of data in the hands of users, allowing them to access the information and delete it. It also calls on Member States to ensure the creation of adequate safeguards for the processing of sensitive data and effective systems of accountability.
Furthermore, the recommendation underlines the need for Member States to equip workers with the necessary skills to adapt to technological changes and thrive in the digital age, including through training and requalification programs. The recommendation calls on Member States to ensure consumer protection and avoid monopolization.
To ensure its implementation, the recommendation includes two tools to be developed by UNESCO that address the challenges related to the impossibility of codifying and regulating all aspects of AI without hardening innovation. They are motivated by the fact that, as in any innovative activity, downside risks can occur at all stages of the AI lifecycle.
The “ethical impact assessment” will help AI stakeholders to assess the impact that data sets, conceptual frameworks and algorithms may have on society, and to address or mitigate potential risks. The “assessment methodology” will help countries identify their readiness to implement the recommendation, with the aim of supporting capacity-building efforts to address the transformation that AI is bringing about in virtually every area.
Audrey azoulay is Director General of UNESCO.
You can follow EL PAÍS TECNOLOGÍA at Facebook and Twitter or sign up here to receive our weekly newsletter.
Sign in to continue reading
Just by having an account you can read this article, it’s free
Thanks for reading EL PAÍS
#ethical #framework #artificial #intelligence