UNESCO has launched a wide process of consultations to obtain the many points of view of stakeholders. This involved experts from 155 countries, members of the public (through a global online survey), United Nations agencies, major stakeholders from the sector such as Google, Facebook and Microsoft, and the world of academe with the University of Stanford and the Chinese Academy of Sciences.
‘We must make sure Artificial Intelligence is developed for us and not against us,” declared UNESCO Director-General Audrey Azoulay. “We need a robust base of ethical principles to ensure artificial intelligence serves the common good. We have made this process be as inclusive as possible since the stakes involved are universal,” she explained.
Over recent years and all the more so since the outbreak of COVID-19, there have been ever more applications drawing on AI, notably with the aim of accelerating research into vaccines and improving tracing of the virus. AI has contributed to the development of telemedicine and distance learning. It has also been used to operate drones for the delivery of medical supplies and the need for a global regulatory instrument has grown on a par with the spread of AI applications.
The potential of artificial intelligence, as described in both scientific publications and works of fiction, gives rise to fear that machines will take decisions out of human hands, that it will erode individuals’ right to and that it will expose people to manipulation to the detriment of their rights and freedoms. The massive quantity of data collected and processed daily raises major concerns over confidentiality, privacy, and the reproduction of discriminatory practices and stereotypes.
The draft recommendation submitted to the international community establishes a number of over-arching concepts. To start with, AI technologies must not exceed what is necessary to achieve legitimate aims or objectives, and should be appropriate to the context.
There is also a requirement for human oversight and determination – ensuring that humans are ethically and legally responsible for all stages in the life-cycle of AI systems.
Stewardship of the environment and peace is no less important. Throughout their life-cycle, AI systems must contribute to the peaceful interconnectedness of all living creatures with each other and respect the natural environment, notably with regard to the extraction of raw materials.
Equally important is gender inclusion. AI technologies must not reproduce the gender inequalities found in the real world, notably with regard to salaries, representation, access, and stereotyping. Political actions, including measures of positive discrimination, are required to avoid these major pitfalls.
UNESCO will assist governments and civil society players (corporations, members of the public etc.) in developing concrete awareness-raising campaigns and ethical impact assessment tools for AI in all fields.
Beyond wishing to establish international consensus on the subject, UNESCO’s experts urge Member States and AI players further to raise public awareness and stress that it is important for everyone to be made aware of their digital rights.