The ITU opened its AI for Good Summit in Geneva with the goal of highlighting how artificial intelligence can be used to help achieve some of the world’s major objectives, as identified in the United Nations’ Sustainable Development Goals (SDGs).
It seeks to connect AI innovations with global problems – like universal access to food supply and healthcare – to find sustainable solutions.
Many organizations are using artificial intelligence and big data to help them achieve their objectives. The International Committee of the Red Cross (ICRC), which provides relief to parts of the world subsumed by violence, war and disaster, has identified the potential that AI has to do good. But it has also identified some of the risks.
Not just data, but people
The ICRC is responsible for managing vast quantities of data. Much of this data is sensitive, such as the information it collects during its visits to war prisoners, and its disclosure can lead to deadly consequences to individuals. The ICRC must manage its data with integrity and ensure its proper protection. According to its Director General, Yves Daccord, “it is not just data but people. Our management of data can have life or death consequences”.
The ICRC already makes use of big data and algorithms to complement its work. Big data and smart mapping can be used to identify where places where the ICRC needs to provide support. Social media, like Twitter, can help pinpoint problems such as the epicentre of a disease outbreak.
But big data and machine learning can also have harmful consequences. The way war is waged is changing and Daccord noted that we “need a principled design of technology. We need to keep human control and accountability”.
Role of standards
Standards can mitigate and address some of the concerns around the use of artificial intelligence.
The IEC and ISO joint technical committee on artificial intelligence, (ISO/IEC JTC 1/SC 42), was established in 2017, to develop international standards for artificial intelligence on topics such as AI trustworthiness, use cases, applications, governance implications of AI, computational approaches of AI, ethical and societal concerns.
As Wael William Diab, Chair of ISO/IEC JTC 1/SC 42 noted, “it’s important that we address both the technical challenges as well as the ethical, societal concerns and trustworthiness aspects of the technology. Addressing these technological perspectives from the start by bringing in as numerous and diverse stakeholder views, will enable broad adoption.”
During its recent 3rd plenary meeting in Dublin, ISO/IEC JTC 1/SC 42 approved of a new project to develop an overview of ethics and societal concerns in the area of artificial intelligence.