Artificial Intelligence and Education: Protecting the Heritage of Humanity

The COVID-19 pandemic has transformed our lives in more ways than one. It has not just alerted us to the vulnerabilities of our health systems but also how ill-equipped our education systems are to cope with disruptions of this scale. When the pandemic forced schools to shut down and learners had to completely switch to online learning systems, the transition was anything but smooth.

As part of the coordinated global education response to the COVID-19 pandemic, UNESCO, UNICEF and the World Bank conducted a Survey on National Education Responses to COVID-19 school closures. According to this joint report, 108 countries reported missing an average of 47 days of in-person instruction due to school closures – the equivalent to approximately one quarter of a regular school year – a long gap in the life of a student.

The mad rush to get online learning platforms will be one externality that will definitely emerge from the pandemic. Only time will only tell us if this externality will be positive or negative for humanity. But we can tilt it towards it being positive by taking action sooner rather than later.

In the rush to get a quick fix solution, governments are allowing private companies to offer their online learning platforms, tools and products for free-at least for the time of this crisis. This will no doubt ensure children continue with their studies and disruption in the education system is kept to a minimum.

A number of challenges have emerged, ranging from teacher preparedness, internet access, digital content suitability and student flexibility, among many others. These are all ongoing issues many countries have been grappling even before the crisis hit the world.

But one important issue that has not received enough attention revolves around data ownership and management. As students get enrolled on online platforms, the data generated about themselves of their learning is documented without much regard to privacy and accountability. The ethics underlying data ownership has yet to be discussed and the rush to get learners online on a host of predominantly privately owned learning platforms suggests another potential social virus will gather momentum to the deterrent of individual learners. 

We draw on President Clinton, who had already announced in March of 2000 that the genome sequence as such could not be patented when the Human Genome was finally cracked in April 14, 2003 by a truly international endeavour headed up by The International Human Genome Sequencing Consortium and involving more than 20 research institutions in the US, UK, France, Japan and China. This had significant ramifications as it could not be appropriated by private interests and it belonged to all of humanity – a commons like the air we breathe. This begs the question, should data like the genome sequence be above appropriation by private corporations? 

This is why UNESCO in its Universal Declaration on the Human Genome and Human Rights acknowledges that: « The human genome underlies the fundamental unity of all members of the human family, as well as the recognition of their inherent dignity and diversity. In a symbolic sense, it is the heritage of humanity. »  and “The human genome in its natural state shall not give rise to financial gains.”

It seems inevitable that the same questions should now be raised with respect to « human data » (and in particular information collected on human behaviour with regards to education).

Is such data a common heritage of humankind?

One of the reasons why this question is becoming increasingly urgent is that “human data” has become an indispensable  commodity  in the development of artificial intelligence (AI) in education. Controlling the supply chain of this commodity amounts to controlling AI.

It may well be that the real question is whether AI for education is a common heritage of humankind. If yes, then the indispensable human data commodity will invariably be a common heritage of humankind.

Just as it is important to safeguard the data of the users, there is an economic argument of efficiency that clearly emerges. The power of the AI algorithms comes from the quality and quantity of data. Pooling the data suggests efficiency and effectiveness improvements that can only help the end user but also pushes corporations to continuously improve their algorithms to capture the growing market of individualised learning.

If the international community resolves to no longer allow this groundbreaking and disruptive technology to rest in the hands of a few corporations for utilization for education we need to develop an international treaty designating a custodian for such data.

A special purpose organization could oversee and manage such exchanges modelled on the basis of ICANNS’s (Internet Corporation for Assigned Names and Numbers) stakeholder structure but with the exception that this new proposed body must be multilateral and multi-stakeholder.

We are in a crisis now and establishing such a body can be a time-consuming process. But there is no time for complacency. We can’t afford to wait for the privacy crisis when it happens but start work immediately. What we must do now is to ensure that all private data collected during this crisis must be purged by the private corporations on a weekly basis and the identity of the user completely anonymous until such an agreement can be put in place.

Dan Shefet, Lawyer at the Paris Court of Appeal (France) and Anantha Duraiappah, Director, UNESCO MGIEP