Declaration aims to encourage responsible Artificial Intelligence development

Open this photo in gallery:

In keeping with its leading role in the emergence of Montreal as a global hub for research in artificial intelligence (AI,) Université de Montréal (UdeM) is at the forefront of efforts to control the responsible use of the technology and to ensure it is used for the benefit of society.

When Yoshua Bengio, head of the MILA and Canada Research Chair in Statistical Learning Algorithms, began his pioneering work in neural networks and machine learning at UdeM more than 25 years ago, AI wasn’t attracting the attention it receives today, and it wasn’t clear if and how AI research would benefit Canadian industry, he says.

Nevertheless, UdeM encouraged research at the MILA and supported the growth of the faculty, which boosted the development of a critical mass of machine-learning experts in the city.

As Montreal’s AI cluster earned worldwide attention and tech giants including Microsoft and Google established labs in the city, Prof. Bengio opted to stay in academia. “Knowing my work is going to be available for everyone and for society at large and helping develop the next generation of leaders in AI is incredibly satisfying,” he says.

In addition to the pioneering research at UdeM, Prof. Bengio believes the institution’s values and ethics play a role in attracting tech industries to the city and enhancing its reputation.

But while AI is being used in many real-world applications, ranging from smartphones to health care, the powerful technology is destined to have a significant impact on society. “We have to make choices about how we want this technology to be used,” says Prof. Bengio.

Catherine Régis, an associate professor in UdeM’s Faculty of Law, the Canada Research Chair in Collaborative Culture in Health Law and Policy, and the co-director of the newly created Health Hub – Politics, Management and Law, agrees that AI raises difficult legal questions that are intertwined with ethical issues.

The development of the Declaration for a Responsible Development of Artificial Intelligence, initiated by UdeM and Fonds de recherche du Québec, is a way to involve the public in decision-making around the issues of AI, she says, adding the declaration is a valuable educational tool.

“AI is a complex subject, and the declaration provides a unique opportunity to engage people in discussion – not only for citizens but also for scientists to find out what people think about the subject,” says Prof. Régis.

The declaration also raises social and ethical issues not only for the legal profession but also for developers and researchers, says Prof. Régis, noting that forums and workshops held under the aegis of the declaration gives those working in AI valuable opportunities to reflect on their own practices.

Knowledge and understanding of the potential impacts of AI are essential to ensure a legal framework that is as relevant as possible, she adds. “The more people are involved and the more public the discussion becomes, the more difficult it is to put it aside and say that it’s not a problem,” she says. “The declaration also reminds everyone that we all have a duty to make sure we develop AI for good [causes].”

Produced by Randall Anthony Communications. The Globe’s editorial department was not involved in its creation.