1. What's new

New partnership research project examines AI’s impact on the quality of work

Posted

SSHRC-funded project led by Institute for Work & Health draws more than 30 partners spanning many sectors

The Institute for Work & Health (IWH) has been awarded funding to lead a seven-year partnership research project to study artificial intelligence (AI), job quality and worker wellbeing. The project is supported by funding from the Social Sciences and Humanities Research Council (SSHRC), through its flagship Partnership Grants program.

The project, titled Partnership on AI and Quality of work (PAIQ), comprises 34 government agencies, civil society organizations, AI research organizations, unions, employers and applied research institutes. The research team, led by IWH Associate Scientific Director and Scientist Dr. Arif Jetha, comprises nearly 40 researchers spanning occupational and public health sciences, engineering, business, policy, ethics and other disciplines.

Historically, the introduction of new technologies has led to productivity gains and economic opportunities for some, while worsening job quality and employment outcomes for others. The rapid adoption of AI, systems designed to perform tasks that typically require human intelligence, has the potential to reshape the working world. While most experts believe AI will impact the nature and availability of work, less efforts have been made to understand how these impacts might differ across occupations and industries, and how these impacts will influence occupational health and safety and wellbeing.

Despite the potential of AI to fundamentally shift the interface between work and health, there is currently limited evidence that can be used to develop solutions that can ensure that the health and safety of workers are protected, says Jetha. The need to keep up with the rapid advancement and use of the technology is one reason this research partnership is important, he adds.

Research led by IWH has shown that all jobs will have some tasks that can be performed by AI. No jobs will be untouched, says Jetha, citing findings from a study he recently completed that explored the characteristics of occupations most impacted by machine learning, an AI subfield. What’s different about AI is the worker groups most affected may be different from the worker groups impacted by past technological shifts. This might lead to different patterns of inequity emerging in the labour market compared to those seen in the past.

The partnership has three priorities. They are:

  1. to regularly monitor insights and share research findings on the impact of AI adoption and use in different workplace settings and their impact on job quality and worker health, safety and wellbeing;
  2. to use existing population-level data sources to examine Canadian workers and occupations most likely to be impacted by AI and direct policy and programmatic attention to protect workers; and
  3. to conduct field research at job sites to directly observe the effect of AI on workers, including how AI may impact job tasks, perception of work, and worker wellbeing.

When taken together, the research aims to identify and explain how different worker groups and work context characteristics may be related to harms or benefits arising from the use of AI at work.

The involvement of partners representing different sectors and perspectives will help to strengthen uptake of the project’s outcomes and co-design practical strategies, says Jetha. Our partnership with government, standards organizations, unions and civil society organizations will enable our work to inform legislation and organizational practices to improve conditions at work and protect workers adversely affected by AI, says Jetha.

Our collaboration with national research organizations at the forefront of AI provides PAIQ with a unique opportunity to integrate research evidence into recommendations for responsible and safe AI design that can be shared with AI developers as they commercialize the technology, he adds, pointing to AI organizations such as the Canadian Institute for Advanced Research (CIFAR), the Vector Institute, Mila (l’Institut Quebecois d’Intelligence Artificielle) and Amii (Alberta Machine Intelligence Institute).

To stay on top of findings and updates from this project, including updates about the project launch event in the fall, go to the project page: www.iwh.on.ca/projects/partnership-ai-quality-of-work.