Quebec’s national library is embarking on an ambitious project aimed at developing a comprehensive database that will enhance the capacity of artificial intelligence systems to accurately reflect the province’s diverse cultural landscape, including Indigenous languages. The Bibliothèque et Archives nationales du Québec (BAnQ) has initiated the experimental phase of this cultural and governmental databank, following a thorough feasibility study completed earlier this year. This initiative seeks to address the notable deficiencies in AI systems’ access to reliable data pertaining to Quebec’s society, economy, and culture.
Addressing Data Deficiencies
The motivation behind this innovative project stems from a growing concern that existing generative AI technologies often lack a nuanced understanding of Quebec’s unique context. According to Valérie D’Amour, who spearheaded the feasibility study, the landscape is wide open for exploration. “All scenarios are a little bit on the table right now,” D’Amour stated. “We have a plethora of ideas and are keen to validate these possibilities with cultural stakeholders, as well as with data owners and providers who will engage in the discussions.”
BAnQ envisions a platform that will not act as a public distribution channel for creative works. Instead, it will impose strict controls over access to the data, ensuring that it is used responsibly and effectively. Marie Grégoire, the CEO of BAnQ, emphasised the goal of equipping AI systems with a wealth of Quebec-specific references. “This includes contributions from both the research sector and the business community, whether they pertain to smaller models or larger frameworks,” she noted.
Learning from Global Initiatives
This project mirrors similar efforts in other countries, such as Sweden, where extensive collections of Nordic-language texts have been compiled to enhance the development of generative AI models tailored to Scandinavian languages. BAnQ plans to commence with its own collections before potentially integrating data from external sources.
The initiative originates from recommendations made in a report by Quebec’s innovation council, which highlighted a critical issue: the “very small quantity of data on Quebec” present in current AI training datasets. Destiny Tchéhouali, a co-holder of a research chair focused on French-language artificial intelligence at Université du Québec à Montréal, voiced the urgency of this undertaking, stating, “Quebec culture remains underrepresented in the corpora currently circulating in the AI world.” He highlighted the risk of perpetuating linguistic and cultural biases, particularly regarding Indigenous peoples.
Protecting Cultural Creators
As BAnQ progresses with its plans, concerns have emerged regarding copyright implications for the cultural sector. Grégoire is hopeful that the proposed database could provide greater protections for creators compared to the current landscape. “At the moment, it’s a bit like the Wild West,” she remarked. “Data is being harvested for free, and that should not continue.”
The envisioned database could serve as a centralised gateway, facilitating fair compensation for creators whose works are incorporated into AI systems. Grégoire argued that a collaborative approach among cultural organisations would enhance the sustainability of the sector, ensuring that creators are adequately remunerated for their contributions.
However, not all artists are convinced. Some express apprehension that sharing their work for AI training could jeopardise their livelihoods. Maxime Harvey, a postdoctoral researcher at the National Institute of Scientific Research, articulated this concern, saying, “The primary criticism we encounter is that, even if artists see some financial benefit, they are ultimately feeding a system that may replace contracts they otherwise rely on.”
Timeline and Budget
The feasibility study outlines a projected operational timeline for the platform, aiming for it to be fully functional by 2029. D’Amour indicated that this timeline would be reassessed following the experimental phase. The project is estimated to require a budget of nearly $10.5 million over five years, encompassing both operational and capital expenditures. To support this initiative, BAnQ has already secured $340,000 from the Quebec government for the feasibility study and an additional $750,000 for the 12-month experimental phase.
Why it Matters
This initiative represents a significant step towards ensuring that artificial intelligence systems can accurately reflect the rich tapestry of Quebec’s culture, languages, and societal nuances. By developing a dedicated database, BAnQ not only aims to empower local creators and safeguard their rights but also to combat the cultural biases that have emerged in the AI landscape. As the world increasingly relies on AI, ensuring that these systems are informed by diverse and representative data is crucial for fostering a more equitable digital future.
