In a groundbreaking initiative, Quebec’s national library, Bibliothèque et Archives nationales du Québec (BAnQ), is advancing plans to establish a comprehensive database aimed at enriching the training of artificial intelligence systems. This project seeks to provide a more nuanced understanding of Quebec’s diverse society, including its cultural tapestry and Indigenous languages. Following a thorough feasibility study completed earlier this year, BAnQ has embarked on an experimental phase to develop this government and cultural databank, which will focus on French and Indigenous linguistic content.
Addressing Data Deficiencies in AI
The primary motivation behind this initiative stems from the recognition that many prominent generative AI models often lack the depth needed to accurately represent Quebec’s unique cultural and social environment. Valérie D’Amour, who spearheaded the feasibility study, emphasised in an interview, “All scenarios are a little bit on the table right now. We have a lot of ideas and we want to validate the possibilities with cultural stakeholders, as well as with data owners and providers, who will be involved in the discussions.”
BAnQ aims to create a tightly controlled platform that will not function as a public distribution channel for creative works. Instead, it seeks to ensure that AI systems are better grounded in authentic Quebec references, whether these are derived from small-scale models or large-scale research initiatives.
Learning from Global Efforts
Quebec is not alone in its quest to build AI resources tailored to regional cultures. Similar projects have emerged in other countries, such as Sweden, where extensive collections of Nordic-language texts have been compiled to bolster AI models for Scandinavian languages. BAnQ plans to commence its data collection with its own archives before considering supplementary data from external sources.
This initiative is rooted in a recommendation from a 2024 report by Quebec’s innovation council, which highlighted the troubling scarcity of Quebec-specific data in AI training datasets. Destiny Tchéhouali, co-holder of a research chair focused on French-language AI at Université du Québec à Montréal, pointed out that Quebec’s cultural contributions are significantly underrepresented in the current AI landscape. He warned, “We run the risk of reproducing linguistic biases and cultural biases. And when we also talk about Indigenous peoples, we run an even greater risk of all these biases.”
Cultural Protection and Copyright Concerns
As BAnQ develops its proposed database, copyright issues emerge as a significant concern within the cultural sector. However, Marie Grégoire, president and CEO of BAnQ, expressed optimism about the potential benefits of the platform, stating it could offer enhanced protection for creators compared to the existing system. “Right now, it’s a bit like the Wild West,” she remarked. “Data is being harvested for free, and that should not be the case.”
The envisioned database is designed to serve as a centralised gateway, facilitating fair compensation for creators whose works are utilised in AI training. Grégoire believes that by collaborating, cultural organisations can better safeguard creators’ rights and ensure long-term sustainability for the sector.
Despite these assurances, some artists remain apprehensive. Maxime Harvey, a postdoctoral researcher at the National Institute of Scientific Research, noted that there is a prevailing fear that contributing work to AI systems could jeopardise their future livelihoods. “The main criticism we hear in the field is that, even if artists earn income from it, they are still feeding the beast that will eventually be used to replace contracts they may lose because of AI,” he stated.
Looking Ahead: Timeline and Budget
The feasibility study envisions making the database operational by 2029, although D’Amour indicated that this timeline will be reassessed after the experimental phase concludes. The study has outlined a budget of approximately $10.5 million over five years, which will cover both operating and capital costs. To support this ambitious project, BAnQ has already secured $340,000 from the Quebec government for the feasibility study, along with an additional $750,000 to fund the upcoming 12-month experimentation phase.
Why it Matters
The establishment of this cultural and government databank by BAnQ is a pivotal step towards ensuring that Quebec’s rich heritage and linguistic diversity are adequately represented in the rapidly evolving AI landscape. As artificial intelligence increasingly shapes our world, the potential to mitigate biases and enhance cultural understanding is not only vital for local creators but also crucial for fostering a more inclusive digital future. By investing in this initiative, Quebec is taking meaningful strides to protect its cultural identity while navigating the complexities of modern technology.