Quebec is taking a significant step towards enhancing the representation of its unique society, culture, and Indigenous languages in artificial intelligence systems. The Bibliothèque et Archives nationales du Québec (BAnQ), the province’s national library and archives, has embarked on an ambitious project to create a database of cultural and governmental content. This initiative follows a recent feasibility study and aims to improve the understanding of Quebec’s diverse landscape within generative AI frameworks.
Addressing the Data Gap
The push for this database arises from concerns that existing AI systems often lack reliable information about Quebec, leading to a skewed representation of its society and culture. Valérie D’Amour, who spearheaded the feasibility study, emphasised the importance of collaboration with cultural stakeholders and data providers. “All scenarios are a little bit on the table right now,” she stated, highlighting the project’s exploratory nature and the variety of ideas being considered.
BAnQ aims to ensure that the data collected is not merely a public repository of creative works but rather a tightly controlled resource that accurately reflects Quebec’s identity. Marie Grégoire, BAnQ’s president and CEO, pointed out that the ultimate goal is to enrich AI models with Quebec-centric references, whether sourced from academic research or the business community.
Learning from Global Initiatives
This initiative is not isolated; similar efforts have emerged internationally. For instance, Sweden has successfully compiled large collections of Nordic-language texts to support the development of AI models tailored to Scandinavian languages. BAnQ will initially focus on its own collections before expanding to data from other sources, ensuring a robust foundation for the database.
The project aligns with recommendations from Quebec’s innovation council, which noted a significant deficiency in the volume of Quebec-related data available for AI training. Destiny Tchéhouali, an academic specialising in French-language AI, echoed these concerns, asserting that Quebec’s cultural output is significantly underrepresented in global AI datasets. He warned of the risks inherent in this oversight, particularly for Indigenous communities, who face even greater challenges in terms of representation and bias in AI outputs.
Navigating Copyright Challenges
As BAnQ progresses with its database, copyright issues are emerging as a vital concern within the cultural sector. Grégoire contended that the proposed platform could offer creators enhanced protections compared to the current landscape, which she likened to a “Wild West” scenario. “Data is being harvested for free, and that should not be the case,” she remarked, indicating a need for a more structured approach to data usage.
The database is envisioned as a centralised gateway, potentially simplifying the process of compensating creators whose works are utilised in AI training. However, there are apprehensions among artists regarding the implications of their contributions. Maxime Harvey, a postdoctoral researcher, noted that many creators fear that their involvement might ultimately threaten their livelihoods by supporting technologies that could replace traditional contracts.
A Vision for the Future
The feasibility study anticipates that the database could become operational by 2029, subject to reassessment after the experimental phase. The estimated budget for the next five years stands at approximately $10.5 million, covering both operating and capital expenses. The Quebec government has already allocated $340,000 for the feasibility study and an additional $750,000 for the project’s 12-month experimentation phase.
Why it Matters
This initiative represents a crucial effort to ensure that Quebec’s rich cultural landscape is accurately reflected in the rapidly evolving world of artificial intelligence. By building a database that centres on local content, BAnQ is not only addressing the current data gap but also paving the way for a more inclusive and representative future for AI technologies. The outcomes of this project could significantly impact how Quebec culture is perceived and integrated into global narratives, ultimately influencing the broader discourse on technology and identity.
