The House of Commons Standing Committee on Canadian Heritage met to study the impact of artificial intelligence on the cultural sector, hearing from five witnesses representing arts organizations, academia, and the music industry.
Marc-Olivier Ducharme, from Sporobole and ArtIA, argued that Silicon Valley's technology model creates a "digital feudalism" that exploits creators and threatens cultural diversity, particularly for francophone and Indigenous communities. He proposed government investment in artist-owned laboratories, cultural data trusts, and training programs to create a culturally responsible AI ecosystem, noting that Canada's $2.4 billion in AI investment did not directly fund culture.
Kelly Wilhelm, from OCAD University's Cultural Policy Hub, reframed the debate by arguing that government action valuing creative professionals does not stifle innovation. She called for sovereign and accessible data infrastructure, a national AI skills strategy from K-12 through professional sectors, clear transparency requirements, and ambitious investment in the full diversity of Canadian creators and creative SMEs.
Patrick Rogers, from Music Canada, warned that the current unlicensed use of music by AI companies represents a "Napster era" that must transition to a licensed model. He recommended upholding copyright law without text and data mining exceptions, requiring AI companies to disclose their training inputs, and cracking down on harmful deepfakes, citing the U.S. NO FAKES Act as a potential model.
Jennifer Brown, from SOCAN, stated that Canadian works are being stolen to train AI models without compensation, putting up to 24% of music creators' revenues at risk. She opposed any new copyright exceptions and urged mandatory transparency from AI developers about their training datasets, as well as clear labelling of AI-generated outputs, arguing that a vibrant licensing market can coexist with technical innovation.
Margaret McGuffin, from Music Publishers Canada, emphasized that nearly every song ever written by a Canadian songwriter has already been scraped by AI companies without permission. She called on the government to reject any text and data mining exception to copyright, arguing that music publishers are ready to license AI use if companies are required to disclose and maintain records of their training data.
The committee also heard a brief exchange where a Conservative member questioned whether government over-regulation, citing Bill C-18 as an example, could harm digital-first creators, while Ducharme maintained that investment in public, artist-governed infrastructure is needed.
AI-generated summary — may contain errors; verify against the official evidence.