I prompted Anthropic’s Claude Sonnet 4.6
You are an expert on the Canadian library and archives systems with a broader perspective on the international scene. In particular, you have a profound understanding of the mandate, role and activities of Library and Archives Canada (LAC). Given its resource limitations, how well is LAC positioned to serve Canadians across the country? What three priority actions, perhaps outside-the-box ideas, would you recommend LAC do to improve?
Here’s the response.
Understanding LAC’s Mandate and Its Weight
LAC’s mandate is fourfold: to preserve the documentary heritage of Canada for the benefit of current and future generations; to serve as a source of enduring knowledge accessible to all, contributing to the cultural, social and economic advancement of Canada as a free and democratic society; to facilitate cooperation among communities involved in the acquisition, preservation and diffusion of knowledge; and to serve as the continuing memory of the Government of Canada and its institutions. That is an extraordinarily broad mandate for a single institution serving a geographically vast country of 40 million people.
How Well Is LAC Positioned to Serve Canadians?
Honestly, the answer is: inadequately, and getting worse. Here is a frank reckoning.
The resource squeeze is real and worsening. Budget 2023 affirmed a commitment to refocus government spending, which resulted in reductions to LAC’s operating budget of $3.6 million in 2025–26 and nearly $5.4 million in 2026–27. Further reductions from the Comprehensive Expenditure Review in Budget 2025 have compounded the pressure. In practical terms, LAC undertook a workforce adjustment exercise in August 2025, reducing the number of executive positions and reorganizing teams, while acknowledging that reductions in its operating budget will impact how it delivers its mandate.
The ATIP crisis is a fundamental failure of public accountability. Almost 80% of access to information requests completed by LAC did not comply with the time frames set out in the Act. The causes are structural, not merely managerial: LAC’s collection is primarily composed of physical, analog records — billions of pages originating from over 300 government institutions, existing in a wide variety of formats and dispersed across multiple storage facilities nationwide. One extraordinary case illustrates the fragility: an unexpected access request related to the Deschênes Part II Report required the work of approximately 50 employees from across the organization, forcing LAC to adjust its other priorities and causing delays across the board.
Digitization is moving far too slowly. LAC aims to add 500,000 new digitized and described images to its online holdings in 2026–27 — a respectable-sounding number until one recognizes that the analog backlog runs to billions of pages. Meanwhile, LAC’s capacity to describe its collections so they are discoverable to Canadians will be further impacted due to fiscal reductions.
Regional access remains a deep structural problem. LAC is fundamentally a centralized, Ottawa-centric institution. Researchers outside the National Capital Region face prohibitive barriers — travel costs, long wait times for records to be pulled, and limited remote service options. The new Ādisōke building, the joint facility LAC will share with the Ottawa Public Library opening in 2026, won’t have enough room for the national library’s books, so they will have to be trucked in to waiting readers. The Walrus, reviewing the situation, observed that LAC is making deals to survive and seem relevant — a pointed critique of an institution scrambling rather than leading. The WalrusThe Walrus
Researchers are being failed. Historian Stephen Azzi of Carleton University has been waiting more than two years for records needed for a book on the 1988 federal election. His colleague Norman Hillmer remarked that today it is the responsibility of researchers to justify their right to know what governments did, rather than the responsibility of government to put its records into sunlight. The Walrus
The broader ecosystem is also underserved. Small communities rarely have sufficient budgets to equip themselves with permanent resources for the preservation of their documentary heritage, and provincial and territorial governments struggle to provide even basic professional archival expertise to them. LAC, as the national anchor institution, is not effectively backstopping these gaps. Librarianship
To be fair, LAC is not standing still. It is harnessing AI through Transkribus to more quickly process digitized handwritten documents, and it is fostering partnerships with Internet Archive Canada to broaden free access to its collections. Its five organizational priorities for recent years have included deepening reconciliation commitments, stabilizing ATIP, transforming services for new audiences, advancing digitization, and integrating EDIA principles throughout. These are the right directions. But they are pursued with insufficient resources against a scale of challenge that demands a fundamentally different operating model. Canada.caCanada.ca
Three Outside-the-Box Priority Recommendations
Recommendation 1: Launch a “Distributed Heritage Hubs” Model — Federate, Don’t Centralize
The most structurally bold thing LAC could do is stop pretending it can serve a country the size of Canada from one building in Gatineau. Drawing on international models like New Zealand’s DigitalNZ and Sweden’s Digisam, LAC should establish a secretariat to develop an inventory of existing digitization projects across memory institutions, pool in-kind and financial resources, and create a unifying framework for multi-institutional digitization projects — but go further than the strategy documents suggest by formally delegating regional custodianship responsibilities. Library and Archives Canada
The idea: negotiate memoranda of understanding with 5–7 major regional universities and provincial archives (e.g., Dalhousie, Université de Montréal, University of Manitoba, University of Alberta, UBC) to serve as distributed regional nodes of the national collection. LAC would provide standards, training, digital infrastructure access, and a shared preservation framework. Regional institutions would provide physical space, local expertise, and community outreach — especially to Indigenous communities in their regions. A pan-Canadian committee of peers drawn from CARL, CULC, and other library associations should underpin the standards and protocols for such a network of regional preservation and storage facilities. Librarianship
This is not outsourcing — it is building a network rather than a warehouse. The British Library has done this with Legal Deposit Libraries; the Smithsonian does it with affiliate institutions. Canada has the university infrastructure to do it too. The payoff: dramatically improved regional access for Canadians who currently cannot afford to travel to Ottawa, a distributed risk model for disaster recovery, and meaningful engagement with provincial heritage communities.
Recommendation 2: Establish a National AI-Powered Transcription and Description Corps — A “Wikipedia Moment” for Canadian Archives
LAC is sitting on billions of undescribed and untranscribed pages that are essentially invisible to Canadians. The bottleneck is description — you can digitize a document, but if it has no searchable metadata, it might as well still be in a box. LAC is already using Transkribus, an AI tool that recognizes characters in handwritten and printed text, to transform analog finding aids into usable data. But this needs to be scaled by an order of magnitude through a radically more open and participatory model. Canada.ca
The recommendation: create a Canada Heritage Transcription Initiative — a structured, quality-controlled crowdsourcing program modeled on the U.S. Library of Congress’s “By the People” program and the Australian Newspapers digitization project at the National Library of Australia, both of which have generated millions of verified transcriptions from volunteers. LAC should partner with genealogical societies (there are hundreds across Canada, already motivated), university history departments (undergraduate courses could earn credit for verified transcription work), Indigenous language organizations (for records in Indigenous languages), and Francophone cultural organizations.
The AI layer does the heavy lifting on legible text; humans handle handwriting, marginalia, and quality control. The combination can process collections at a pace no internal team could match. This simultaneously solves the discoverability crisis, engages Canadians with their heritage in a direct and meaningful way, and generates a community of invested stakeholders who then become LAC’s loudest advocates for funding. It turns a resource limitation into a civic participation opportunity.
Recommendation 3: Decouple and Radically Accelerate the Proactive Disclosure Strategy — Make Government Records “Open by Default”
The ATIP crisis is not primarily a staffing crisis. It is an architecture crisis: a system designed around reactive, request-by-request disclosure of analog records cannot cope with a digital era. LAC continues to face backlog challenges due to the high volume and complexity of requests, which makes meeting legislated timelines difficult. The deeper fix is to change the model entirely. Canada.ca
LAC should champion — and itself model — a shift from “access on request” to “open by default, with exceptions.” Concretely, this means: pressing the Treasury Board and Privy Council to establish a policy that all federal records older than 20 years (with defined national security and privacy carve-outs) are automatically transferred to LAC pre-cleared for public release, rather than arriving classified and requiring decades of redaction work. For records older than 30 years, the default would be full public release with no ATIP trigger required.
This mirrors what the UK’s National Archives achieved with the 20-Year Rule and what several Nordic countries have long practiced. It fundamentally shifts the bureaucratic burden: instead of LAC spending tens of thousands of staff hours redacting records one request at a time, departments would be responsible for classifying and flagging sensitive material before transfer. As critic Norman Hillmer observed, the current system requires researchers to justify their right to know — that is the wrong default in a democracy. Oic-ci
The political case is also timely: in an era of declining trust in government institutions, proactively open archives are one of the most powerful signals of democratic health a government can send. LAC should be making this argument loudly to its Minister and to Parliament.
A Final Word
LAC is a world-leading national library and national archives with a rich, broad mandate, but it is seeking to strengthen its visibility and beneficial impact on the lives of Canadians. That is a careful understatement. The truth is that LAC is an institution of genuine national importance — the custodian of Canada’s collective memory — operating under conditions that make a full performance of its mandate impossible. The three recommendations above share a common philosophy: stop trying to do everything alone with inadequate resources, and instead become the architect and coordinator of a national heritage ecosystem. That pivot — from warehouse to network hub, from gatekeeper to open-platform enabler — is the strategic transformation LAC needs. Carl-abrc
COMMENT
That’s what Claude Sonnet 4.6 suggests. There are other AI facilities that would undoubtedly surface additional ideas, especially with a more refined prompt that better accounts for LAC’s limitations in authority. Has LAC leadership considered asking AI?
Is there anything in the response(s) of value?
Last year, on holiday on the Mediterranean island of Malta, I had the privilege of visiting the National Archives in Rabat. Three of us from Ottawa, courtesy of a connection through Canada’s former Librarian and Archivist, Ian Wilson, were given a conducted tour. They laid out some particularly significant documents.
Attend in person, or online.