Hacking AI Governance: Exploring the Democratic Potential of Canada's Algorithmic Impact Assessment
Bibliographic record
Abstract
Amid growing concern over the adoption of artificial intelligence systems, algorithmic impact assessments (AIAs) have increasingly been proposed as a means of measuring and mitigating the impacts of AI. Proposed AIA methods vary significantly in their approaches, but even within this heterogeneous group, the AIA tool released by the Government of Canada in 2019 stands out. This AIA tool—an open-source, online questionnaire platform—represents one of the first-ever attempts at putting principles of “responsible AI” into practice. In this research-creation thesis, I explore Canada’s AIA tool as a media object, looking at the online questionnaire as a strategic opportunity to intervene in the growing debates about AI governance. Building on methods in critical making and civic hacking, this project includes the creation of both a critical guide to the AIA tool (aia.guide) and a series of “AIA hackathon” workshops designed to explore the tool’s use by the Government of Canada and its potential in the broader AI governance context. Informed by a deep ambivalence over the technology (Bucher 2019), I argue that the AIA tool is largely performative but also represents an important site for tactical intervention. In particular, I argue that collaborative processes of questionnaire design may prove to be effective methods for participatory and community-based AI governance.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.021 | 0.051 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.003 | 0.006 |
| Science and technology studies | 0.021 | 0.021 |
| Scholarly communication | 0.017 | 0.004 |
| Open science | 0.002 | 0.007 |
| Research integrity | 0.002 | 0.003 |
| Insufficient payload (model declined to judge) | 0.004 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".