Mapping geographical biases of AI principles
Bibliographic record
Abstract
This research investigates geographical biases of AI principles. Although we assume countries participate equally in discussions for the direction of AI development and deployment, value biases are worrisome. Some economically dominant countries might codify most of the AI principles for their regional benefit. On the other hand, other less powerful regions might be underrepresented or unrepresented. To measure a geographical bias of principle declarations, the study collected AI principles (n=94) by filtering datasets of former research and analyzed them using computational tools. The research team found that institutions from 25 countries (12.95%) directly declared their own AI principles out of 193 countries of the world. Among leading countries for AI principle declarations, three main regions mostly participated in AI ethics and technology principle discussions: region 1 is North America (the U.S. and Canada), region 2 is Europe (the U.K. and the E.U.), and region 3 is East Asia (China, Japan, and South Korea). These three regions produced 89.36% (about 90%) of the AI principle declarations. This geographical bias implies biases of coded values. Since participation itself is notably correlated to educational, economic, and technological backgrounds, further developments based on AI principles might underrepresent technically developing countries’ viewpoints.
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 | 0.018 | 0.112 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.007 | 0.013 |
| Science and technology studies | 0.002 | 0.002 |
| Scholarly communication | 0.003 | 0.003 |
| Open science | 0.001 | 0.004 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.004 | 0.001 |
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".