Privacy and data protection regulations for AI using publicly available data: Clearview AI case
Bibliographic record
Abstract
Data are pivotal resources in artificial intelligence (AI) research and development. Acquiring enormous quantities of high-quality data is essential for improving AI performance. However, obtaining such data poses significant challenges, including high costs and accessibility barriers. A critical question arises: does using publicly available data from the Internet lead to violations of privacy and/or data protection laws? This paper examines privacy rights and data protection laws concerning the use of publicly accessible online data in AI research across the United States, Canada, Europe, and Australia. Focusing on recent controversies, particularly the Clearview AI case, it compares the legal and regulatory frameworks in these jurisdictions. The analysis highlights how digital governance in privacy and data protection must evolve in response to the growing demand for publicly available data in AI development. The central finding of this paper is that significant privacy and data protection risks and debates arise when publicly available data are used without adequate consent or legitimate purposes. The enforcement of these legal principles varies across countries, with substantial dependency on specific circumstances. The Clearview AI case, in particular, has exposed several ironies and unresolved issues surrounding data protection, privacy rights, and enforcement practices. These include disputes over the extraterritorial application of laws, inconsistent and impracticable legal enforcement, and settlements reached without judicial rulings. To foster a fair and trustworthy environment for AI development, compliance with privacy and data protection laws is critical. Achieving this requires more consistent, practical, and cooperative legal frameworks across nations to ensure safe and practical protection of personal data.
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 | 0.067 | 0.100 |
| Meta-epidemiology (narrow) | 0.000 | 0.001 |
| Meta-epidemiology (broad) | 0.001 | 0.002 |
| Bibliometrics | 0.002 | 0.003 |
| Science and technology studies | 0.011 | 0.034 |
| Scholarly communication | 0.016 | 0.017 |
| Open science | 0.003 | 0.011 |
| Research integrity | 0.012 | 0.017 |
| Insufficient payload (model declined to judge) | 0.003 | 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".