The Impact of AI and Cross-Border Data Regulation on International Trade in Digital Services: A Large Language Model
Bibliographic record
Abstract
The rise of artificial intelligence (AI) and of cross-border restrictions on data flows has created a host of new questions and related policy dilemmas. This paper addresses two questions: How is digital service trade shaped by (1) AI algorithms and (2) by the interplay between AI algorithms and cross-border restrictions on data flows? Answers lie in the palm of your hand: From London to Lagos, mobile app users trigger international transactions when they open AI-powered foreign apps. We have 2015-2020 usage data for the most popular 35,575 mobile apps and, to quantify the AI deployed in each of these apps, we use a large language model (LLM) to link each app to each of the app developer's AI patents.(This linkage of specific products to specific patents is a methodological innovation.) Armed with data on app usage by country, with AI deployed in each app, and with an instrument for AI (a Heckscher-Ohlin cost-shifter), we answer our two questions. (1) On average, AI causally raises an app's number of foreign users by 2.67 log points or by more than 10-fold. (2) The impact of AI on foreign users is halved if the foreign users are in a country with strong restrictions on cross-border data flows. These countries are usually autocracies. We also provide a new way of measuring AI knowledge spillovers across firms and find large spillovers. Finally, our work suggests numerous ways in which LLMs such as ChatGPT can be used in other applications.
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 | 0.009 | 0.028 |
| Meta-epidemiology (narrow) | 0.001 | 0.001 |
| Meta-epidemiology (broad) | 0.002 | 0.003 |
| Bibliometrics | 0.002 | 0.002 |
| Science and technology studies | 0.002 | 0.006 |
| Scholarly communication | 0.008 | 0.010 |
| Open science | 0.003 | 0.004 |
| Research integrity | 0.003 | 0.004 |
| Insufficient payload (model declined to judge) | 0.018 | 0.002 |
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".