Data-Biased Innovation: Directed Technological Change and the Future of Artificial Intelligence
This report analyzes how privacy regulations, particularly the GDPR, have influenced artificial intelligence (AI) innovation across different regions, revealing a shift from data-intensive to data-saving AI methods.
(Generated with the help of GPT-4)
| Quick Facts | |
|---|---|
| Report location: | source PDF |
| Language: | English |
| Authors: | Carl Benedikt Frey; Giorgio Presidente; Pia Andres |
| Publisher: | London School Of Economics; Institute For European Policy-making, Bocconi University; Oxford Internet Institute, University Of Oxford; Oxford Martin School, University Of Oxford |
| Geographic focus: | Global |
| Publication date: | December 2024 |
Methods
The research method involved constructing a novel taxonomy of AI technologies based on data intensity and analyzing patent applications from 57 countries. The study utilized keyword searches in patent databases and assessed the impact of GDPR exposure on patenting behavior through a Two-Way Fixed Effect model.
(Generated with the help of GPT-4)
Key Insights
The research examines the impact of privacy regulations on AI innovation trajectories globally, particularly focusing on the European Union's General Data Protection Regulation (GDPR). By analyzing patent applications from 57 countries across 76 industries from 2010 to 2021, the study identifies a significant shift from data-intensive AI methods, such as deep learning, to data-saving approaches, including transfer learning and synthetic data generation. The findings indicate that firms exposed to the GDPR altered their technological strategies, increasing data-saving patents while decreasing data-intensive ones. This shift is particularly pronounced among EU-based firms, which faced higher compliance costs due to the GDPR. The report also highlights geographic disparities in AI patenting, with the U.S. leading in data-saving innovations and China dominating in data-intensive patents. The research contributes to understanding how regulatory environments shape innovation patterns and suggests that privacy regulations may reinforce market concentration by favoring established firms over smaller entrants.
(Generated with the help of GPT-4)
Additional Viewpoints
Categories: 2024 publication year | global geographic scope | English publication language | artificial intelligence | privacy regulation | innovation | directed technical change | data-saving | data-intensive | GDPR | geographic heterogeneity | market concentration