AI tools used across the economy rely on well-trained AI models. It is critical that AI policies support robust training processes, so that those AI tools can work as intended.

Policymakers should not only promote the responsible development and use of AI systems, they should also ensure that policies allow companies to train AI models on a robust set of data. Training AI models on a significant quantity of data and different types of data can improve the accuracy and reliability of AI models, leading to better results for the companies and consumers that use AI tools built atop those models. Policies can inadvertently limit data used for AI training, including through laws on copyright, privacy, disclosure requirements, and cross-border access to data. BSA recommends specific actions that policymakers can take in each of these areas to promote robust training of AI models.

 Translation: Korean