![isabella-fischer-d5cdVcNb9Lc-unsplash](https://www.architectureandgovernance.com/wp-content/uploads/2024/07/isabella-fischer-d5cdVcNb9Lc-unsplash-678x381.jpg)
A new report by researchers at the Digital Trade and Data Governance Hub at the George Washington University found that the United States lags far behind its peer G-7 countries when it comes to comprehensive data governance policies. The fourth annual report finds the United Kingdom, Australia, Germany, France and New Zealand take the most comprehensive approaches to data governance at the national and international levels for the fourth year in a row. Spain, Brazil, the Netherlands, the European Union and Canada, respectively, round out the top 10.
These top-performing countries have adopted laws, strategies and policies such as data protection and freedom of information laws; established an open data portal and a personal data protection body; and held public consultations related to data governance. They have also created a stable environment in which data-driven sectors such as AI can flourish. According to the authors of the report, the citizens of these countries can trust that, generally, their data is protected, and their laws and institutions are evolving to meet the challenges of data-driven technologies such as AI.
By comparison, the United States has not kept up with its peers, because it lacks many of the key indicators of data governance, such as a federal data protection law or an agency devoted to protecting personal data.
“Data governance is important because there is no artificial intelligence without various types of data. Every day, Americans create and utilize huge troves of data and yet, U.S. policymakers are not doing nearly enough to govern data effectively,” says Susan Ariel Aaronson, founder and director of the GW Digital Trade and Data Governance Hub and one of the report’s co-authors. “For example, AI developers utilize personal, proprietary and public data to create variants of AI. Policymakers and governments need to find ways to ensure that the datasets underpinning AI are as accurate, complete and as representative as possible. In doing so, we’ll be able to create AI systems that are worthy of our trust.”
The GW Digital Trade and Data Governance Hub is the only organization in the world that maps and analyzes the governance of various types of data in 68 countries and the European Union. Researchers created a metric of 26 indicators built on six attributes, including strategies, policies, laws, and other institutional changes that are used to govern data.
Additionally, the researchers identified several key takeaways in this year’s report, including:
- Despite the worldwide interest in responsible AI, only 42% of the 68 countries and the EU analyzed in this report have adopted responsible AI initiatives. Here, the U.S. shines—with policies such as the President’s Executive Order on AI and the establishment of the new U.S. AI Safety Institute.
- The countries performing the best on the data governance metric are high income countries. The countries doing the worst tend to be lower income countries.
- Most countries in this sample sought public comment on one or more of their data governance initiatives. Yet these same countries did not, in general, respond to these comments. The report found only four of the 68 countries and EU listened and responded to public comments. In many countries, data governance is not yet a fully democratic process that provides a feedback loop between the government and the governed.