GeoAI and the Law Newsletter
Keeping geospatial professionals informed on the legal and policy issues that will impact GeoAI.
Summary of Recent Developments in GeoAI and the Law
This issue will focus on legal and policy developments outside of the U.S. Given the geospatial community’s global nature, geospatial professional often must consider the impact of laws and policies across the globe. Countries are taking a very different approach to AI: some are proactively considering laws and regulations, while others are taking more of a measured approach, for example developing policies or convening groups of experts, either intentionally or due to internal politics and/or other priorities. Nonetheless, almost every nation is considering both the benefits and the risks of AI to their citizens.
Recommended Reading
Italy’s Privacy Regulatory asks OpenAI for detail on Sora (translated version)
According to the report, the Italian agency responsible for data protection has requested information about the data used to train the text-to-video application.
International AI rights treaty hangs by a thread
Reporting suggests the U.S. is working to exclude the private sector from being subject to the treaty being negotiated by the Council of Europe.
EU Passes AI Rules Despite Doubts It Got the Right Balance
EU passes comprehensive law regulating AI.
NEW AI LAW ALERT!: Utah to Add Disclosure Requirements for Artificial Intelligence
Bill will require disclosure for certain AI applications that interacti with consumers.
Reddit Says FTC Investigating Deals to License Data and Train AI Models
FTC is reportedly investigating Reddit’s arrangements to license its data to train AI models.
The Deep Dive
Each week, the Deep Dive will provide a detailed analysis on how a particular legal matter (e.g., a case, law, regulation, policy) pertaining to AI could impact the geospatial community and/or GeoAI in particular.
The EU passing of the AI Act is certainly the most significant legal and policy development the past week. It is intended to balance innovation with perceived risks. As it will take several years before the AI Act is fully implemented, a more detailed analysis of its impact on GeoAI will follow in subsequent issues. However, companies that are selling GeoAI applications in Europe for critical infrastructure should already be determining whether they will be subject to its provisions that regulate ‘high-risk’ AI systems.
Two countries that wish to lead in AI - China and India - have taken different approaches. For example, over the past several years, China has published several regulations on algorithms and generative AI. As noted in last weeks’ edition, India recently published a draft AI policy. However, the situation is evolving in both countries, as China is reportedly considering a comprehensive law and India has already reportedly changed its draft policy.
Elsewhere, Brazil has convened a meeting of experts in anticipation of developing a legal framework it reportedly intends to present to the United Nations. The African Union Development Agency reportedly developed a blueprint of AI regulations for African governments to consider adopting within their own countries. In the Middle East, the Saudi Data & Artificial Intelligence Agency (SDAIA) published an Artificial Intelligence Ethics Framework (2023) and the UAE’s Council for Artificial Intelligence has published several guidelines on AI matters, including ethics and use of AI in government services.
In a legal context, definitions matter. This is particularly true in the context of AI, since the definition of an artificial intelligence system differs between countries. As a result, any analysis of an AI law or regulation should begin with a review of the definitions to determine what the law applies to (and who is covered). This is particularly true in the context of GeoAI, given its potential for dual-use. Organizations may find that a certain application using GeoAI can be used for one purpose, but is heavily regulated for another.