Panel Discusses International Space Law, AI Role

Science & Technology Law Section

The panel, moderated by Dr. Stefan Geirhofer and featuring George Long and Nick Reese, examined how international treaties and national laws interact in space, the legal risks of managing intellectual property there, and the emerging role of AI in space applications

George Long started the discussion with highlights of the Outer Space Treaty. He clarified that it prohibits countries from claiming any part of outer space; assigns responsibility to nations and private actors for space activities; establishes liability for damage caused by space objects; addresses the jurisdiction for those held responsible; and requires consultations if a country’s space activity could interfere with another country. Long also shared how the Moon Treaty has only been ratified by 17 countries because it requires equitable profit sharing and multiple nations oppose it. Additionally, he discussed the Liability Convention, which sets liability standards for damage caused by space objects, and 18 U.S.C. § 1367, the federal law that punishes satellite interference. He stressed how vague and outdated these laws are when it comes to applying them to AI. Long questioned whether AI could be classified as a space object or activity, and who would be held accountable if the AI is autonomous when applying the Liability Convention.

Ultimately, there are many weaknesses in the international legal framework in relation to space due to the ambiguity of where AI fits into the application, the lack of updates, and minimal acknowledgement of privatized space involvement except for liability and communal use. Long believed that the international community should modernize to manage AI and update the Outer Space Treaty.

The risk for commercial companies to go beyond lower Earth orbit is too high due to the outdated international legal framework. While there is a need for critical infrastructure in space, if that were to happen, it would technically become common use under international laws, and AI would be required for maintenance because of space's inaccessibility. Nick Reese addressed the common point of sending data centers into space by stating it would be extremely expensive and unrealistic with the current state of legal liability. If we were to send data centers into space at the current cost of $1,500 per kilogram, it would amount to $1.4 trillion. Even if the cost drops to $20 per kilogram, the total would be $54.4 billion; private companies would be reluctant to bear these costs. For these reasons, commercial companies will not accept the current risks of going beyond lower Earth orbit because the infrastructure isn't there, and the legal exposure is too great.

Dr. Geirhofer concluded by highlighting three new applications of AI in space: AI in Earth observation, autonomous satellite steering, and quantum key distribution. Earth observation lowers costs by filtering satellite data to send only essential information. Autonomous steering raises unresolved liability questions of whether responsibility falls on the satellite owner or the AI model owner, since owning a space object typically makes one liable for any damage. Quantum key distribution offers another avenue for cost reduction. Across all three, a common thread emerged. AI is already reshaping what's possible in space, even as the legal frameworks struggle to catch up.

Endnotes

Author

View Bio →

Author

Emily Joinville

Committees

This content was produced by:

• Law Student Engagement Committee

Related Content