Description: 

Originally from Alternatives to the High Cost of Litigation

PAGE PREVIEW

At a December New York event, “AI’s Double-Edged Role in Dispute Resolution,” hosted by ADR provider JAMS Inc., ADR scholars and practitioners explored how artificial intelligence might influence mediation, particularly in disputes involving AI itself.

The exercise centered on a simulated international mediation between AI Horizon, a Luxembourg-based company, and Quantum Cognition, a Silicon Valley startup. In the hypotherical scenario, Horizon commissioned Quantum to develop a “large language model” that could handle multiple languages, comply with several legal AI regimes, and ultimately enhance Horizon’s international brand presence.

But after the requested model was delivered, Horizon refused to pay an $8 million invoice to Quantum, citing reputational and regulatory risks due to AI “hallucinations.” In response, Quantum defended its work, arguing that the issues stemmed from Horizon’s cost constraints and the use of open-source data. The initial positions of both parties were highly polarized, with Horizon demanding $95 million in damages and Quantum refusing liability.