With the increasing ease of manipulating videos, photos, and other evidence using artificial intelligence, a team of researchers in Canada is embarking on a mission to assist the legal system in distinguishing between authentic and fake content.
Comprising experts in technology and law from universities in Ontario and British Columbia, the team is dedicating the next two years to developing a user-friendly and free tool that can be utilized by courts, individuals navigating the justice system, and others to identify AI-generated content.
According to Maura Grossman, co-director of the project and a computer science professor at the University of Waterloo, the current reliance on commercial tools in courts is problematic. These tools are often unreliable, showing bias against non-native English speakers and generating false positives. Hiring experts for authentication is not only expensive but can also significantly delay court proceedings.
Speaking about the importance of accurate tools in court decisions, Grossman stressed the necessity for transparency and reliability. She emphasized that any tool used to determine the authenticity of evidence must provide clear explanations for its decisions and instill confidence in its accuracy.
Funded by the Canadian Institute for Advanced Research’s Canadian AI Safety Institute program, the research team aims to create a tool that can be utilized across North America. Yuntian Deng, an assistant professor at the University of Waterloo, highlighted the challenges posed by the rapid evolution of generative AI technology, stating that countering these advancements would be an ongoing process.
While the team anticipates developing an initial version of the tool within two years, Grossman acknowledged the need for continuous improvements and potential alternative funding sources beyond the grant’s duration. The focus is not solely on sophisticated criminal cases but also on assisting in everyday legal matters, including family court disputes.
In summary, the research team’s goal is to equip courts with a reliable tool to detect AI-generated content, ensuring the integrity of legal proceedings in the face of rapidly advancing technology.