Behavioral Use Licensing for Responsible AI
share
Scientific research and development relies on the sharing of ideas and artifacts. With the growing reliance on artificial intelligence (AI) for many different applications, the sharing of code, data, and models is important to ensure the ability to replicate methods and the democratization of scientific knowledge. Many high-profile journals and conferences expect code to be submitted and released with papers. Furthermore, developers often want to release code and models to encourage development of technology that leverages their frameworks and services. However, AI algorithms are becoming increasingly powerful and generalized. Ultimately, the context in which an algorithm is applied can be far removed from that which the developers had intended. A number of organizations have expressed concerns about inappropriate or irresponsible use of AI and have proposed AI ethical guidelines and responsible AI initiatives. While such guidelines are useful and help shape policy, they are not easily enforceable. Governments have taken note of the risks associated with certain types of AI applications and have passed legislation. While these are enforceable, they require prolonged scientific and political deliberation. In this paper we advocate the use of licensing to enable legally enforceable behavioral use conditions on software and data. We argue that licenses serve as a useful tool for enforcement in situations where it is difficult or time-consuming to legislate AI usage. Furthermore, by using such licenses, AI developers provide a signal to the AI community, as well as governmental bodies, that they are taking responsibility for their technologies and are encouraging responsible use by downstream users.
READ FULL TEXT