Early in his career (late 1980s), Waldridge realized that computer networks would become ubiquitous and that this would intersect with AI; the natural question was: what happens when AI systems can communicate with each other across networks? This motivated his work on multi-agent systems.