Cognitive assistants are a promising solution to the increasing complexity of large-scale collaboration. By providing support in data processing and decision making, they could lighten the cognitive load put on human collaborators. More precisely, conversational agents are especially suited to large-scale collaboration as they are user-friendly and could be integrated into existing collaboration tools. However, to be successfully integrated into the collaboration process, the assistant needs to be trusted. My doctoral research aims to identify which factors determine trust in a chatbot's advice during collaboration.