The rise of digital donations and the potential of AI in fundraising are reshaping charitable giving. Digital donations - contributions made via online platforms, crowdfunding websites, mobile apps, or directly through a charity’s website - are increasing worldwide. Despite this increasing shift to digital platforms, the impact of human-machine interaction on donation behavior remains unclear. Research suggests that people respond differently to machines than to humans, often exhibiting weaker emotional and social reactions. However, little is known about how this affects actual donation decisions. This study addresses this gap through an online experiment investigating how interactions with digital agents (chatbots) compared to human interactions influence donation behavior. Using a between-subjects design, participants are randomly assigned to one of three donation-reporting-conditions: a form, a chatbot conversation, or a human chat. In the first phase, participants earn money through a real-effort task; in the second phase, participants decide in a dictator game, what amount of their earned money they want to donate to a charity. Based on previous research, we hypothesize that donations will be higher when reporting to a chatbot or a human compared to a form, with the highest levels in the human interaction condition. Possible underlying mechanisms include social image concerns, social approval, and perceived closeness. This research contributes to digital philanthropy and human-machine interaction literature. Insights will help charities and non-profit organisations refine their fundraising strategies by understanding how digital communication channels influence donation behavior.