An increasingly popular vehicle for charitable donations is the use of donor-assisted funds or DAFs. DAF grantors can receive the greatest possible tax deduction while maintaining a high level of control over which charitable organizations receive the funds and when those donations are triggered. DAF funds are unique in that the grantor no longer has ownership of the funds, but continues to have control. Furthermore, donations can be made electronically at the push of a button. Enabling not-for-profit organizations to identify DAF contributors presents several potential opportunities to maximize a charity’s funding. For example, organizations could customize their suggested gift to request larger donations from DAF directing contributors or use more personal channels to reach out. Identification of DAF contributors is of great interest to many different audiences including individuals responsible for donor base solicitations, those focused on donor acquisition, and individuals responsible for end-of-life bequests.

While there are well over one million DAF accounts, there is no publicly available list of DAF contributors. Instead, GIANT AI has taken on the challenge of predicting which individuals are most likely to be DAF grantors based on that individual’s similarity to grantors already known by the charitable organization. This is not an easy task. Every person has countless characteristics such as age, sex, geography, religious affiliation, level of education, donor history, various metrics of affluence, and many more that could be indicative of DAF grantor status. To explore this massive volume of data, GIANT uses artificial swarm intelligence, an analog of the natural collective intelligence observed in flocks of birds, schools of fish, and swarms of bees. In artificial swarm intelligence, thousands of intelligent agents simultaneously search a data landscape for patterns and communicate their findings to each other. When applied to predictive analytics, the swarm of intelligent agents subdivides a problem into homogeneous segments, creates a solution for each segment, and finally reunites the solutions into an ultimate equation that is more nuanced and more powerful than the sum of its parts.