Builders face challenges in creating and testing forecasting bots
Programming Prescience: Issues in Human Construction of Forecasting Bots
Human-Computer InteractionArtificial Intelligence
Summary
Forecasting bots use AI to predict the future by learning from past data. The researchers talked with 15 people who built these bots to understand their experiences. They found it is hard to teach bots to find correct facts, to make logical predictions, and to know how well the bot really works. Builders often tried complicated solutions that didn’t help and tested their bots in incomplete ways. This study helps identify where new tools and processes could make building forecasting bots easier and better.
What this means in practice
- •For forecasting developers: Improve tool design to support builders in tuning and testing forecasting bots more effectively based on detailed user challenges.
- •For ai product managers: Guide development workflows that better integrate human and AI strengths for forecasting systems, reducing wasted complexity and opportunistic tests.
Authors
Priya Dhawka, Benjamin Wilson, Andrew Head
Abstract
AI models' growing capacity for complex reasoning has made them attractive tools for forecasting, or the systematic prediction of the future based on past and present data. However, should AI forecasting systems be adopted, human builders will shoulder the work of building and evaluating them. In this paper, we explore what this work looks like with a qualitative interview study with 15 builders of forecasting bots, sampled from among successful participants in a recent AI forecasting competition. Our analysis surfaces the challenges around (1) tuning a bot to acquire factual research, (2) implementing coherent reasoning, and (3) understanding how well one's bot actually does. During their work, builders frequently introduced complexity that did not pay off, and tested in ways that were opportunistic and piecemeal. Our detailed account of their challenges helps clarify what problems should be targeted in upcoming innovations in process, tools, and human-AI teaming for forecasting.