Project Summary

Issues like the erosion of analytical faculties and real-world skills are becoming more common due to the automation bias that many AI tools are developing. There's evidence that users of automation tools develop an inability to monitor their reliance on such tools (automation agents). Notably, users develop a real dependence on the agent due to their inability to work independently. The vast majority of people that work with AI do so through AI chatbots. For this reason, we can carry out large observational studies. Since users of AI chatbots usually have a high degree of automation bias, it is almost certain that this bias is reflected in their digital biometrics. In this case, digital biometrics is used to capture the physical symptoms of automation bias. This project ultimately aims to develop a software tool that offers the real-time diagnostics of automation bias.

What are this project's goals? How will you achieve them?

The core goal of this project is to identify the behavioral patterns associated with automation bias. The technical goal of this project is to build a data collection tool. The data collection tool will be based on an AI chatbot and user movement will be tracked.

How will this funding be used?

The budgeting plan consists of three main allocations: (1) cost of submission of the research project to an IRB (~$1,000); (2) payment to research subjects on Prolific or MTurk to collect research data (~$1,500); (3) purchase of a Claude AI subscription.

Who is on your team? What’s your track record on similar projects?

This is a self-directed project. This is my first self-directed project, but as an early adopter of technology, I am particularly interested in the rapid absorption of AI and the corresponding decline in critical engagement. I complete the majority of the required skills for this project, including the creation of the web interface, the integration of the chatbot APIs, and the writing of the frontend telemetry tracking code. I plan to conduct a thorough and ethical experiment to offer open-source research in this area and this motivates me greatly.

How much money have you raised in the last 12 months, and from where?

I have not raised any money in the last year. I have budged my own funds, as this project is completely self-initiated. Currently, this is my primary method of funding the project through my application to Manifund.