Descriptive Ethics of the AI Developer Community

I am currently recruiting participants from the artificial intelligence, machine learning, and data science community for research about ethics. If you are an AI/ML developer or a data scientist I hope you will consider talking with me. I can be reached via the links at right, or on Signal at plus one four one five seven two two seven two six four.

This research is part of my PhD in Interdisciplinary Normative Science at Maastricht University. It has been approved by the Ethical Review Committee Inner City Faculties under the identification number ERCIC_318_12_01_2021.

Why Research AI Developer Ethics?

Much of the dialogue about ethical AI is happening around developers about what they build. This study seeks to engage directly with developers. The goals of this research are to:

  1. Describe the experience of developers as distinct ethical agents;
  2. Identify the values and goals that motivate developers and data scientists; and
  3. Outline the restrictions and protections developers think are important for them to act ethically in their field.

What this Study is and What it is Not

This study is about professional ethics. It is about the ethical agency of developers and what they think about the future of their field. It is not about the ethics of algorithms or AI/ML deployments.

Who Can Participate

If you are an artificial intelligence engineer, machine learning engineer, or data scientist please consider participating in this study. This includes developers working with natural language processing, recommender systems, vision & multimodal sensing, general purpose APIs, chatbots & conversational AI, adaptive tutoring & mentoring, decision support, robotics, predictive analytics, deep learning, and artificial neural networks.

What Participation Looks Like

Participation is limited to a single 45-minute interview with me. Participation is anonymous. I will not ask questions about proprietary projects, algorithms, or your employer, and further request that you not discuss them either.

Data Collection

I will take manual notes during our interview. I will save the notes for no longer than ten years on a university network drive that only I can access. The notes will not be made available for any research other than what is described here.

Benefits & Risks

This research is intended to contribute to generalized knowledge about how AI/ML developers conceive of their ethical agency. There will be no direct benefits to participants during this phase. Physical risks associated with this project are no greater than what you encounter in everyday life. That is, talking on the phone or using internet-based conferencing.

I will take great care to avoid exposing participants to reputational or employment risks that could arise. In addition to anonymity, participants will be asked to speak only in general terms about how ethics are applied and discussed within the developer community. As noted above, I will not ask about proprietary matters and will ask that you also not them. I will also ask that all communication be conducted through personal email or encryption-based messaging (Signal). Furthermore, interviews will be scheduled during non-work hours to avoid any association with your official employment role or duties. These protocols are intended to reduce as much as possible whatever risk of reprisal may exist from an employer. This risk cannot be completely eliminated, however. If this risk seems too high for you, you may simply decline to participate.

Future Use

I will seek to present the findings at professional meetings and publish the results in relevant journals. I may also use the findings to inform future research to test ethical decision-making models. In all cases, your participation will remain anonymous.