Purpose
PAI’s ABOUT ML research team is conducting 60-minute, one-on-one, online interviews to gather opinions from AI/ML industry practitioners and researchers to better understand their experience with using XAI tools or principles. The purpose of this study is to improve the explainability of AI/ML projects and gather feedback on PAI’s “XAI Toolsheet,” a documentation framework and a template to assess the functionality and usefulness of (XAI) tools. We define an XAI tool as any means for directly implementing an explainability algorithm (including but not limited to LIME, SHAP, AIX360, Amazon Sage Maker). The XAI Toolsheet includes a list of tool features informing the functionality and usability of XAI tools.
REQUEST TO PARTICIPATEProcess
This study will consist of an hour-long interview over Zoom. The interview contains two components:
- You will be asked a series of questions about your experience with using XAI tools or principles to improve the explainability of AI/ML projects at your organization or in an academic setting.
- You will be asked to complete a short activity to provide feedback on PAI’s XAI Toolsheet prototype.
Frequently Asked Questions
Interviews will be conducted from August to October 2022.
The study is being conducted by Dr. Surya Karunagaran, Research Fellow at the Partnership on Artificial Intelligence (PAI), and supervised by Sarah Villeneuve, Program Lead for ABOUT ML and Fairness, Transparency, and Accountability at PAI.
Participants will be compensated $50.00 (USD) for their participation in the interview.
There are no direct risks to participating in this study. Participation is completely voluntary, and you may avoid discussing any topics that you would rather not.
Data gathered from this study will be to refine PAI’s XAI Toolsheet and inform artifacts such as white papers, blog posts, and publications. PAI’s XAI Toolsheet will be made publicly available upon completion. The final research products will not contain any personally identifiable information, but will rather discuss general themes, and may include de-identified illustrative examples and quotes. The researcher will contact the participant for any permission to use a quote.
Interviews will be recorded and transcribed. Once the transcribed, recorded video files will be deleted. The transcript files will be password protected, stored on the researcher’s PAI issued laptop, and inaccessible to those outside the ABOUT ML research team at PAI.
Data gathered from the screening questionnaire or interviews will not be shared outside the research team at PAI. Each participant will be sent a copy of their interview transcript to add to, clarify, or redact any additional information from the interview.
We are seeking input from participants whose values and principles align with making AI/ML systems understandable to all the relevant stakeholders and having some familiarity with XAI tools. We are interested in gathering opinions from XAI tool developers and XAI tool users: including but not limited to data scientists, industry and academic AI/ML researchers, AI/ML UX Researchers, and policy-makers.
If you are interested in helping us improve our research, please express your interest in participating in our interview study through this form. We’ll follow up with you shortly based on your response with further details. Selected participants will receive $50 as compensation for participating in this study.
Please feel free to share this with anyone who might be interested. You can contact Surya (surya@partnershiponai.org) or Sarah (sarah.v@partnershiponai.org) directly with any questions.
Interest to Participate in PAI’s XAI Toolsheet Interview Study