Farsight: Fostering Responsible AI Awareness During AI Application Prototyping

Abstract

Prompt-based interfaces for Large Language Models (LLMs) have made prototyping and building AI-powered applications easier than ever before. However, identifying potential harms that may arise from AI applications remains a challenge, particularly during prompt-based prototyping. To address this, we present Farsight, a novel in situ interactive tool that helps people identify potential harms from the AI applications they are prototyping. Based on a user's prompt, Farsight highlights news articles about relevant AI incidents and allows users to explore and edit LLM-generated use cases, stakeholders, and harms. We report design insights from a co-design study with 10 AI prototypers and findings from a user study with 42 AI prototypers. After using Farsight, AI prototypers in our user study are better able to independently identify potential harms associated with a prompt and find our tool more useful and usable than existing resources. Their qualitative feedback also highlights that Farsight encourages them to focus on end-users and think beyond immediate harms. We discuss these findings and reflect on their implications for designing AI prototyping experiences that meaningfully engage with AI harms. Farsight is publicly accessible at: https://PAIR-code.github.io/farsight.

Accepted to CHI 2024 (Best Paper, Honorable Mention). 40 pages, 19 figures, 5 tables. For a demo video, see https://youtu.be/BlSFbGkOlHk. For a live demo, visit https://PAIR-code.github.io/farsight. The source code is available at https://github.com/PAIR-code/farsight

References in corpus (22)

• Improving fairness in machine learning systems: What do industry practitioners need?
• Power to the People? Opportunities and Challenges for Participatory AI
• On the Educational Impact of ChatGPT: Is Artificial Intelligence Ready to Obtain a University Degree?
• Human Factors in Model Interpretability: Industry Practices, Challenges, and Needs
• Predictability and Surprise in Large Generative Models
• Jury Learning: Integrating Dissenting Voices into Machine Learning Models
• Seeing Like a Toolkit: How Toolkits Envision the Work of AI Ethics
• Investigating How Practitioners Use Human-AI Guidelines: A Case Study on the People + AI Guidebook
• A Systematic Review and Thematic Analysis of Community-Collaborative Approaches to Computing Research
• Walking the Walk of AI Ethics: Organizational Challenges and the Individualization of Risk among Ethics Entrepreneurs
• Supporting Human-AI Collaboration in Auditing LLMs with LLMs
• CrowdWorkSheets: Accounting for Individual and Collective Identities Underlying Crowdsourced Dataset Annotation
• Sensible AI: Re-imagining Interpretability and Explainability using Sensemaking Theory
• `It is currently hodgepodge'': Examining AI/ML Practitioners' Challenges during Co-production of Responsible AI Values
• Incorporating Ethics in Computing Courses: Perspectives from Educators
• "That's important, but...": How Computer Science Researchers Anticipate Unintended Consequences of Their Research Innovations
• Unpacking the Expressed Consequences of AI Research in Broader Impact Statements
• RECAST: Enabling User Recourse and Interpretability of Toxicity Detection Models with Interactive Visualization
• REAL ML: Recognizing, Exploring, and Articulating Limitations of Machine Learning Research
• StickyLand: Breaking the Linear Presentation of Computational Notebooks
• Interpretability, Then What? Editing Machine Learning Models to Reflect Human Knowledge and Values
• Angler: Helping Machine Translation Practitioners Prioritize Model Improvements

Cited by in corpus (9)

• Misty: UI Prototyping Through Interactive Conceptual Blending
• The Value-Sensitive Conversational Agent Co-Design Framework
• VeriPlan: Integrating Formal Verification and LLMs into End-User Planning
• Supporting Industry Computing Researchers in Assessing, Articulating, and Addressing the Potential Negative Societal Impact of Their Work
• Impact Assessment Card: Communicating Risks and Benefits of AI Uses
• From Hazard Identification to Controller Design: Proactive and LLM-Supported Safety Engineering for ML-Powered Systems
• Vipera: Towards systematic auditing of generative text-to-image models at scale
• Linting is People! Exploring the Potential of Human Computation as a Sociotechnical Linter of Data Visualizations
• Dynamite: Real-Time Debriefing Slide Authoring through AI-Enhanced Multimodal Interaction