Hidden Layer Interaction: A Technique to Explore the Material of Generative AI
This pictorial describes the process of developing an interaction technique for directly engaging with the hidden layers of a generative AI model for image synthesis. First, we give some background to generative AI in HCI, arguing that current interaction techniques prevent us from directly interacting with the material of AI, foreclosing its use in design. Drawing on inspiration from the Computer Science field of feature visualization, we investigate the materiality of our prototype, a GAN model trained to generate fashion imagery, and show how Hidden Layer Interaction offers an alternative to standard prompting. In doing so, we illustrate how this change in approach leads to new forms of interaction with the internal semantics of generative AI, and demonstrate how one might use Hidden Layer Interaction to engage with AI as a material in design.
Research Questions / Practical Problems
Question signals indexed for this paper.
- 75%
Dream Lens: Exploration and Visualization of Large-Scale Generative Design Datasets
CHI '18· Generative AI (Text, Image, Music, Video) +1
- 75%
Exploring Challenges and Opportunities to Support Designers in Learning to Co-create with AI-based Manufacturing Design Tools
CHI '23· Generative AI (Text, Image, Music, Video) +1
- 75%
RoomDreaming: Generative-AI Approach to Facilitating Iterative, Preliminary Interior Design Exploration
CHI '24· Generative AI (Text, Image, Music, Video) +1
- 75%
AI-Assisted Causal Pathway Diagram for Human-Centered Design
CHI '24· Generative AI (Text, Image, Music, Video) +1
- 75%
Better Little People Pictures: Generative Creation of Demographically Diverse Anthropographics
CHI '24· Head-Up Display (HUD) & Advanced Driver Assistance Systems (ADAS) +1
- 75%
The Impact of Sketch-guided vs. Prompt-guided 3D Generative AIs on the Design Exploration Process
CHI '24· Generative AI (Text, Image, Music, Video) +1
- 75%
GenQuery: Supporting Expressive Visual Search with Generative Models
CHI '24· Generative AI (Text, Image, Music, Video) +1
- 75%
I-Card: A Generative AI-Supported Intelligent Design Method Card Deck
CHI '25· Generative AI (Text, Image, Music, Video) +1
- 75%
HapticGen: Generative Text-to-Vibration Model for Streamlining Haptic Design
CHI '25· Vibrotactile Feedback & Skin Stimulation +1
- 75%
Learning Personal Style from Few Examples
DIS '21· Generative AI (Text, Image, Music, Video) +1
Based on Jaccard similarity of research subtopics & professions (≥60%)