Home

Work

Contact

About

Home

Work

Contact

About

Home

Work

Contact

About

Eidolon

One week , 2026

Team:

Failenn Aselta

The Challenge +

Project Overview

What happens when the internet is no longer human-navigated, but optimized for machine-to-machine interaction? Some users may prefer direct, human-led engagement and remain skeptical of automation. while other may welcome AI assistance. The future internet will likely be hybrid space of, allowing individuals to choose how much control they delegate and how much they retain. Eidolon uses AI as a visual tool to not only showcase its actions, but make it an easier place for those who fear its new shape.

Cindy

75 year old Cindy struggles to understand AI agents. She has only recently become comfortable using a computer, and to her these systems feel intimidating. In a world that is evolving rapidly, she longs for stability and prefers not to rely on AI.

75 year old Cindy struggles to understand AI agents. She has only recently become comfortable using a computer, and to her these systems feel intimidating. In a world that is evolving rapidly, she longs for stability and prefers not to rely on AI.

Ashleigh

22 year old Ashleigh is an AI whiz. She loves the idea that her vacations and social media posts can be streamlined using AI. However, she worries that AI does not always provide accurate information, she wishes there were a clearer way to understand AI.

22 year old Ashleigh is an AI whiz. She loves the idea that her vacations and social media posts can be streamlined using AI. However, she worries that AI does not always provide accurate information, she wishes there were a clearer way to understand AI.

How might we improve trust in AI by 40% and decrease cognitive load by 63%?
Develop agents who not only showcase their interactions visually, but clean up environments to make the web a safer place for all to use.



How might we improve trust in AI by 40% and decrease cognitive load by 63%?
Develop agents who not only showcase their interactions visually, but clean up environments to make the web a safer place for all to use.



Case Studies

The low-fidelity iteration of this project was developed efficiently, as interface conventions within AI platforms and video game systems are well established and supported by substantial research. I conducted comprehensive analysis of current large language models to inform the layout strategy, synthesizing industry best practices and identifying key strengths to integrate into the design.

The low-fidelity iteration of this project was developed efficiently, as interface conventions within AI platforms and video game systems are well established and supported by substantial research. I conducted comprehensive analysis of current large language models to inform the layout strategy, synthesizing industry best practices and identifying key strengths to integrate into the design.

The low-fidelity iteration of this project was developed efficiently, as interface conventions within AI platforms and video game systems are well established and supported by substantial research. I conducted comprehensive analysis of current large language models to inform the layout strategy, synthesizing industry best practices and identifying key strengths to integrate into the design.

Lo-Fi Wireframe

User's Journey

A solution emerged that allows customization of an AI agent’s style. As AI continues to progress, one major component may be the use of Bitcoin for transactions instead of relying on the selling of user data. This could lead to the development of an AI agent that maintains the current structure of the web for a small fee rather than monetizing personal data. Alternatively, the AI agent could act as a liaison throughout the web, securing transactions and managing interactions on behalf of the user.

Cindy’s Journey

Ashleigh’s Journey

I opted for a stylized persona over a 3D mesh because pure abstraction often feels cold, leading to a "trust gap." While meshes signal objectivity, they fail to build the emotional rapport necessary for complex interactions. By utilizing a gold-skinned, non-gendered character, I bypass the Uncanny Valley—signaling the agent is not human while leveraging the "Buddha-like" associations of deep reasoning. This stylized form allows for "anthropometric trust," where the agent can visually "look" at data or point to decision paths, making its internal state legible.

I opted for a stylized persona over a 3D mesh because pure abstraction often feels cold, leading to a "trust gap." While meshes signal objectivity, they fail to build the emotional rapport necessary for complex interactions. By utilizing a gold-skinned, non-gendered character, I bypass the Uncanny Valley—signaling the agent is not human while leveraging the "Buddha-like" associations of deep reasoning. This stylized form allows for "anthropometric trust," where the agent can visually "look" at data or point to decision paths, making its internal state legible.

Throughout each design decision, the corresponding execution was guided by extensive research into the optimal use of the software. As one moves through each research redline, the intention behind the overall composition, color palette, and micro-interactions becomes clear. Each element was designed to minimize cognitive load while maximizing user clarity and satisfaction.

Throughout each design decision, the corresponding execution was guided by extensive research into the optimal use of the software. As one moves through each research redline, the intention behind the overall composition, color palette, and micro-interactions becomes clear. Each element was designed to minimize cognitive load while maximizing user clarity and satisfaction.

Conclusion

By treating the web as a fragmented and high-friction "digital territory" rather than a set of static pages, Eidolon applies systematic sanitization to the chaos of future browsing. The project moves the user experience from a state of "Environmental Anxiety" to "Systemic Trust" by utilizing a Recursive Sanitizer to prune predatory UI and dark patterns in real-time. This intervention proves that when agentic workflows are anchored in clinical research, such as Hyunjin Song’s findings on processing fluency, we can engineer a 63% reduction in cognitive load, transforming the internet into a predictable, low-stress workspace that facilitates human success rather than exhausting it.


The "Uncanny Valley" of AI often stems from unpredictability; when an agent's actions are opaque, user trust collapses. Eidolon bridges this gap by introducing a 2D Negotiation Canvas that makes the agent’s decision-making process visible and legible to the human eye. By grounding the interface in anthropometric trust and haptic authorization gates, this project demonstrates that we can solve the "black box" problem of AI. The result is a system that doesn't just perform tasks, but actively safeguards the user's focus, driving a 40% increase in trust by returning agency to the human at the center of the loop.

By treating the web as a fragmented and high-friction "digital territory" rather than a set of static pages, Eidolon applies systematic sanitization to the chaos of future browsing. The project moves the user experience from a state of "Environmental Anxiety" to "Systemic Trust" by utilizing a Recursive Sanitizer to prune predatory UI and dark patterns in real-time. This intervention proves that when agentic workflows are anchored in clinical research, such as Hyunjin Song’s findings on processing fluency, we can engineer a 63% reduction in cognitive load, transforming the internet into a predictable, low-stress workspace that facilitates human success rather than exhausting it.


The "Uncanny Valley" of AI often stems from unpredictability; when an agent's actions are opaque, user trust collapses. Eidolon bridges this gap by introducing a 2D Negotiation Canvas that makes the agent’s decision-making process visible and legible to the human eye. By grounding the interface in anthropometric trust and haptic authorization gates, this project demonstrates that we can solve the "black box" problem of AI. The result is a system that doesn't just perform tasks, but actively safeguards the user's focus, driving a 40% increase in trust by returning agency to the human at the center of the loop.

Final Click Through

Ethical Considerations

Ethical Considerations

This project situates itself within broader debates in AI ethics, acknowledging that contemporary AI systems continue to raise concerns around surveillance capitalism, algorithmic manipulation, opaque decision-making, and concentrated data power. Rather than assuming AI is inherently neutral or benevolent, the proposal recognizes its potential to subtly influence user behavior and prioritize commercial interests. In response, it introduces a dual-agent framework as a structural safety network, where a personal AI advocates for the user by monitoring persuasive tactics, limiting unnecessary data exchange, and making algorithmic influence visible through a protective interface layer. By embedding safeguards directly into the system’s architecture, the project argues that trust in AI must be constructed through transparency, distributed control, and user autonomy, not convenience alone.

This project situates itself within broader debates in AI ethics, acknowledging that contemporary AI systems continue to raise concerns around surveillance capitalism, algorithmic manipulation, opaque decision-making, and concentrated data power. Rather than assuming AI is inherently neutral or benevolent, the proposal recognizes its potential to subtly influence user behavior and prioritize commercial interests. In response, it introduces a dual-agent framework as a structural safety network, where a personal AI advocates for the user by monitoring persuasive tactics, limiting unnecessary data exchange, and making algorithmic influence visible through a protective interface layer. By embedding safeguards directly into the system’s architecture, the project argues that trust in AI must be constructed through transparency, distributed control, and user autonomy, not convenience alone.

Existing

Existing

Existing

Thinking

Thinking

Thinking

Discussing

Discussing

Discussing

Idle

Idle

Idle

Running

Running

Running

Confused

Confused

Confused

Bibliography


Brennan-Marquez, Kiel, and Stephen Maher. "Micro-Costs." Georgetown Law Journal. May 2025.

https://www.law.georgetown.edu/georgetown-law-journal/wp-content/uploads/sites/26/2025/05/Brennan-Marquez_Maher_Micro-Costs.pdf.


Liao, Yan, et al. "The Influence of Interface Layout and Color on the User Experience of Older Adults." International Journal of Environmental Research and Public Health 19, no. 15 (2022): 9251

. https://www.mdpi.com/1660-4601/19/15/9251.


Nass, Clifford, Jonathan Steuer, and Ellen R. Tauber. "Computers are Social Actors." CHI '94: Proceedings of the SIGCHI Conference on Human Factors in Computing Systems.

https://www.researchgate.net/publication/221517042_Computer_are_social_actors.


Schroeter, Ronald, et al. "Agent vs. Avatar: Comparing Embodied Conversational Agents Concerning Characteristics of the Uncanny Valley." ResearchGate. 2020.

https://www.researchgate.net/publication/344153513_Paper_Presentation_Agent_vs_Avatar_-_Comparing_Embodied_Conversational_Agents_Concerning_Characteristics_of_the_Uncanny_Valley.

Song, Hyunjin, and Norbert Schwarz. "If It's Hard to Read, It's Hard to Do: Processing Fluency Affects Effort Prediction and Motivation." Psychological Science (2008).

https://dme.engin.umich.edu/wp-content/uploads/sites/204/2014/09/Song-Schwarz-2008.pdf.


Wortham, Robert H. "Robot Transparency: Improving Understanding of Intelligent Behaviour for Designers and Users." ResearchGate. 2017.

https://www.researchgate.net/publication/318538795_Robot_Transparency_Improving_Understanding_of_Intelligent_Behaviour_for_Designers_and_Users.


Anthropic. "Model Context Protocol." Anthropic News. 2024.

https://www.anthropic.com/news/model-context-protocol.

APA. "Stress in America 2025." American Psychological Association. 2025.

https://www.apa.org/pubs/reports/stress-in-america/2025.


Gartner. "Gartner Predicts Search Engine Volume Will Drop 25% by 2026." Gartner Press Release. February 19, 2024.

https://www.gartner.com/en/newsroom/press-releases/2024-02-19-gartner-predicts-search-engine-volume-will-drop-25-percent-by-2026-due-to-ai-chatbots-and-other-virtual-agents.


Google Design. "Gradients: Designing the Visual Language of Gemini AI." Google Design. 2024.

https://design.google/library/gemini-ai-visual-design.


MIT Media Lab. "Ambient Intelligence Research Group." MIT Media Lab.

https://www.media.mit.edu/research/groups/ambient-intelligence.


Pew Research Center. "Growing Public Concern About the Role of Artificial Intelligence in Daily Life." Pew Research. August 28, 2023.

https://www.pewresearch.org/science/2023/08/28/growing-public-concern-about-the-role-of-artificial-intelligence-in-daily-life/.


Stanford Institute for Human-Centered AI. "Demographic Stereotypes in Text-to-Image Generation." Stanford HAI. 2024

. https://hai.stanford.edu/news/demographic-stereotypes-text-image-generation.


Apple. "Human Interface Guidelines: Materials (Liquid Glass)." Apple Developer.

https://developer.apple.com/design/human-interface-guidelines/materials.


Nielsen Norman Group. "Aesthetic-Usability Effect." NN/g.

https://www.nngroup.com/articles/aesthetic-usability-effect/.


Nielsen Norman Group. "Designing UX for Seniors." NN/g.

https://www.nngroup.com/articles/designing-ux-seniors/.

Failenn Aselta