Sonder 2040: Futures Research & Design Fiction Film
A 2025 Master of Human-Computer Interaction Capstone Research & Design Project for 99P Labs, Honda Research Institute USA
THE SOLUTION
Augmenting human perception to see—and decide—beyond the visible in 2040.
Sonder 2040 imagines a decentralized AI decision-making tool, Sonder, that is designed to augment human perception in high-stakes organizational contexts.
By surfacing invisible factors—ethical tradeoffs, systemic risks, and long-term consequences—and connecting perspectives across a network of agents, Sonder empowers tomorrow’s leaders to make more balanced, transparent, and adaptive choices.
In a future where complexity and pressure will only intensify, Sonder offers a vision for ethical resilience: helping organizations avoid collapse by seeing further, together.
[Authored by ChatGPT for clarity]
THE PROBLEM
The world of 2025 feels ethically motivated by capitalism and centralized technology practices that put most people at a disadvantage.
Honda Research Institute, USA (HRI USA) is uniquely positioned to create a brighter future for leaders and every day consumers, especially in the realm of artificial intelligence.
Through 99P Labs, HRI USA’s student innovation hub and the MHCI Capstone Program at Carnegie Mellon University’s Human Computer Interaction Institute, exploring the future of ethical human-AI relationships that inspire and support the mobility of everyday people for decades to come is possible.
The 2025 MHCI x 99P Labs Futures Research uses human-centered and speculative practices to understand how an AI system can augment the unique values of individual end users to encourage ethical decision-making in 2040.
If the goal is to build an AI system that could intake unique end-user values to make ethical decisions, then a holistic understanding of human life at the individual and cultural level would be necessary.
This includes understanding abstract and concrete aspects of human values and behavior.
RESEARCH
Realm of the User: Product Design Lens
Understanding the current state of decision-making required interviewing high-stakes decision-makers in real life to better understand their self-reflections on decision-making and how AI could augment their process via an interview and speed dating session.
High-stakes decision-makers (n = 8) from various domains were interviewed on their current decision-making processes and shown a collection of storyboards via speed dating to uncover their reactions to the AI system ranging in levels of autonomy and ethical responsibility.
Participants were asked to recall how they make decisions as a high-stakes decision-maker at work including questions like, “what are current decision-making challenges?” and “have you ever made a decision that had a bad impact?”
“The decision-maker’s proximity to the problem they are trying to solve will dictate how much information they have when solving the problem.”
CEO, Social Services
Participants were shown a series of storyboards, including the final storyboard exploring how Sonder could balance human control, ethical choices & creative augmentation during the decision-making process:
“The CFO uses Sonder to make proactive decisions and allows the AI system to make decisions on his behalf with his pre-approval.”
“What would it be like to show someone a 30 second video of each possible future as a story?”
Senior Designer, Technology Design
Human-Centered Research Insights
Clear Needs
- Decision-making is a nonlinear process
- Data needs to be trustworthy–whether it’s originating from team members or the organization
- Data that comes from tangible or knowable sources appears easier to trust
Under the Surface
- Human nature is limited in perception, so visible or material factors are easy to comprehend and sense
- Decision-making is also influenced by “invisible” factors that exist outside of data itself like time, the abstraction of trust or emotional responses
World at Large: Speculative Design Lens
Designing the future is highly variable and requires many research methods to plausibly understand what is just over the horizon, especially when considering the future of ethical human-AI relationships. The use of PESTLE analyses to understand the past and present was the key to unlocking patterns that would influence the possible futures leading to 2040’s ethical human-AI relationships landscape.
Looking backwards at the steel industry’s rise and fall throughout 20th Century Pittsburgh and comparing the beginning of AI in the 1950s to the AI consumer revolution of the 2020s, inflection points revealing the impact of new technologies on work and leisure shaped the worldbuilding of the Sonder 2040 design fiction video final cut.
An initial worldbuilding rough cut was created and tested with viewers (n = 7) as a quick speculative design prototype test.
Viewers responded better to aspects of the worldbuilding that connected more to their perception of the present day and overly dystopian language that felt off-base distracted them.
Speculative Research Insights
- 2025’s unstable geopolitical climate & global supply chain/tariff policy disruptions could lead to 2040’s need for diverse resource access, including domesticating raw material extraction & recycling
- 2025’s deregulation and/or defunding of centralized services creates new value opportunities in the private sector could lead to 2040’s “fragmented” or decentralized regions and/or local communities socially, economically & politically
- 2025’s introduction of multi-agent AI systems and agentic AI into the world for consumers & AI’s widespread inclusion in work and governance could lead to 2040’s normalization of sophisticated consumer-facing complex AI systems leads to increased AI literacy and more open-source AI spin-offs by various user groups
DESIGN
Logline In 2040’s Pittsburgh, Appalachia Region, a disillusioned chief resources officer at a manufacturing & robotics company shares her hopes for a more ethical resource extraction plan with Sonder, a contextual and trans-regional multi-agent AI system.
Narrative Design Principles
The AI system, Sonder, should be able to capture visible and invisible decision-making factors
Sonder needs to consult trusted data to support decision-making, including how it’s calculated and where the data originated from
Sonder needs to help augment and visualize the user’s understanding of the decision-making environment
IMPACT
Using Emotion to Explore the Future of Ethical AI
The final design fiction film Sonder 2040 was screened to an audience of HCII faculty & students (n = 13) to determine if the film could meet the goal of encouraging optimism about the future of human-AI relationships.
After being presented with the research that led to the design and creation of the design fiction film, the audience was encouraged to discuss together the following questions:
What is the central message of this film depicting Sonder 2040? Do you agree or disagree with it?
“It’s interesting that the future is one where, in order to play in the corporate world, you have to pretend that you’re analytical and logical. But she’s emotional. And it’s interesting that you’ve used this really analytical and logical tool to help her achieve her emotional state with analytics and reason… it allows her to have those values.”
Design Fiction Screening Participant
How do these imagined social systems reflect your current anxieties or hopes for human-AI partnerships?
“If I could just find whatever node supports what I want the answer to be… what is truth, even? How would I know it when it happens? So I thought it was interesting that both messages came through for me—the hope and also the risk of this kind of system.”
Design Fiction Screening Participant
The viewers’ perspectives on technooptimism were measured before and after the screening, moving from a marked pessimistic view to one of optimism.
“I just want to say that I think the work that you’re doing is really, really, really important… I think the fact that you did the work to dredge up what was happening in past decades—how can we use lessons from that to inform what might happen in the current decade? …This is the crux of what we’re here for in human-computer interaction, so props for really taking that on.”
Design Fiction Screening Participant
Into the Future
The design fiction & collective discussion is clear: using emotions and multimedia storytelling encourages the exploration of possible futures regarding ethical AI.
By framing these media explorations as optimistic, we can begin to shift the current perception of human-AI relationships from one of fear and apprehension to one of augmentation and empowerment for everyday users.
SONDER (N)
the realization that each random
passerby is living a life as vivid and
complex as your own