As the curtain parted on moooi’s 2022 exhibit, the audience was treated to a multi-sensory performance unlike any other: four synchronized robot scent diffusers telling a story through dance. As the robots—called Piro—swayed and dipped to the beats of a custom-designed soundscape, it was easy to forget the performers were made by a team of industrial designers and engineers in San Francisco.

The work expressed a future vision for what it means to humanize technology. Rather than design technology to make our lives more efficient, what if instead we designed technology to make life more beautiful?

For twenty years, moooi has seduced the world with bold, imaginative design. The firm, named after the Dutch word for beautiful, made its name by designing furniture and home goods that appeal to an interest in pushing the edges of form as much as function. Moooi came to IDEO looking to create an exhibit for the largest furniture fair in the world that might provoke the audience to imagine a world where design’s primary function was to inspire joy.

“Defy Gravity”: one of several experiential “worlds” moooi created for the 2022 Salone del Mobile. Each room had its own scent, soundscape, color, and digital experience.

Our process started by going wide. We presented 20 potential conceptual designs to the team, one of which moooi founder Marcel Wanders lovingly referred to as the “farting pet,” a robot scent diffuser that could walk around the home.

A scented smoke ring lingers in the air after Piro’s final crescendo.

From there, a team of IDEO designers enlisted the support of Catie Cuan, a choreographer who specializes in human-robot interaction, and Dutch sound design firm KLOAQ, to refine the concept into a four-minute performance piece.

Piro learned to mimic balletic human gestures.

The final result was technically involved: four robots moved independently, controlled by custom electronics and software designed to mimic the movements and grace of ballerinas. But in the performance itself, the complex mechanics disappeared, allowing the audience to experience not just what technology can do, but how it can make us feel.

‍

What happens when a designer and a classically-trained choreographer collaborate? Attendees of Salone del Mobile.Milano, the largest furniture fair in the world, got to see for themselves.

To build a climate-resilient future, we need an all-hands-on-deck approach across industries and organizations, as well as the leadership and collaboration of frontline communities themselves. 

To that end, IDEO and CSAA Insurance Group worked together to unite nine partner organizations, including Aon and the Environmental Defense Fund, to launch an open call for solutions to help communities prevent, prepare for, and recover from climate-related disasters. The goal of the Challenge was to surface actionable and scalable ideas that experiment with new technologies, models, and partnerships, then provide innovators with the support and funding they need to get their ideas off the ground and into the hands of frontline communities.

IDEO set these entrepreneurs up for success throughout the submission process, holding one-on-one conversations with participants, sharing constructive feedback on potential solutions, and developing their ideas and skill sets. The team also hosted workshop and panel sessions to grow community bonds and provide expertise on critical topics, including assessing climate risk and storytelling for impact. Hundreds of ideas came in from around the world, ranging from harnessing the power of AI, to building resilient infrastructures, to democratizing access to climate data, and more.

Once all of the ideas were in, a cohort of judges with a wide diversity of work and lived experience screened for concepts that were relevant, impactful, novel, community-informed, and equitable. At the end of the Challenge, the team chose 13 winning concepts with high potential for impact and partnerships, including Reef Rocket, which seeks to fabricate natural biocement reefs that reduce coastal flooding and erosion, and BurnBot, which is working to prevent wildfires by conducting controlled burns at scale.  

For these winning entrepreneurs, the end of the Challenge is just the beginning. Now, they are equipped with critical early funding, as well as a cohort of climate entrepreneurs, along with mentorship from IDEO, CSAA, and the other organizations. It’s a big step forward as they introduce new forms of equitable and transparent data, promote the voices of BIPOC communities, and bring to market new technologies that will protect and support frontline communities through this ongoing crisis. 

Read the Challenge Impact Report to learn more.

In the US alone, climate-related disasters cause $120 billion in damages annually. That figure is five times greater than 50 years ago, and projections are $2 trillion by the end of the century.

Nearly 90% of counties in the US declared a natural disaster over the last decade.

418

submissions

13

winners

$1M

in prizes

Every year, the NSF provides around 30 teams with up to $750,000 and a year’s worth of coaching, training, and collaboration to develop solutions for research areas like sustainable ocean use or combating fake news. Over the course of nine months, the teams learn how to uncover insights about human needs and design durable solutions for end users. But many of the scientists and experts in the program felt stretched by the idea of putting the needs of end users in the mix with their scientific testing—something few of them have been trained to do in their previous work.

To make that process feel more achievable, the teams built out prototypes and received guidance on refining and pitching their ideas along the way. IDEO joined the project as an innovation partner, facilitating hands-on workshops and creating “Science TV” episodes that could help the program reach a broader audience, deepening its impact now and in the future.

To help grantees get vulnerable, and evolve their ways of working, IDEO sent grantees to improv training  so that they could learn to build off each other. The team also took them to magic shows and on ghost tours in downtown San Francisco to learn the value of compelling storytelling in their work. The fun was part of the plan—gelling as a team and building relationships meant participants could make mistakes in front of one another. Being wrong doesn’t come naturally to academics, but it’s a key part of the design process, and a crucial part of solving complex problems .

Each year, a select group of teams move onto the next phase of the “coop-etition,” receiving up to $5 million in additional funding. And all of the participants leave with valuable new skills they can use to break out of their own silos and create human-centered, ground-breaking solutions for the rest of their careers.

When the NSF launched the Convergence Accelerator and made the radical move of bringing together teams from different disciplines, these academics, scientists, and industry partners didn’t have a shared language to move their ideas forward, and needed a structured approach to coming up with new solutions.

164

trained research teams

$200M

in federal support granted to AI-driven innovation, quantum technology, food security, sustainable materials, opportunities for people with disabilities, and more

IDEO and The Earthshot Prize built a nine-month Fellowship program and a five-day, in-person retreat to give Finalists the best shot at reaching the Prize’s ultimate goal: to scale the solutions that will repair our planet this decade.

To build out the new Fellowship, IDEO took a deep dive with the previous year’s Finalists to uncover where they needed critical support. This research revealed that the Finalists needed more ways to leverage the Earthshot platform, connect with potential partners to help scale their ideas, and build community with each other. The IDEO team then used those insights to create a set of recommendations The Earthshot Prize team could implement to create a more supportive and structured experience. Ideas they took forward included assigning a designated “Leap Advisor” who could provide guidance and assistance for each Finalist, as well as an advisor from the Prize’s Global Alliance Network who could partner on specific growth initiatives, like scaling product design, or policy change. 

Next up: the design of the retreat itself. Together, IDEO and the Earthshot team had three goals for the cohort’s week together in Windsor: focus on new forms of leadership and movement building, spur action through storytelling, and foster community.

The team developed the curriculum with leaders from Google X and Speed & Scale, alongside startups like UPSIDE Foods. They designed moments to build community, like paired walks and a chocolate-making workshop where Finalists could get to know each other away from the noise of meetings. 

Photographers captured the event to create assets for Finalists to use when pitching their ideas, and Finalists honed their storytelling skills. But what most remember is the magical moments of coming together as a cohort of next generation climate leaders who will help each other make an impact on the problems facing our planet, long after The Earthshot Prize has concluded.

The problems facing our planet are enormous, spanning industries, borders, currencies, and regulatory environments. Meaningful solutions will require leaders who can work across those lines.

30

climate solutions accelerated in first two years

Tucked away behind the brightly-colored restaurant fronts of East London’s Brick Lane is Second Home, a co-working space housing the design team of “vibe coding” company Lovable, one of the fastest-growing start-ups in history.

Founded in 2023 by Swedish software engineers Anton Osika and Fabian Hedin, and currently valued at $6.6 billion, Lovable’s mission is to make software development accessible to everyone, not just the 0.6 percent of the planet who are professional developers. The firm talks about their goal as “building the last piece of software.”

Nad Chishtie, Head of Design at Lovable, discovered his passion for building through his childhood obsessions with music, online gaming, and computers. Initially drawn to coding at the age of 8, he later “fell into design by accident” when he discovered he preferred problem-solving to programming. After stints at two hypergrowth companies, Cmune and Element, he had an epiphany. “With the launch of OpenAI’s GPT-3, I was able to do 70 percent of the coding work of seven full-time developers. I thought, ‘Everything's going to change.’” This realization led Chishtie to seek out the most ambitious people in this space, which ultimately led him to Lovable co-founder Anton Osika.

I spoke with Chishtie at Second Home about the skills people need to design with AI, Lovable’s unique culture, the pros and cons of collaborating with agents, and why humans—and optimism—will always matter. 

Ed White (EW): We’re increasingly hearing about a newer discipline called “design engineering” at companies like Lovable. What does that mean in practice?

Nad Chishtie (NC): It's essentially applying design thinking to code. Depending on your technical skills, design engineering could focus more on front-end development, microinteractions, or the architecture needed to support multiple initiatives. 

An interesting example of a design engineer is Niklas, who joined us as the team’s second designer and has a background in industrial design. Until last year, he hadn’t written code professionally. Now, he’s the third-largest contributor to the Lovable codebase. That’s because today there are two paths: Path A involves ideation, creating comps, and managing stakeholders, while Path B is about directly translating ideas into code with 100 percent fidelity. Path B wins a lot. It’s why Niklas’ learning curve was just a couple of months.

It’s difficult to find people with this skill set, but a single person with this shape can be as valuable as an entire team. 

Path B also means that, in design, we can drive initiatives forward without needing technical support or advocating for inclusion in the engineering roadmap. We can just implement ideas directly. 

Initially, some of the engineering leaders questioned why the design department was hiring engineers. But after we became productive, their response changed. Now, it's one of the most requested roles within our company.

EW: What mindsets or skills are required to be a successful design engineer?

NC: You need to be a very strong systems thinker. Instead of focusing on a specific solution, you’re developing the system that enables that solution to exist.

In classic lean startup methodology, the focus has been on specializing both people and products. With AI, you need to do the inverse and adopt a generalist mindset—for two reasons. First, it’s more challenging to constrain AI models than to let them operate freely. Second, by allowing models to evolve and get better without constraints, your product improves, too. By thinking like a generalist, your product expands its footprint in tandem with AI’s growth. To design at the frontier of AI, you need to invert your perspective.

‍EW: What’s different about designing with LLMs?

NC: When you design software as a service (SaaS), the main task is to design constraints that make the system more deterministic. With AI, the goal is to develop something that is somewhat deterministic and also truly generative. Because if it isn’t generative, then it’s not truly intelligent; it just mimics patterns or picks templates in the background. This marks a shift in design from focusing on hard constraints to providing a broader design direction, while maintaining certain guardrails. 

EW: How is working with these models changing design?

NC: From a technological perspective, we’re ruthlessly trying to reduce the time between having an idea and making it tangible. This applies to Lovable as a product and also to our internal practices. For instance, in our development pipeline, we’re trying to streamline processes as much as possible. If someone here has an idea and is curious about its viability, they can quickly tag an agent and receive immediate feedback. Today, it’s standard for a tech company to spend minutes, hours—even days—testing out a single part of their code base. Our goal is to reduce that time to zero. 

In terms of people’s time here at Lovable, there are no sunk costs with LLMs. We can adopt a more experimental mindset and go from a customer insight to a tangible iteration within a few minutes. Then, we can share that with more people, gather feedback, and decide whether to double down on the idea or delete it. It's totally acceptable to pivot because the turnaround cycle was minutes. 

EW: Lovable is one of the organizations that’s early to work with agents on a daily basis. What’s that experience like?

NC: We try to think of agents as if they were people. They need to have the proper context about how our team works, what’s important to us, the principles and philosophies behind our decisions, and what constitutes a good or bad decision. We have banks and banks of knowledge and skills that codify our thought processes for ourselves and for agents. They’re like a human employee onboarding Wiki, but for agents.

‍EW: What do agents do in practice, and what roles do they play in design?

NC: One role is called a “Linter.” You write some code, and then a Linter will dynamically check it against our codebase. A simple example is keyboard focus. Imagine you have an application with 20 different features that need to be added at various times. Keyboard focus can be quite invisible, making it challenging to maintain coherence. We have a skill bank dedicated to keyboard focus. If you’re designing or implementing a feature, a Linter might review your work and say, “With my end-to-end view of the application, this is how we think about keyboard focus. In the specific view you’re working in, you definitely should do it this way.” 

Another role is a “Sweeper.” Whenever you propose changes in a pull request, Sweepers conduct a variety of checks. For example, a Sweeper agent might say, “You’re introducing a new mental model for the user here. Here’s a list of related mental models that you should consider. Should we extend an existing mental model instead of adding a new one?”

The final role we’ve found very helpful is the “Critic.” These agents critique your implementations as thoroughly as possible. You can even ask them to imitate experts. For instance, you could prompt, “You are [Danish usability pioneer] Jacob Nielsen. I want you to critique this from a purely UX perspective.” The models are trained on Nielsen’s entire body of work, so you’ll gain insights that reflect Nielsen’s expertise.

EW: What’s been particularly effective for you when you use these agents?

NC: Prompting Critics to ask questions rather than make statements. General-purpose AI is overly sycophantic. Having Critics ask questions flips that dynamic. It keeps you in control, which is crucial since humans are ultimately responsible for the outcomes of these systems, and it encourages you to continuously critique your thinking. 

EW: As you work more often with powerful agents, what do you think the role of humans is?

NC: Human designers must be held accountable. We must create usable, desirable products that feel cohesive. I believe that across all mediums—be it film, TV, music, food, etc.—we have an innate ability to sense whether the people behind them genuinely care and what values guide their work. Do they value speed, a sense of luxury, quality, or something more experiential? I think the same applies to software. For Lovable, the experience of using our product is always the responsibility of the humans on our team; we don’t delegate that responsibility to AI. 

EW: How are team structures changing?

NC: I keep coming back to the word “asymmetry.” We used to build teams in very symmetrical ways at tech companies: you’d have a product manager, engineers, a designer, and possibly a data analyst. This cookie-cutter approach was easy for hiring and HR purposes, but it didn’t necessarily lead to the best products or user experiences. 

Today, team dynamics are evolving. Generalists can manage the entire design process—from user insights and analyzing data to building an initial product and testing it with users. A single generalist can deliver 20 times the value of someone focused on just one part of the process.

EW: What does that mean in terms of the people you look for?

NC: The “K-shaped” economy is an apt analogy for what we look for in design engineers. We’re seeing a split between those who are “AI native” and driven, and those who are more cautious about AI and hesitate. These two types of people are diverging significantly, with compounding effects in both directions. We hire people who are optimistic about AI’s potential and are eager to learn.

We also focus heavily on the slope of someone’s career trajectory—not just where they are today, but where they could be in this quickly evolving environment. During my introductory calls with designers, I spend time discussing their motivations for entering design, major influences from past roles, and self-initiated projects rather than just work history. 

Ultimately, a company’s success comes down to its teams and the products they build. The best team is the one with the highest learning aptitude.

EW: How do you create an environment or culture of experimenting and learning?

NC: One of our core values—“våga vara annorlunda”—means “dare to be different” in Swedish. We encourage our employees to challenge norms and embrace innovation by fostering an environment where failure is acceptable. We encourage rapid experimentation and urge individuals to continually challenge their own workflows. 

Many companies believe they need to adopt a single AI tool to write more code and then deploy it organization-wide. In contrast, at Lovable, we allow everyone to choose their own tools. It’s not only fun, but as the tools evolve and our understanding of how to use them expands, our organization adapts and grows as well. 

We have a few helpful rituals, too. Every Friday, we host open demos to showcase what we’ve been doing during the week. Someone might share, “I figured out this new approach that required overhauling our development environment. If we use agents in this new way, we could be 20 times more productive.” We celebrate these achievements as part of our onboarding process, in our Monday morning meetings, and during our internal company awards. 

EW: What’s important to remember when designing with LLMs?

NC: First, the models themselves are incredibly idiosyncratic; it's almost like having different children, each with their own strengths and weaknesses. To produce truly great outcomes, you need to understand these idiosyncrasies and build around them.

Second, you need to consider the entire system. You can't evaluate models, core product architecture, and user experience in isolation. You have to understand them all and integrate them to create a cohesive user experience.

EW: Any parting thoughts for designers?

NC: Be optimistic. Every door is open right now. I feel like a kid again, discovering technology for the first time. 

‍

“Design in the Age of AI” is a series of conversations with designers and makers from across industries and disciplines, building the future with AI, today.

“To design at the frontier of AI, you need to invert your perspective."

Every CEO we talk to is focused on the same thing: using AI to become more efficient.

We call this the “AI Dividend.” The surplus of human bandwidth, creative energy, and organizational capacity that automation liberates. And right now, most leaders don’t have a strategy to reinvest it.

The organizations that win in the coming decade will not be the ones that automated fastest. They will be the ones that invested the resulting AI Dividend most wisely, redirecting it toward the unmeasured frontier where human creativity, judgment, and sensibility still reign. The AI Dividend is not a bonus. It is the seed capital for a fundamentally different kind of organization.

The efficiency trap

The instinct to automate is understandable. AI can now handle tasks that once consumed enormous human effort: synthesizing reports, managing logistics, coordinating schedules, and writing first drafts of code. Early adopters are seeing real gains. At Anthropic, they estimate that 90 percent of the code written to build Claude Code will soon be written by Claude Code itself. These numbers are not anomalies. They are signals.

But here’s the trap: If every competitor achieves the same efficiencies—and they will—then efficiency alone produces no lasting advantage. When the marginal cost of execution approaches zero, what differentiates one offering from another? Not speed. Not cost. The real advantage comes from pairing the efficiency gains with investment in an innovation capacity. 

The race to the bottom, where every product and service converges on the same AI-optimized median, isn’t a hypothetical. It’s already underway. Browse any social media feed, and you can see the first wave of it: AI-generated images, text, and video that looked novel six months ago, now blur into indistinguishable sameness. The term for this is "AI slop," and it will quickly become a problem for even the most well-intentioned companies. 

Leaders who focus only on efficiency will find themselves competing on price in a market where price advantages evaporate almost overnight. The smarter move is to think of efficiency not as the destination, but as the mechanism that creates room for something far more valuable.

The architecture problem

To understand why this moment matters, it helps to look at an earlier technological revolution.

When factories first electrified in the late 19th century, most owners simply swapped out the steam engine for an electric motor. They kept the same layout, the same belt-and-shaft system that transmitted power from a single central source to every machine on the floor. The factory was still designed around the constraint of steam. Everything had to be arranged near the power source, and the entire line ran at the same speed.

It took nearly 30 years for manufacturers to realize that the electric motor had completely changed the game. With small individual motors, you could put power at the point of use. You could rearrange the factory around the flow of production, not the flow of energy. The resistance to change was immense. Plant managers had spent careers optimizing the belt-and-shaft system. But once the unit-drive factory emerged, productivity gains dwarfed anything the old architecture could deliver.

Today’s knowledge-work organizations operate as modern belt-and-shaft factories architected not around value creation, but around the movement of information through large-scale enterprises. The layers of middle management, the endless meetings, the reporting structures, the approval chains—these are coordination mechanisms that evolved to solve a very specific problem. When the primary constraint on execution was how quickly and accurately information could flow between people, bureaucracy was the best available technology. Alfred P. Sloan understood the power of autonomy when he redesigned General Motors in the 1920s. He introduced what he called “coordinated autonomy,” giving division leaders more freedom to make market decisions while centralizing finance and operations. It was brilliant for its era. And every modern organization descends from it, but they have grown and been forced to add ever more layers of hierarchy to deal with the need for coordination. 

This is where the AI Dividend becomes transformative. Sloan’s vision can now be fully realized, and the resources that were once consumed by coordination and the organizational overhead of being big can now be redirected. The dividend is not just a few hours freed up on individual calendars. It is a structural surplus: the entire cost of managing complexity in ways that no longer require human intermediation.

Where the dividend should go

So, where should leaders invest this AI Dividend?

The answer is not “more of the same, faster.” The AI Dividend should be invested in a new, more nimble and dynamic organization at the creative frontier: exploring the unmeasured territory where AI models cannot yet operate and where human sensibility, intuition, and judgment create genuine differentiation. We see two opportunities at the creative frontier: to make existing ideas better and to create fundamentally new ones.

This is not abstract or aspirational. There are historical precedents, and they are remarkably apt.

When the Industrial Revolution flooded markets with cheap, uniform goods, the Arts and Crafts movement emerged in response. William Morris, working in 1870s England, looked at the mass-produced kitsch pouring out of industrial production and saw an opportunity. He built a practice around the premise that quality, taste, and creative judgment could produce things that machines could not replicate and that people would pay a premium for. From Morris came the entire modern design movement and the idea that, by making ideas better, quality and design can create differentiated value in response to technological commoditization. History is rhyming. AI is producing its own version of industrial slop, and the commercial response will follow a similar pattern. The organizations that invest in human creativity, taste, and judgment as a counter to algorithmic sameness will be fit for the new rules of competition. 

And there is a deeper layer to what makes this frontier valuable. Less frequently, but more dramatically, there are times when operating at the creative frontier can lead to fundamentally new discoveries. Consider Jackson Pollock. His drip paintings, created in the late 1940s, expressed fractal patterns with uncanny mathematical precision, decades before Benoît Mandelbrot described fractals as a formal mathematical concept. In the 1990s, a physicist proved that Pollock’s canvases contain precise fractal structures at multiple scales of magnification. Pollock was not doing math. He was operating in the territory between tacit knowledge and the unknown, sensing patterns that had not yet been codified and expressing them through a medium that had no algorithmic equivalent.

This is what the creative frontier looks like. It is the zone where human intuition grasps something real before systematic knowledge catches up. It is where new categories, new markets, and new possibilities are born. And it is precisely the territory that AI, by definition, cannot explore alone. 

A new architecture for productivity

Investing in the creative frontier is not simply a matter of telling people to be more creative. It requires building the organizational architecture that enables creativity and learning to be productive at scale. This is what we call the “Adaptive Organization.” 

An Adaptive Organization integrates both efficiency and innovation, optimizing around change rather than stability or consistency. It operates through small, highly autonomous teams that are maximally interconnected without hierarchical constraints, enabling fast parallel execution. Unlike traditional organizations built for predictable environments, it treats adaptability itself as the core competency—measured not by efficiency alone, but by the speed and capacity to evolve.

The blueprint for the Adaptive Organization is still being defined, but three early principles stand out:

1. Build smaller, more autonomous teams

The most productive organizations we observe today are not large hierarchies pursuing one strategy at a time. They are networks of small groups that operate with high autonomy and minimal coordination overhead. AI handles the connective tissue that used to require middle management: information sharing, resource allocation, and progress tracking. What remains is a team small enough to trust one another, move quickly, and take creative risks. The evidence is already strong. ElevenLabs organizes its 400 employees into 20 “micro teams” of five to ten people. Amazon’s “two-pizza teams” follow the same logic. Rather than hiring more managers, Moderna has deployed thousands of custom AI agents to automate coordination and has trained all 2,400 employees to be data-driven decision-makers.

2. Increase speed by operating in parallel

Legacy organizations often pursue one big bet at a time because their coordination costs make parallel activity prohibitively expensive. When AI absorbs those costs, you can run thousands of experiments simultaneously and execute the most promising strategies far more efficiently. In biological terms, it is the difference between a species that produces one offspring and bets everything on its survival and one that sends a thousand seeds into the wind. The latter adapts faster because it learns faster by doing. Every experiment that fails is data. Every experiment that succeeds is a new capability that accelerates progress.

3. Develop the creative capacity to translate the unknown into a competitive difference

This means investing in people and practices that operate at the frontier: people with curiosity, creative confidence, a bias to action, and a willingness to challenge dogma. Their job is not to optimize the known, but to sense what is coming next. It means building an organization that values judgment and taste, not as luxuries, but as the core capabilities that distinguish a premium offering from a commodity.

The compounding effect

Here is what makes the AI Dividend argument urgent rather than merely interesting: the dividend compounds.

When you invest freed-up capacity in the Adaptive Organization, you generate new insights and new possibilities. Those insights, fed back through AI-enabled systems, create new efficiencies and new capabilities, which free up more capacity, which you can invest in more exploration. It’s a flywheel. The organizations that invest the dividend first will not just have a head start. They will have a compounding advantage that accelerates over time.

Think of it this way. A company that uses AI only for efficiency is like someone who loses weight but never exercises. They are thinner, but not fitter. The company that invests the dividend in creative capacity is like someone who loses weight, feels more energy, starts exercising, builds strength, and finds they can do things they never could before. The fitness compounds. One change enables the next.

This also means that waiting is costly. The gap between early investors and late adopters will not be linear. It will be exponential. First movers who invest the dividend wisely get access to AI-accelerated creative tools, which let them invest even more productively, creating more advantage. The window for catching up narrows with each cycle.

The gardener’s mindset

None of this can be commanded into existence. You cannot mandate creativity. You cannot engineer emergence. And this is where leadership itself must change.

The leader of a belt-and-shaft factory was, appropriately, an engineer. The leader of a dividend-investing, Adaptive Organization is more like a gardener. The gardener does not make the plants grow. The gardener creates the conditions in which growth happens: the right soil, the right light, the right spacing, the right pruning. The gardener sets boundaries and provides resources. And then the gardener gets out of the way.

This means designing organizational structures that enable rather than direct. It means tolerating ambiguity and partial answers. It means understanding that the creative frontier, by definition, cannot be mapped in advance. You are not building a machine. You are cultivating an ecosystem.

The instinct of most leaders under pressure is to tighten control, to demand predictability, to optimize harder. That instinct is perfectly suited to the belt-and-shaft factory. It is lethal in an environment where the competitive advantage comes from adaptation, from sensing and responding to changes faster than the world throws them at you.

The average lifespan of a Fortune 500 company has been declining for decades. Not because these companies got worse at what they do, but because the environment around them changed faster than they could adapt to. AI has the potential to reverse that trend, but only if leaders use it to increase their adaptive capacity, not just their operational efficiency.

The dividend is real. It is already accumulating.

The door to the new economy is open. Are you willing to step through it?

You can find more of Tim and Joe's explorations of the future at theoasis.press.

“The smarter move is to think of efficiency not as the destination, but as the mechanism that creates room for something far more valuable."

‍

If you ask consumers whether they trust their bank or health insurer, the answer is often lukewarm at best. Traditional financial institutions are seen as opaque, distant, and—in the worst cases—extractive. In a recent survey, barely a third of people thought their bank was being honest and transparent about costs and fees. But this narrative misses a subtler, more hopeful trend: The very technologies customers once feared as “creepy” are now opening a path toward rebuilding trust—not by hiding behind compliance and branding, but by providing real, personal value that people experience every day.

At the heart of that opportunity is AI. But not the generic, algorithm-as-black-box that so often gets headlines in the press. I’m talking about AI designed to help people, surfaced in ways that feel insightful rather than intrusive, and that elevates the relationship between individual and institution. It has the potential to rewrite our skepticism of digital products. 

The myth of the “creepy algorithm”

We’ve long assumed that consumers recoil at AI and the use of personal data. But emerging evidence and our own work at IDEO suggest a more nuanced truth: People don’t mind institutions using their data if the outcome is genuinely helpful.

What consumers resist is not personalization per se, but value extraction masquerading as personalization. They reject opaque upsells, hidden fees, and pricing strategies that leverage data solely to maximize profit. Anyone who has filed their taxes with the most popular online filing software can attest to the repeated (and difficult to decline) requests to use their tax data and share it with third parties, or to pay for add-on features of dubious value. But consumers welcome insights that feel like guidance—a personalized budget forecast that suggests paying down a mortgage faster, or a care recommendation that anticipates a chronic condition based on patterns in medical claims.

Financial institutions need to know their customers to help them meet their goals, not just the company’s bottom line.

When AI becomes a bridge, not a barrier

This distinction reframes the role of AI from a technical feature to a trust-building mechanism across two dimensions:

Predictive empathy: AI can surface insights that anticipate people’s needs before they articulate them. Imagine a health insurer that detects early patterns of risk in claims data and nudges members toward preventive care—with clear explanations and options—rather than waiting for a crisis.

Clarifying complexity: Financial and health decisions are inherently complicated. Consumers often feel alone with high-stakes choices—choosing a plan, managing costs, planning a life event. AI can simplify decision pathways, outline “what good looks like,” and benchmark personal choices against broader patterns of positive outcomes.

This is not theoretical. Customers are more open to institution-driven AI guidance than third-party general-purpose AI tools, precisely because of contextual relevance. While 79 percent of US consumers are uncomfortable with AI providing medical advice, a majority are open to AI supporting tasks like after-visit summaries and follow-ups.

In recent work we completed for a network of financial advisors, for example, we found that a customer’s trust in their financial advisor alone was not enough. They needed to believe that the plan presented to them met their needs and was likely to succeed, given the current market and their own financial situation. Unfortunately, too many financial institutions are seeing an opportunity to replace people with machines. When, in fact, there’s now an entirely new opportunity space of using generative AI to show, rather than tell, customers why they should have confidence in the path forward. People build trust. Digital tools can build confidence through what-if scenario planning, natural-language explanations, and the ability to have back-and-forth conversations whenever people want, without fear of asking “stupid” questions.

The unleveraged assets: contextual data and human intent

Banks service millions of accounts across life stages. Health insurers care for diverse populations across care journeys. These institutions are regulated to protect consumers, and in theory, succeed when their customers succeed. Yet traditionally, the relationship has been transactional: accounts, premiums, claims, and products.

AI changes that calculus. With careful design, institutions can translate the data they already hold into guidance that feels humane and useful. What might that look like?

• A bank dashboard that explains how spending patterns today impact long-term financial goals tomorrow in plain language, with actionable suggestions.

• A health insurance experience that models likely care pathways based on clinical evidence, population outcomes, and expected out-of-pocket costs, helping members choose care confidently.

• Alerts that not only flag risk—say, missed medication adherence—but also provide contextual support options, such as care coaches or financial assistance resources.

We don’t have to imagine far into the future to see what this looks like—we can actually see it in the not-too-distant past. When IDEO worked with PNC to develop their online banking platform, Virtual Wallet, one of the most compelling features for customers was a calendar that visualized “Danger Days”—days when their balances would be low, and they were most likely to overdraft. During a time when banks were making big profits by optimizing their technology to increase overdraft fees, PNC went in the opposite direction, helping customers avoid fees they despised and improve their financial lives. The result was an award-winning product that led to millions of new accounts.

This reframes data use from an internal optimization tool to a shared resource that benefits the member first. As banks bring generative and agentic AI to the masses, that could start to look like suggesting specific actions customers can take to reduce spending when cash flow is tight, or automatically splitting bills that fluctuate in cost, such as groceries and utilities.

Why incumbents have a strategic advantage

New fintech and health tech startups often don’t have to contend with years of technology debt and legacy systems, but they do suffer from a cold-start problem: They lack deep context. They may ask users to upload data, recreate history, or deduce preferences from limited interactions. Meanwhile, incumbent banks and insurers already have longitudinal data spanning life events, regulatory frameworks ensuring baseline protections, and direct experience managing risk at scale.

These are more than just advantages—they are assets waiting to be mobilized to build trust.

A new contract between people and financial institutions

Rebuilding trust in financial services and health insurance isn’t just a matter of better algorithms. It’s about redesigning relationships and using AI to make institutions advocates for individual success.

It’s easy to imagine continuing with the status quo: adding a chat interface to a mobile banking app that relies on existing functionality, for example, or offloading the most commonly-requested customer support tasks to an agent to reduce call center volume. 

That’s not to say that there isn’t merit in some of those approaches. But, in a world where AI can feel impersonal and inscrutable, there is an opportunity for these institutions to shine by making AI personal. That’s not only good design. It’s good business, and it’s good for society. The question isn’t whether AI can help. It’s whether we will use AI to help people—boldly, compassionately, and with trust at the center.

‍

Working to build trust with your customers? We’d love to help. Get in touch.

“People don’t mind institutions using their data if the outcome is genuinely helpful."

For decades, the typical relationship between humans and hardware was a straightforward one of command and control. We pushed buttons; machines performed tasks. But as AI migrates from our screens into our physical environments, the nature of that relationship is shifting: Computation is becoming a nondeterministic presence.

So far, the industry seems to be caught between what users actually want and the sci-fi fantasies currently captivating the tech community. Rather than marrying what’s technologically feasible with what people need, companies are bolting AI onto everything, shipping tech demos as finished products. The result is half-baked experiences traded for focus, and speed traded for agency and privacy. A better path means establishing a new standard in which trusted, useful intelligence serves the physical world.

For us, it means designing under these seven principles for good AI hardware, inspired by the timeless rigor of Dieter Rams, and setting the conditions to put them into practice. 

1. Good AI hardware is mutualistic

AI is not a deterministic layer of software trapped in plastic. Designed well, it is a partner in human flourishing. We build hardware rooted in mutualism: the device learns from the perpetually shifting nuances of human intention, and the human grows through what the device makes possible. More than optimization, it is a shared evolution between person and machine.

What might that look like?

The Household Reflection Mirror

A bathroom mirror is designed without predetermined health goals. It comes equipped with sensors, models, and processing power, but without a built-in definition of what “better” means for you. That definition emerges through use.

In the first few weeks, the mirror observes your behaviors without interpreting them. Over time, patterns surface that you implicitly confirm or contradict through your actions. You linger on certain readouts. You ignore others. Slowly, the device constructs a model of what matters to you, not what a wellness framework says you should care about.

After three months, the mirror notices your sleep quality degrades during a particular recurring time of the week. On Monday morning it surfaces a single line beneath your reflection: “Sleep is typically shorter on Mondays.” The next Sunday, you go to bed an hour earlier. The mirror notices. Over time it learns what kind of attention you actually respond to, because you taught it through the way you live and react. The device you have at the end of the year is not the one you started with—and neither are you.

Without your input, it is a capable system with no inherent sense of purpose. Without it, you miss a form of self-knowledge you didn’t know you needed. Closing that gap together is the essence of mutualism.

2. Good AI hardware is off by default

Truly human-centered hardware requires intentional consent to engage. Privacy isn’t an afterthought; it is a foundational design requirement. Until there is consent and utility, we reject the always-on model of technology. We design to respect the privacy of our homes and lives, so that technology enters our cognitive space or observes our physical space only when we consent. Good AI hardware is something you want around, that makes you feel safe.

What might that look like? 

The Consent Door

A front door knob has a small illuminated ring at its base. When you arrive home, the ring pulses once, dimly and unhurried. It is asking for your attention.

Twist the knob the way you normally would to unlock the door, and the ring goes dark. The house stays quiet. No sensors wake. No systems are activated. You are home, yet the home does not know it.

Twist the knob the opposite way before you enter, and the ring glows brightly. The AI features of the home come online. The thermostat starts learning. The mirror observes. The kitchen listens. You’ve made a conscious choice with your body before you crossed the threshold.

Guests see the same dim, illuminated ring when they arrive. They face the same choice. The house never assumes.

3. Good AI hardware is aesthetic

A piece of hardware should improve the aesthetic or emotional vibe of a space or interaction. With all the focus on the AI, it’s easy to lose sight of how intentional and sensorial the hardware has to be to provide value. We lean into creativity by using ergonomic forms, tactile textures, and haptics to create objects that feel like art or furniture while having capabilities suited to their use. The best AI hardware is crafted with intention for its role in our environments. 

What might that look like? 

Climate-Aware Window Glass

A sheet of AI-embedded  glass replaces a traditional living room window. The glass learns sunlight patterns, outside temperatures, and occupants’ daily rhythms. When the afternoon sun becomes harsh, the glass subtly softens and diffuses the light, giving the room a warm, painterly glow. On cold winter mornings, it lets in full sunlight to warm the space. There are no visible controls and no notifications. The window functions as an intelligent material, shaping light and heat in ways that feel natural, healthy, and beautiful.

4. Good AI hardware makes the invisible honest

AI hardware should communicate its state, data usage, and limitations through intuitive physical cues. Sensors now extend far beyond cameras and microphones into biosignals, radar, emotion recognition, and neural intent. Technology can understand us in unprecedented ways, which creates unprecedented room for distrust. It’s crucial that we replace vague terms and conditions and labyrinth privacy settings with transparent, real-time feedback, and make the invisible visible. Honesty is the only foundation for a lasting relationship with intelligent technology.

What might that look like? 

Visible Cognition Display

A small home security camera sits near the front door. Inside the door is a narrow strip that displays what the AI believes it sees in plain language. When someone approaches at night, it writes: “Face scan attempted. Low light. Result unreliable. Door stayed locked.”

The system reaches the edge of what it can confidently do and stops there, telling you exactly why—and that restraint is the point. A device that surfaces its own limitations in real time is making a fundamentally different promise than one that acts with false confidence and hides its reasoning. That promise, repeated across thousands of small moments, is how trust is genuinely built.

5. Good AI hardware is minimally intrusive

‍

The ultimate goal is to improve our lives in the physical world, not to keep us tethered to a digital one. We need to design for the disappearing act by ruthlessly stripping away unnecessary screens and notifications, and removing friction between humans and their environments. We need to design to follow user intent, using the superpowers of AI to understand user needs and respond in simple, clear ways. Whenever possible, we embed AI into existing rhythms and daily flows, rather than requiring people to pick up new habits. 

What might that look like?

The Context-Aware Notification Pebble

A small, smooth stone is connected to your devices. When something truly important comes in—a call from family, a hard-to-get dinner reservation, a flight deal on a trip you’ve been planning—the pebble gently glows. There are no sounds. No vibrations. No screens. Just light.

If you pick it up, the message is faintly projected onto the desk. For matters that require a decision, the system has already done the legwork. It held the reservation. It queued the flight deal. The pebble presents you with a choice, not a task. A single tap confirms your decision. The pebble stops glowing, and life continues as normal.

If you choose to ignore it, the pebble turns off on its own. The system respects your attention instead of competing for it.

6. Good AI hardware is honest about its capabilities 

We design hardware that delivers on what technology can actually do, while meeting real user needs. Good design speculates and continues to evolve as technology improves, but it does not hinge on speculative promises it can’t meet (and which inevitably let users down). We focus on the high-fidelity reality of what a device can do now, making the hardware a reliable anchor as technology improves rather than a vessel of far-reaching hype.

What might that look like? 

The Tutor Tablet for Children

A child asks the intelligent tablet a complicated science question. The device pauses and displays the message: “I am not certain about this answer…  Would you like me to look it up with you?” When the model is unsure, it offers sources or invites the child to explore the topic together. Sometimes it even says: “I might be wrong. Let’s test it…”

In this way, the child learns two important lessons: the scientific method and the understanding that intelligence, whether human or artificial, involves humility.

7. Good AI hardware respects the Earth

‍

We reject creating unnecessary devices that contribute to the global e-waste crisis by staying attuned to real user needs. When we build, we design for longevity through repairability, modularity, and circularity. From the selection of materials to the ease of recycling at the end of a product’s life, good AI hardware takes responsibility for its physical footprint. We build tools that are meant to last, not to be discarded when the next software update arrives.

What might that look like?

The Modular Home Intelligence Hub

A small wood and aluminum hub sits on a shelf, running local AI models for the home. Each component slides out like a drawer: Compute, Storage, Radio, Battery, and Sensor Array. While these modules are functionally co-dependent, each one is discrete. For instance, whenever new AI chips are released, the owner can simply replace only the Compute drawer.

The device also maintains a lifetime material dashboard:

Operational life: 8 years
Upgraded components: 2
Material saved vs. replaced: 4.3 kg
Energy used this month: 2.1 kWh

The hub processes most workloads locally to minimize unnecessary cloud energy consumption. Instead of being replaced every few years, it ages alongside the household.

Putting the principles into practice 

It’s one thing to create a list of principles. It’s another to make them part of our work. To uncover real human needs, we do research with people in their homes, learning where they want machines to help and where they want to stay in control. Leveraging emerging tools like 3D printing, AI-accelerated prototyping, and generative design, we move from insight to artifact quickly. Rather than just rendering a sensor, we build it. Rather than speculating about how friction feels, we test a physical interface in someone’s home and let their reaction tell us what to do next. We hold creativity accountable to what’s technically viable and commercially scalable, so the visions we make are ones businesses can actually act on.

We are moving past the era of experimental AI and into the era of specialized, high-performance hardware that permeates our everyday lives. The only way to succeed in the market is to earn a permanent place in people’s  lives through trust and utility. By anchoring what we build in these principles, we move beyond the hype and into producing hardware that people actually want around. That is the standard we are building toward.

‍

(Looking for a partner to build AI products with? Get in touch.)

‍

"A better path means establishing a new standard in which trusted, useful intelligence serves the physical world."

‍

I help organizations cut through complexity by shaping how they show up, through brand strategy, identity, and storytelling.

Working at the intersection of brand and business, I bring an editorial eye and a bias toward making—turning ideas into tangible experiences that people can understand, trust, and believe in.

My work spans government, healthcare, financial services, media, and consumer goods, from launching Gen Z-focused ventures to building innovation labs and reimagining legacy brands. I relish moments of ambiguity and enjoy translating across teams, perspectives, and priorities to move ideas forward.

I started my career in editorial and digital design, shaping cross-platform experiences and identity systems in publishing and agency environments. That foundation still shapes how I work today: I’m detail-oriented, collaborative, and overreliant on the Oxford comma. I hold a BFA in Visual Communication from the School of the Art Institute of Chicago.

For 25 years, I've asked the same question: Who does this design exclude—and what are we going to do about it?

My work spans human-centered strategy, inclusive design, and qualitative research, with clients ranging from Microsoft, Google, and Verizon to the National Science Foundation, and San Francisco Unified School District.

Before IDEO, I taught elementary school in East Palo Alto and spent a decade doing design work with social service organizations—where I learned firsthand what it costs people when systems are built without them in mind.

I am currently writing User Error, a nonfiction book about digital access and the design decisions behind it. I live in Oakland, California.

‍

My expertise spans brand, communication, and product design across tech, education, the arts, and social impact.

I believe great work demands both high-level vision and obsessive attention to detail, and only happens through collaboration.

I bring an inclusive, hands-on approach and a deep understanding of business, which enables me to consistently deliver excellence while always asking why.

Before joining IDEO, I worked at the brand firms Pentagram and VSA Partners. I began my career as a designer, leading teams at the School of the Art Institute of Chicago and the Milwaukee Art Museum. I hold an MFA in graphic design from Maryland Institute College of Art and a BFA in communication design from the Milwaukee Institute of Art & Design, and have taught at some of the best design schools in the US.

I am responsible for IDEO’s long-term success in China and working with clients to use design as a tool to enable growth. Specifically, I have helped Chinese companies design holistic brand solutions through the development of their products, communication, services, and innovation teams, and I have helped multinationals expand their presence and influence in China.

I advise global leaders on developing China-led innovation and capabilities.

In over 15 years in IDEO Shanghai, I have worked on projects that use design to elevate the quality of the experience of healthcare products and services, streamline processes that increase productivity, create spaces and programs that promote and enable inclusive communities, and build next generation mobility solutions that are planet-positive.

Before joining IDEO, I worked at Philips Electronics and the Electrolux Group in Italy, the Netherlands, and Singapore on a number of breakthrough commercial products. I am a member of the Young President Organization in Shanghai.