“
“We're building a device that has a vibe that is not like walking through Times Square, but like sitting in the most beautiful cabin by a lake and just enjoying the peace and calm.”
Stop the slop, design AI products that make life better
Welcome to 2026. AI continues to dominate the conversation.
Last year, a staggering 1.5 trillion dollars were invested globally in AI, fueling a surge of products and services across nearly every category. If you pay attention not just to markets, but also to consumers, you will see there is a growing dissonance between the supply and demand sides of the AI equation.
People are not confident that AI actually makes their lives better. Less than half (46%) of people globally actually trust the AI systems they are using. 50% of Americans express more concern than excitement, specifically citing a loss of “personal control” and the degradation of human creative thinking.
AI is a transformative technology, no doubt, but it is not a magic bullet. Some products are genuinely groundbreaking. Some are poorly conceived and executed. The difference doesn’t come down to model or compute. It comes down to whether AI can offer a way of life that people see as genuinely better.
In 2026, as the AI bubble starts to deflate, we are seeing the emergence of four domains where AI can potentially provide lasting value. In these spaces, technology moves beyond novelty to solve profound human needs, offering ways of life that are not just more efficient, but meaningfully better.
Personal Devices: Technology that gets out of your way
The human challenge:
Digital fatigue from high screen usage
Since the invention of screens, we’ve been forced to interact with technology on its terms, adapting our posture, thought-processes, and social behaviors to “the machine”. As a result, profound “digital fatigue” has set in. Record-high screen usage is now directly linked to “technostress,” chronic neck pain, and a state of “hyperconnectivity” that paradoxically increases social isolation and loneliness.
The opportunity:
Products that enable you to be more present
People desire technology that recedes into the background, allowing us to reclaim our physical health, attention, and social presence. AI offers a path to “zero UI” environments where technology finally adapts to the human, rather than the other way around. This new interaction paradigm allows people to interact using active inputs (voice, gestures, touch) and passive inputs (routines, biometrics, usage patterns), so they can be more present in their environment but still have access to the benefits of AI’s assistance.
What we’re seeing:
Early-movers are investing big in edge and ambient computing, paving the way for broader adoption:
- The automotive sector has moved aggressively toward screenless displays to combat driver distraction and enhance safety, paving the way for other consumer goods.
- Healthcare has the highest adoption rate of ambient intelligence, projected to hold a 24% market share by the end of 2025.
- The smart home market is expected to reach $162.78 billion by late 2025, with a core focus on “invisible automation.”
Design recommendation:
Trade digital noise for quiet assistance, using AI to orchestrate seamless support so people can remain present in the moment.
Wellbeing: Technology that helps you see the whole picture
The human challenge:
Not just living longer, but better
People are worried not just about their health, but also their health care. Medical care across the globe is getting more expensive and harder to access. Meanwhile the gap between total life expectancy and healthy life years is widening, people are living longer but not necessarily better. This unique combination of factors has put new emphasis on longevity and prevention.
The opportunity:
Better informed life-long health decisions
People want to feel more empowered to take their wellbeing into their own hands. AI is transforming the direct-to-consumer wellness space, moving from trailing indicators (what happened) to leading indicators (what is about to happen). By identifying sub-clinical changes before symptoms appear, this technology empowers users to be more proactive and self-reliant.
What we’re seeing:
Innovations incubated by research labs are starting to show up in direct-to-consumer products:
- Predictive biometrics forecast illness, health events, or burnout.
- Vocal biomarkers screen for early depression or cognitive decline.
- Ambient diagnostics using optical imaging to estimate long-term health risks.
- Personalized health simulations (aka Digital Twins) help develop long term treatment plans
Design recommendation:
Shift from tracking what happened to guiding what comes next, connecting data and behaviors with long-term outcomes in a way that feels personal, comprehensible, and actionable.
“
“I think we all in this room are focused on what is effectively some kind of preventative care... our view is that we can drive preventative care, and we can do that through a device.”
Education: Technology that learns how you do
The human challenge:
Inability to think for ourselves
Critical thinking is a crucial skill that is often underdeveloped in educational settings (for a variety of reasons). At the same time, there is a rising societal anxiety that over-reliance on AI will lead to “mental atrophy” and the loss of independent judgment, making us even less capable of thinking for ourselves.
The opportunity:
New ways to support every learner
Recent research shows that AI can can help or hinder learning. When AI is used to summarize content before a student reads it, critical analysis scores drop by 22%. However, when used to debate a topic after reading, critical analysis scores rise by 18%. When designed correctly, AI can reinforce critical thinking and help educational concepts stick by making them relevant, engaging, and personalized. When designed poorly, AI can subvert learning and create over-reliance.
What we’re seeing:
Experts in learning science are demonstrating what AI designed for deep-learning looks like:
- Restricted from giving answers, Khanmigo uses “scaffolding” to break down complex problems and prompt students for the next logical step.
- Duolingo Max provides personalized explanations for responses, including why the answer was correct or incorrect, with grammar and vocabulary insights.
- Microsoft Reading Coach generates custom practice stories on the fly that target a student’s specific phonetic struggles and personal interests.
- Squirrel AI monitors a student’s “cognitive state” and pivots the curriculum in real-time to prevent frustration or boredom
Design recommendation:
Align AI‘s behavior to the moment in the learning journey. Avoid pre-emptive summarization and answer-giving; instead, deploy AI to challenge assumptions, explain errors, make content engaging, and amplify curiosity.
“
“It's easy to put a little wrapper around these models and call it a tutor, but they're essentially like, 'We'll do your homework for you.' Because of its Socratic design, though, Khanmigo will instead ask the student what they think is the first step to solving the problem and then go from there.”
Work: Technology that gives you time back
The human challenge:
Cleaning-up after multi-purpose AI tools
Workers understand they are expected to adopt a “growth mindset” around AI, often with the implicit message that refusal equals obsolescence. But the reality on the ground is messy. Many AI tools are overly generic, producing outputs that feel shallow, incorrect, or misaligned with real business constraints. Instead of saving time, employees often spend hours prompting, correcting, and validating AI-generated work. This creates a new kind of cognitive load.
The opportunity:
Specialized personalized AI partners
Instead of general-purpose AI tools that attempt to do everything reasonably well, workers desire domain-specific systems that do a few things exceptionally well. AI with deep knowledge of their particular industry, function, or workflow can meaningfully reduce friction by understanding the constraints, language, and decision logic that professionals actually use. Instead of creating invisible labor, AI can create space for workers to focus on judgment, creativity, and strategic decision-making, the parts of work that are distinctly human and increasingly valuable.
What we’re seeing
Early signals suggest a shift away from horizontal AI tools toward deeply embedded, role-specific systems:
- Vertical AI startups, focusing on domains such as healthcare, automotive, or retail, captured ~35% of new AI funding rounds in 2025.
- AI embedded directly into enterprise software workflows are showing 150%+ year-over-year growth.
- Gartner predicts 40% of enterprise applications will be integrated with task-specific AI agents by the end of 2026, up from less than 5% in 2025
Design recommendation:
Design AI tools that are deeply opinionated about the work they support. Prioritize specialization over breadth and integration over novelty. Use AI to quietly absorb repetitive, rules-based tasks so humans can focus on judgment, creativity, and leadership.
“
“AI will not destroy jobs — it will destroy tasks. That’s a big difference.”