Artificial intelligence in design is no longer a future fantasy—it’s a present reality. We already see real use cases where AI meaningfully extends a designer’s capabilities. It acts as a force multiplier, enabling individuals and teams to generate ideas, iterate faster, and explore directions that might not have otherwise surfaced.

One of the clearest early wins is in concept generation. Tools powered by generative AI can quickly produce mood boards and product concept renderings. It’s like having a team of digital interns ready to create visual alternatives immediately. This doesn’t just speed up workflows—it expands the scope of creative exploration, which leads directly to higher quality solutions down the road. And on the textual side, AI can now simulate something as fundamental as brainstorming. When I started in design in the '90s, it took eight people in a room for an hour to generate a wide set of ideas. Now, an AI collaborator can do that in seconds.

That’s where AI excels in the design process: during the divergent phase, where the goal is quantity and variety. Accuracy at this stage isn’t critical. After all, even human-generated ideas in a brainstorm are often messy, incomplete, or unviable. It’s in theconvergence phase—where concepts are refined and validated—that human judgment becomes essential. AI can suggest, but it’s up to us to select.

For me, the most prescient issue in business today is this: companies must stop treating AI as a set of experiments and start embedding it into the very fabric of their operations. My work focuses on helping leadership teams transition from pilots to measurable outcomes, tied to revenue, product impact, and customer experience.

I’ve lived through successive waves of transformation—from 2G through 5G and into the Internet of Things. Each wave reshaped how we connect, but none moved with the velocity or breadth of AI. Unlike earlier technologies that disrupted one sector at a time, AI is reshaping nearly every function simultaneously—design, marketing, finance, operations, even legal.

Still, no transformation succeeds without trust. In AI, trust hinges on transparency—how data is treated, how system decisions are explained, how governance is embedded. Trust is not abstract; it’s a design requirement. Confidence indicators, provenance of recommendations, and visible governance structures aren’t optional. They are the mechanisms by which adoption takes root. As regulations fragment across states and nations, mature organizations will distinguish themselves not only by compliance but by weaving governance into workflows from the outset.

Artificial intelligence is not new. It’s newly accessible.

With the recent surge of interest in AI, the excitement feels familiar to us in industrial and manufacturing engineering. Years ago, data science went through the same hype cycle. Today, AI—particularly generative AI—is riding a similar wave. The algorithms have existed for decades. What’s changed is access. Generative tools like ChatGPT have made powerful computing capabilities available to anyone with an internet connection. Accessibility is transformative, but it also creates anxiety, uncertainty, and a significant trust gap.

As an academic and educator, I’m now tasked with helping students navigate that gap. Since I’ll coordinate Capstone Design projects next year, I plan to strongly encourage students to use AI tools like Enzzo to support new product design. But this isn’t just about tool adoption but responsible integration. It’s about learning how to evaluate, question, and apply AI critically.

My background straddles two worlds. I trained in mechanical design and production engineering but transitioned to industrial engineering, specializing in logistics, supply chain systems, and operations research. I’ve lived the shift from hardware to systems thinking and watched AI evolve from abstract academic theory to daily practice.

In every era of product development, there are a few breakthroughs that redefine the rules—moments when the tools change so dramatically that the winners aren’t the ones with the most resources, but the ones who adapt fastest. We’ve seen it with the internet, with 3D printing, and with the rise of global supply chains. Now, artificial intelligence is bringing another of those inflection points, collapsing timelines, stripping away inefficiencies, and rewriting what’s possible for teams of any size. The question is no longer if AI will shape your business—it’s whether you’ll be a leader who is ready when it does.

When I was 26, I pooled savings with my roommate and a few friends and took a gamble. We dove into trading, sourcing products from China, and even flying halfway around the world to meet factories that would produce our products in person. There was significant fragmentation in the product creation process; this was a barrier we were determined to overcome. Having been born in Taiwan and being conversational in Mandarin, I still found working with manufacturers difficult—a realization that if it was hard for me, it must be even harder for entrepreneurs with no language skills at all. Though critical, there were other challenges.

At the intersection of atoms and bits lies a growing frontier—and one venture studio is laser-focused on it. They aim to apply AI, sensory technologies, robotics, and advanced metrology to the physical world, tackling challenges across manufacturing, energy, human performance, medical devices, and industrial systems. They’re not just incubating ideas—they're helping startups scale, bridging the treacherous “commercialization valley of death” that so often claims hardware-based ventures.

The reality is, most innovation ecosystems aren't built for hardware. While software has enjoyed decades of iteration and support, the physical side—design, engineering, manufacturing—remains high-risk and under-supported. This studio’s platform is about de-risking that process, empowering the next generation of entrepreneurs to navigate the most failure-prone part of building physical products.

AI is now reshaping that landscape. From CAD to DFM (Design for Manufacturability), hardware engineering tools have long stagnated. Legacy platforms like SolidWorks and Rhino dominate and are resistant to change in part due to entrenched workflows and highly structured supply chains. But that’s starting to shift. Cloud-based platforms, like Onshape, are early signals of what’s coming. 

And AI? It’s the accelerant.

When we launched Hardware Is the New Salt, we expected strong opinions about AI in product design and engineering. What we didn’t expect was the consistency—seasoned product leaders, innovators, founders, designers, and executives from different domains all describing the same shift. Not hype. Not fear. But curiosity and openness to how AI will change product development and the products we build. AI is not replacing human innovation—it’s accelerating our thinking, internal processes, tasks that previously took significant time, while democratizing who can participate, and changing what is possible in the end products.
The transition isn’t smooth. It’s full of contradictions. Yet the majority of the product innovators featured in Hardware is New Salt are embracing AI and its potential. And that’s what makes this moment exciting. Here are the key themes covered across the following pages:

The New Acceleration Paradox
Every contributor described the same tension: AI lets teams move faster than ever, at the exact moment when moving fast has never been riskier. More people can create—but you still need people who can tell the difference between something that looks right and something that is right. 

More is Not the Goal—Better Products Are
A misconception outside this community is that innovators are using AI to make more things. The interviews reveal the opposite. The most sophisticated teams use AI to make fewer things—but better ones, and to do it faster.

When I think about AI and product design, I often go back to a simple, illustrative example: when we first put a camera on a mobile phone. At the time, something like that required a fifty-page document to capture the requirements. Today, I could ask an AI, "We’re adding a camera to a phone—what are the technical considerations?" It would instantly generate a decent draft. That part of the process can be automated now. But what about the idea of putting a camera on a phone in the first place? That spark—that’s human. That’s the boundary I see with current AI. Maybe someday AI can propose those ideas too, but right now, that leap still comes from us.

That’s how I frame the role of AI today. It’s an accelerator. It’s there to help us move faster, especially in the iteration stage. It can help test ideas, refine requirements, and explore options—but it’s still up to the human to steer the direction. Creativity, at least for now, remains a human trait.

But "faster" doesn’t mean pumping out more products. It means compressing the cycle from idea to market. There are a lot of iterations in that path—technical constraints, challenges to solve, decisions to make. AI can help with those cycles. It can surface best practices across industries and suggest relevant solutions. You don’t need a hundred people to do that digging anymore—you can get insights quickly and start moving forward.

That time savings is powerful. You can bring products to market sooner. But you can also reinvest that time into creativity, stacking more meaningful features before launch. Often, the real inflection point in adoption happens not with a single new feature, but when five or ten come together and create something richer. AI helps us get to that threshold faster.

When we talk about the rise of AI in product design, I like to think of it less as a checklist of right or wrong approaches and more as a framework. The transformations we’re witnessing today aren’t entirely new—disruptive technologies have been shaping the design landscape for decades. But of all the innovations I’ve encountered in my career, AI is easily the most disruptive.

Understanding where we are today, it helps to trace the evolution of product design over the past 30 years. While the roots go deeper, modern product design came of age with companies like Apple. Steve Jobs, Jony Ive, and others helped shift design from functional to emotional—from utilitarian to desirable—anchored in deep consumer insight. Technology continued to evolve, but the fundamental process remained focused on making things that people love and that solve real, often unmet, needs.

The most significant prior inflection point was the dot-com era, which ushered in a transition from purely physical products to interconnected physical-digital experiences. Even then, the role of designers remained focused on translating consumer desires into meaningful physical forms.

Today’s AI tools remind me of the early days of CAD—computer-aided drafting. When CAD was introduced, people worried it would replace drafters and engineers. In reality, it became a powerful augmentation tool. It allowed us to move away from drawing boards and toward spending more time on creative, high-value work. 3D printing followed a similar arc—cutting what used to take days or weeks into hours. AI fits the same trajectory: in the hands of experts, it accelerates discovery without replacing the human core of innovation. That’s the real shift I see today — AI isn’t changing the ‘why’ behind great design, but it’s transforming the ‘how’ in ways that ripple through every phase of development.

Decades of physical product development now intersect with cloud connectivity, AI, and advanced computing, marking a shift in the way we design hardware. In the past, devices like Sony Handycams or Casio organizers stood on their own, disconnected and limited in their scope. Today, a fusion of intelligence and context powered by IoT, powerful chipsets, and machine learning models are accelerating innovation at scale.

This moment is about speed, not just of computing, but of experimentation and iteration. AI allows us to compress what once took six months into six days. At Fluke, during the discovery phase, we’re using tools like Enzzo to quickly mine customer insights, build personas, model value propositions, and simulate bill-of-materials, at pace. That acceleration allows us to play faster, test ideas more rapidly, and refine more precisely.

When we move into the delivery phase, AI helps bridge the historically conflicting models of hardware’s waterfall approach and software’s agile processes. From prototyping to firmware to rendering, AI-fueled tools enable faster iteration, even across physical and digital boundaries. Then comes the sustain phase, where products are in-market and generating data. AI is changing how we ask questions about usage, performance, and customer behavior. No longer limited by dashboards or SQL skills, teams can now pose natural language queries and get answers in seconds. That’s a revolution in operational intelligence, and a lever for continuous improvement.

However, what makes this generation of technology different from past shifts is not just power or capability, it’s usability. With natural language interfaces, you no longer need to write code to be productive; you just need curiosity. That radically lowers the barrier to adoption, speeding up learning curves across entire organizations.

As a product development executive, leader in hardware innovation, and consultant for startups and scale-ups, and currently Vice President of Product Development at Polaroid, I spend my days thinking about how artificial intelligence is reshaping the way we design, build, and think about hardware.

The AI gold rush isn’t hypothetical. It’s here. And like many leaders in hardware and connected product design, I’m experiencing a tension: excitement over a transformative new tool and concern about its rapid integration into workflows, often without structure, oversight, or even visibility.

Across our global teams, engineers are quietly experimenting with AI in their daily tasks. They're not hiding it, they’re just doing what good engineers do: testing the tools at their disposal. But silent adoption creates challenges. When teams work independently, the organization loses the ability to learn together, missing chances to develop coherent strategies for integrating and governing AI tools effectively.

Because in the end, successful integration hinges on trust.

This isn't the first wave of technological change in hardware design. We've previously experienced major shifts with the introduction of sophisticated computer-aided engineering tools, especially advanced multiphysics simulations like thermal analysis, fluid dynamics, and structural simulations. Initially, these simulation tools weren't immediately trusted or widely adopted. Engineers approached them cautiously, validating outputs carefully, and learning the boundaries of their reliability while improving their accuracy.

AI differs not just in scale but in speed and accessibility. Previous tools were introduced methodically and top-down, carefully tested by R&D before wider implementation. AI, however, emerged bottom-up, rapidly adopted by anyone with internet access and a laptop. This democratization is powerful, but it comes with risks.

With every new technology platform or toolset, and with the general maturation of manufacturing alongside design and engineering, we’ve seen something subtle but powerful happen over the past few decades: deeper integration. Teams that once worked in parallel now converge. Those who thrive are those who can bring together supply chain thinking, manufacturing expertise, material innovation, and advanced engineering.

Consumer electronics brought out the best in this convergence, especially for electrical engineers. But I’d argue it was the mechanical designers who defined the interface, who built the shell between emerging technologies and the real world.

On the design side, we’ve always chased that moment when an innovation doesn’t just give us better tools—it gives us more freedom. More expressive range. More opportunity to shape something meaningful.

Whether it was early CAD, SolidWorks, or modern simulation platforms, the deal was always the same: learn the tools deeply and you’ll unlock their potential. Fluency was the price of freedom.

Now something else is happening. The tools are changing—and so is the nature of fluency.

What we’re seeing with AI is a different kind of liberty. You don’t necessarily need to master a tool in the traditional sense. Instead, you need to engage with it intelligently. If you can articulate your intent—if you can curate a process—AI can help you realize ideas that once required years of technical training. That’s a profound shift.

In a business built on expertise—deep, domain-specific knowledge of product development—the rise of AI challenges some of our most fundamental assumptions. If expertise is suddenly more widely accessible and automated, we have to ask: How do we, as experts, continue to differentiate ourselves? How do we evolve in a world where information and insights are no longer scarce but instantly retrievable?

That was the starting point for me. I began asking my team: How are we using these tools? What’s happening under the hood? It turned out quite a bit. Our software engineers—those writing code for embedded systems or user interfaces—were already tapping into AI to debug, generate test scripts, and validate their logic. “You mean it’s building code?” I marveled. And that was my moment. I had to jump in.

I'm a mechanical engineer, not a programmer. But I wondered if AI could enable me to instantly become a JavaScript expert. At least enough of one to build a checkers game. One May 2023 afternoon later, I had a working prototype and an epiphany on par with the first time I saw Netscape in 1995. Watching ChatGPT create functional scripts was a personal eye-opener. I saw, firsthand, how AI democratizes access to tools and knowledge. Practically, it’s no longer about whether you’ve studied a subject for years; it’s about whether you are curious enough to create the first prompt and follow through on the necessary iterations to achieve something new.

I don’t mean to imply that this last realization settled in instantly. Initially, my reaction to AI was to fear the new world order. But over time, that gave way to curiosity—and now, optimism. What we’ve gained is not just a faster way to code or draft, but a fundamentally new way to access, structure, and navigate knowledge. In our line of work, where entrepreneurs walk in the door with wildly diverse backgrounds—surgeons, chemists, software developers—AI lets me rapidly get up to speed. In just a few prompts, I can synthesize enough of their world to ask better questions and add more value. We hear about frontier labs working to “align” their models to expectations of etiquette, bias, and safety. A consulting firm has to re-align constantly, with customer A talking fluid dynamics and customer B diving deep into medical power systems. The faster we can get past the vocabulary to a place of shared understanding of priorities, the sooner we get to the meaningful human interactions. Used correctly, AI facilitates this, and that’s transformative.

There’s no question that AI democratizes design. Anyone with the right prompt can gain vocabulary, context, and ideas in fields they’ve never worked in. That can be extraordinary—or chaotic. The beauty is that AI can organize disordered inputs, establish structure, and frame problems in new ways. Where we used to spend the first two weeks of a product engagement just framing the challenge, AI can now do that in minutes.

I’ve spent my entire career designing hardware—not because I find it easy, but because I love it. There’s something deeply satisfying about solving physical problems, prototyping, testing on real bodies, and refining through hands-on experimentation. That process—filled with ergonomic tests, sketches, and mockups—is still, in my view, irreplaceable.

I’ve tried tools like MidJourney and Vizcom. They’re fun, even entertaining. But they don’t get to the heart of what makes great hardware design meaningful. When it comes to creating products that fit the human form, perform reliably, and stand the test of time, we need more than rendered imagination. We need craft, insight, and iteration. AI can’t replicate that—not yet, and maybe not ever.

It can help offload the tedious parts. I run a design studio, and part of that involves writing emails, marketing blurbs, and proposals. AI helps me get through those tasks faster so I can spend more time where I belong: sketching, prototyping, and solving the hard problems. That’s been the biggest benefit—not automation for its own sake, but creative rebalancing. AI gives me back time to focus on what I love most.

That mindset extends to my team. They’re younger, have more room to experiment, and they do play with AI tools. But while they’re fun and occasionally surprising, most of what’s generated still doesn’t meaningfully advance our design process. Our work requires precision, insight, and physical iteration—things AI doesn’t yet offer in a reliable or integrated way.

What excites us, though, is how AI is showing up in the hardware our clients are building. The real opportunity isn’t how AI helps us design—it’s how it’s enabling new tools in the world. We’re working on products that assist doctors with remote therapy, enhance diagnostic tools, and support patient care in new ways. That’s the kind of application that gets us out of bed in the morning.