The Murky Middle: Representation, Translation, and the Real AI Opportunity in Product Creation

Everyone can see AI at the edges of product creation.

On the front end, generative tools are producing concept imagery at a speed and volume that would have been unthinkable three years ago. On the back end, virtual photoshoots and AI-generated marketing imagery are compressing timelines and budgets for e-commerce at scale. Those applications are visible, well understood, and — for the brands investing in them — already delivering measurable returns.

It’s the middle that’s murky.

The middle is where a sketch becomes a pattern. Where a pattern becomes a simulated garment. Where a 3D model needs to behave like the physical product it represents — materials that drape correctly, structures that hold up, constructions that can actually be manufactured. This is where the most consequential decisions in the product creation pipeline are made, where the technical demands are highest, and where AI’s potential is least understood. It’s also where I spent the latter part of my career at Nike, building the digital product creation infrastructure that sits between concept and commerce. The middle is where the real leverage is. And almost nobody is talking about it.

The Bookends Are Easy to See

The visible applications are attracting the visible investment. According to McKinsey and BoF’s State of Fashion 2026, over 48 percent of global fashion brands have integrated machine learning into collection planning and 3D sample generation. The AI fashion market was valued at $2.78 billion in 2025 and is growing at nearly 40 percent annually. Nike’s A.I.R. project used generative AI and 3D printing to co-create footwear designs with athletes, compressing concept-to-prototype timelines from weeks to hours. Adidas trained a diffusion model on its entire sneaker archive dating back to the 1950s to generate new concepts. On the other end, platforms like Vue.ai are producing on-model fashion imagery at a quarter of the cost and five times the speed of traditional photoshoots.

None of this is the hard part. Concept generation and marketing imagery are the lowest-risk applications because getting them wrong costs relatively little. A concept that doesn’t work gets discarded. A marketing image that misses the mark gets reshot. But in the middle of the pipeline, decisions carry downstream consequences that compound — a pattern that doesn’t work delays manufacturing, a simulation that doesn’t reflect material behavior leads to costly physical corrections, a 3D model that can’t translate to production engineering wastes weeks of work across multiple teams. The stakes are structurally different.

Representation vs. Simulation

For two years there was a race in fashion tech to go from prompt to 3D model. The results looked impressive on screen — visually compelling shapes you could turn in space, light from any angle, drop into a review meeting. But what those tools produced were solid mesh forms with no manufacturing utility. You couldn’t take them into pattern making. You couldn’t simulate fabric behavior on them. You couldn’t grade them, spec them, or hand them off to a factory. They existed in the space between concept and product, which is precisely where representation alone isn’t enough.

The race I’m describing isn’t over — it’s intensified. Open-source models like Microsoft’s TRELLIS.2 now produce watertight meshes with PBR materials in minutes, dramatically better than the demos of two years ago. For 3D printing and additive manufacturing, the outputs are usable. For pattern making, sample engineering, and production specs in soft goods, they still aren’t. Better meshes, same gap.

The story I’ve been telling for the last two years — AI is good at representation, AI can’t do simulation — is starting to get out of date. Not because the AI itself learned to simulate fabric physics. It didn’t. The simulation engines didn’t suddenly need it to. CLO3D, Style3D, Browzwear, Marvelous, Substance — those tools already handle the physics, and they’ve been handling it for years. The question was never whether the math was good enough. It was who or what was driving the app.

That’s the shift actually happening, and anyone working seriously inside a product creation pipeline right now can feel it. The agentic layer is moving in. Not as a replacement for the simulation engine, but as the operator of it. Sketch in, agent runs the pattern step, agent drives the simulation, agent hands the result to the next stage. None of the major platforms is going to publish a press release titled “agents are now operating our software end-to-end” — but if you’ve been in the rooms where the work is happening, you know roughly where every one of them is. Some of it’s public. Most of it isn’t. Style3D has been the most direct about positioning around an agentic layer. Adobe rebuilt its Substance 3D pipeline on OpenUSD and automation and pushed 55,000 renders in three months. The line between automation and agent is getting thinner by the week.

So the rep/sim binary is collapsing into something more useful. AI does representation natively. The simulation engines do simulation. The middle of the pipeline isn’t waiting on AI to learn fabric physics — it’s waiting on the orchestration layer that lets an agent drive the engines that already exist. That’s a translation problem.

The brands already operating this way are quietly pulling away. Tapestry trained Adobe Firefly on proprietary design elements they call “Coach codes” — brand-specific materials, hardware, construction details — so a digital twin stays on-brand from concept through marketing. That’s not a demo. That’s a production workflow where AI operates at the level of fidelity each stage demands. And what’s public is a small fraction of what’s actually happening inside these organizations. The real experimentation is below the waterline.

The Translation Layer

The most immediate AI opportunity in the middle isn’t full simulation. It’s translation.

Once you stop expecting AI to do the physics itself and start asking how an agent operates the engines that already do, the whole problem reframes. The question isn’t “can AI simulate a draping garment.” The question is “can AI take a sketch, prep the pattern, set up the simulation in the right tool, evaluate the result, and hand it off to the next stage.” That’s translation, end to end. And it’s exactly the kind of structured problem AI is already good at.

Every product creation pipeline has handoff points where work crosses from one system, one team, or one format to the next. Creative output needs to become engineering input. A sketch needs to become a pattern a designer can adjust in the software of their choice. A pattern needs to become a simulated garment without someone manually constructing every seam. In footwear, a sketch needs to become a procedural model — something editable and adjustable — and then transform into a real-time rendering that teams can evaluate together. These translation steps are where enormous amounts of time and fidelity get lost.

The numbers support this. Brands using digital twins and AI-assisted workflows report physical samples dropping from 15–25 per style to as few as 1–2, with pre-production costs falling 60 to 80 percent and development timelines compressing from 12–16 weeks to 2–4. Rendering time for a fully layered outfit dropped from five minutes in 2025 to under 90 seconds in 2026 using GPU-optimized AI pipelines. CLO3D reports 28–40 percent time savings in pattern-to-simulation workflows with AI-assisted features.

These gains come not from AI replacing the people in the middle, but from AI handling the translation between stages — converting formats, bridging systems, maintaining fidelity across handoffs — so that designers and technical developers can spend their time on the decisions that actually require human judgment. Translation is a tractable problem. Full simulation isn’t — not yet. The distinction matters because it changes what you build toward today.

The Human in the Loop

The industry has an underdeveloped understanding of how to use AI in stages alongside the people who provide oversight, quality control, and domain expertise. Most conversations about AI in product creation jump straight to capability — what AI can generate, how fast, at what resolution. Very few start with workflow design — at which step does AI contribute, at what level of fidelity, with what kind of human review at each gate, and who owns the decision when the AI output needs correction.

This is a design ops problem, and it’s the same one I argued in my last piece. The brands that built the operational foundation around 3D pulled ahead while others invested in tools they couldn’t actually deploy. AI in the middle of the pipeline is heading toward the same pattern. The talent bridge — the people fluent in both the creative process and the digital tools — is the connector between where AI is now and where it needs to go inside a working product creation pipeline. Eliminate that capability in a downturn and you don’t just lose what you built. You lose your best shot at what comes next.

AI doesn’t need to replace the people in the middle. It needs to work alongside them — in stages, with oversight, at increasing levels of fidelity. Build that workflow and the middle of the pipeline stops being murky. Leave it unbuilt and AI just becomes another tool that promised more than it delivered — not because it couldn’t, but because nobody designed the conditions for it to work.

Design Ops in Footwear and Fashion: What It Is, What Transfers, and Why Getting It Wrong Is Already Costing You

Most conversations about design ops start in the wrong place.

People reach for project management, resource scheduling, or process analysis and call it design operations. It’s none of those things. At its most precise, design ops is the discipline of designing the conditions under which creative and technical work happens — the workflows, calendars, gates, tools, team structures, and cultural norms that determine whether an organization can consistently convert creative ideas into market-ready products at speed and at scale. And embedded in all of that is something less talked about: a genuine understanding of creatives and the creative process. Done well, design ops isn’t just infrastructure — it’s an act of nurturing. You are creating the environment in which talent can do its best work, feel supported and challenged in equal measure, and have enough autonomy to pursue genuinely innovative solutions.

I held the role of Global Design Operations Director at Nike, spanning Footwear, Apparel, and Equipment. Before that I was a designer. After that I led digital product creation and 3D studio strategy. That arc — from making products, to running the operations behind making products, to transforming how products are made digitally — is why I care about getting the definition right. And the clearest signal that design ops is working is that it becomes invisible — it removes friction so completely that teams stop noticing it. When it fails, the opposite is true: everything feels harder than it should.

That distinction matters more than ever right now. The industry is at a point where significant investment has been made in design technology — 3D, digital product creation, AI — and the results are… uneven. For leading brands that built the operational foundation, AI tooling has moved from experimental to operational. For others, AI is arriving faster than they can absorb it — adding new capability, new complexity, and new pressure onto teams still finding their footing with the last wave of change. The conditions you create for your people matter more, not less, when the technology landscape is moving this fast.

The Transferability Trap

There is a version of this conversation where someone points to automotive or architecture as the benchmark — look how mature their design operations are, let’s import that here. And they’re not entirely wrong. There are first principles that transfer universally.

Decisions made early are cheaper than decisions made late. Digital tools need to be integrated into workflows, not bolted onto them. Asset governance and data infrastructure are foundational, not optional. Structured milestones protect creative exploration from premature engineering pressure. These truths hold whether you’re building a car, a building, or a boot.

But that’s where the direct transfer ends.

The three variables that separate industries operationally are timeline, volume, and complexity of variety. On all three, footwear and fashion sit at the opposite end of the spectrum from automotive and architecture. A car platform might run for seven years. A fashion season runs for months. Architecture manages one-off complexity inside long programs. Fashion manages high-mix, high-volume complexity inside a relentless annual calendar. The operational model that works in one context doesn’t survive contact with the other.

Gaming and VFX offer a more tempting comparison — both are digital-native and have built sophisticated pipelines for managing creative work at scale. But their final output is digital. There is no physical product, no factory, no materials lead time, no last-mile supply chain. The handoff that defines so much of fashion’s operational complexity simply doesn’t exist for them.

There’s a further layer of complexity that rarely gets named. When digital workflows enter footwear and fashion, they don’t replace the physical process — they run alongside it. Other industries are navigating this too — automotive and architecture are increasingly leveraging the same real-time visualization platforms that power gaming pipelines. But they’re doing it with longer timelines and far fewer SKUs. In footwear and fashion, the hybrid physical-digital pipeline has to operate at the same relentless speed and scale as everything else. That’s what makes the operational challenge structurally different — and why no existing model transfers cleanly.

Design ops in footwear and fashion has to be built for footwear and fashion — adapted to the industry, and then calibrated further to the size, scale, and specific business objectives of each organization.

What Good Actually Looks Like

The brands making the most meaningful progress share a consistent pattern. They embedded operational capability at team level rather than deploying it as an initiative from above. Where process becomes the way of working, it sticks. Where it remains a program owned centrally and visited periodically, it doesn’t.

Operational roles can be genuinely valuable in this — raising issues, managing risk, facilitating alignment, maintaining momentum. But they work best when they’re running alongside the people doing the work, not observing from a distance. That proximity matters. It allows ops to understand the work at a deeper level, to see the whole field of play and not just their own responsibilities — and from that vantage point, to plan better, predict problems earlier, and help teams maneuver when things shift. Ops doesn’t need to be the enforcer. But the metrics and signals they surface need leaders who will act on them. Without that, ops becomes a reporting function nobody responds to.

The other consistent differentiator is how technology sits relative to product teams. The brands that have placed technical capability close to — or inside — the product teams move faster and adapt better. Those keeping technology at arm’s length spend enormous energy on translation, and the translation always costs time and fidelity.

Leadership immersion matters too, and it’s underappreciated. Research consistently shows that nearly half of employees in digital transformation programs don’t feel adequately prepared — and that gap almost always traces back to leadership that endorsed the direction without genuinely understanding the work. The brands making real progress have leaders who are in it, not just behind it.

The Failure Modes That Repeat

Despite genuine progress across parts of the industry, the same breakdowns appear with regularity.

Cross-functional alignment is the most common. Design ops that functions well inside the design team frequently collapses the moment it hits the handoff to development, sourcing, or manufacturing. Part of the problem is structural — some functions have mature ops capability, others have none at all, and when the process moves between them it loses coherence. Ops functions that span organizational boundaries rather than sitting inside a single function can help here, providing the connective tissue that keeps the end-to-end pipeline readable and accountable.

Data governance is a failure that runs the full length of the pipeline. Design and BOM data created early in the process becomes disconnected from what follows. Designers struggle to find the reference data they need for their own work. And for the teams who depend on that data downstream, it can be a black hole — assets that exist but can’t be found, versioned, or reused at scale because the underlying infrastructure was never built. The work happens; the knowledge doesn’t accumulate. It’s the same failure at every stage, showing up differently depending on where you sit.

Change management is arguably the most honest gap in the industry — and one of the most under resourced. New tools and new processes get deployed without giving teams the time and genuine support to adopt them. This is not a people problem — it’s a planning problem. Dedicated change management roles are critical during any significant transition, and they tend to work best when partnered closely with ops — one focused on moving people through the change, the other on keeping the operational picture clear throughout it. The industry moves fast enough that carving out space for this is hard. But the brands that don’t make that space pay for it in slow adoption, low confidence, and capability that never fully lands.

The Scaling Back Moment

Over the past two years, a pattern has played out across the industry. Organizations — including significant ones that had been treated as lighthouses — scaled back their DPC strategies. In some cases, entire 3D design departments were reduced or eliminated. The common reading was that the technology didn’t work. That reading was wrong then, and the evidence since has confirmed it.

Most of these weren’t technology failures. They were design ops failures. The tools had matured significantly — the problem was that the operational infrastructure required to deploy them effectively was never built alongside them. No clear workflows. No data governance. No change management. No cross-functional alignment beyond the design team. When the investment was made in the tool but not in the conditions that make the tool valuable, the ROI was always going to disappoint. That’s a design ops problem, not a 3D problem.

The resolution of that period is now becoming clearer, and it tells the same story from a different angle. Many of the brands that “scaled back” were actually right-sizing — moving from deploying 3D everywhere to deploying it where it creates the most value. That’s organizational learning, not failure. The brands that are in genuine trouble are the ones that discarded the capability entirely rather than asking the harder question: did we actually build the operational foundation this required?

The talent question is where this gets most costly. The people who understand both the creative process and the digital tools are rare, developed over years, and not easily rebuilt. Disbanding those teams in a downturn means starting again from scratch in the upturn — at significant cost and with significant delay. And there’s a compounding risk that rarely gets named: these are the same people who would be driving AI adoption into the design process. The designers and operators fluent in digital tools are the natural bridge between where AI is now and where it needs to go inside a product creation workflow. Eliminate that capability and you don’t just lose what you built — you lose your best shot at what comes next. That’s a risk that doesn’t always make it into the cost-cutting analysis.

Where This Is Going

The discipline of design ops becomes more important the more complex the technology landscape gets — and for the brands treating it as a strategic capability rather than a support function, the picture looks different. The operationalization of digital creation in footwear and fashion is progressing, but unevenly. The brands that have built the operational foundation are pulling ahead. The gap between them and those still treating DPC as a project with an endpoint is widening.

AI is adding both opportunity and complexity to this picture simultaneously. More tools, more outputs, and — for anyone thinking clearly — significantly more governance challenges. The regulatory and IP landscape around AI-generated design has shifted from largely undefined to rapidly solidifying. USPTO revised its inventorship guidance for AI-assisted work in 2025. The Copyright Office published its analysis of AI and training data. Congress introduced the NO FAKES Act. The rules are no longer unclear — they’re tightening faster than most organizations anticipated. Brands that haven’t built data governance infrastructure will find AI compounds that weakness rather than solving it.

Design ops exists precisely to absorb that complexity so that creative and technical teams don’t have to carry it themselves. The brands that treat it as a strategic capability will be the ones that realize the full value of their digital investment. The ones that don’t will keep wondering why the tools aren’t delivering what they promised.

The tools were never the problem.