Top 10 UX design trends to know about

The biggest forces shaping design right now are AI-assisted design tools and prompt-based prototyping, design systems maturing into governance layers, accessibility shifting from principle to legal requirement, and the rise of agentic experiences—products that act on a user’s behalf rather than waiting for clicks.

A few years ago, generative AI in design meant novelty image tools. Today it’s embedded in the everyday workflow: designers prompt working prototypes into existence, design systems constrain what AI can generate, and the hard questions have moved from “can we use AI?” to “how do we keep quality, trust, and craft intact while we do?”

If you work in UX, keeping a current view of these shifts helps you decide where to invest your time—and where to stay skeptical. Here’s our take on the ten trends most worth your attention, and what each one means for how you and your team work.

The top 10 UX design trends right now

These trends come from what’s actually visible in the industry—in the tools teams are adopting, the regulations now in force, and the way design work is being reorganized—rather than speculative predictions.

AI-assisted design and prompt-based prototyping

The clearest shift in day-to-day design work is prompt-based prototyping. Instead of starting from a blank canvas, designers describe a flow in plain language—or supply a reference image—and get a clickable starting point in minutes.

Early versions of these tools produced generic mockups that needed heavy rework. The current generation is more useful because it can draw on a team’s real component library, producing output that looks and behaves like the actual product.

That doesn’t make design judgment less important—it makes it more important. The broader role of is still being negotiated, but one thing is clear: when anyone can generate a plausible-looking screen, the differentiator is knowing which option actually serves . are spending less time on production work and more time framing problems, evaluating options, and editing with intent.

Design systems as governance, not just documentation

Design systems used to be reference material: a library of components and a style guide. They’re maturing into something closer to infrastructure—an enforceable source of truth that governs every interface, including AI-generated ones.

This shift is partly driven by AI itself. Generation tools are only as good as the constraints you give them, so teams with a well-structured design system get usable output while teams without one get generic noise. A mature design system is now the thing that makes AI-assisted design safe to adopt.

For design teams, that means design system work has moved up the priority list. Tokens, component APIs, and usage rules aren’t housekeeping anymore—they’re the foundation for how fast the whole organization can ship.

Agentic experiences

A growing share of digital products now include AI agents that complete tasks for users—rebooking a flight, triaging a support request, drafting and filing an expense report—rather than just answering questions.

Designing for this is genuinely different from designing screens. The user isn’t navigating; they’re delegating. That raises new questions: How does the user know what the agent is doing? How do they review and correct it? When should the agent stop and ask?

Trust is the working currency here. Users grant autonomy to systems they understand, so the craft is in transparency—visible reasoning, clear checkpoints, easy undo. Expect “designing the handoff between human and agent” to be a core UX skill, not a niche one.

Accessibility under regulation

Accessibility has been good practice for decades. What’s changed is enforcement. The European Accessibility Act (EAA) came into force in June 2025, requiring a wide range of products and services—including e-commerce, banking interfaces, and consumer apps—to be accessible to people with disabilities. It applies to companies selling into the EU whether or not they’re based there, and national regulators have been ramping up market surveillance and penalties since.

The practical effect: accessibility is moving from a backlog item to a release requirement. Teams are building accessibility checks into design reviews and CI pipelines, auditing against WCAG, and treating screen reader support, keyboard navigation, and contrast as defaults rather than enhancements.

The upside is that this enforces what UX professionals have argued all along—designing for diverse users produces better products for everyone. Teams that treated accessibility seriously before the deadline are now at an advantage.

Ethical design and data transparency

Closely connected to both AI adoption and regulation is the continued push for ethical design—respecting users’ privacy, being clear about how data is used, and avoiding manipulative patterns.

AI raises the stakes. Products that personalize, predict, or act on a user’s behalf depend on data, and users increasingly want to know what’s collected, what the model does with it, and how to opt out. Vague privacy language and dark patterns that were merely annoying a few years ago are now trust-breaking—and in some markets, legally risky.

The teams handling this well treat transparency as a design material: clear consent moments, explanations at the point of action, and honest defaults. It’s slower than the alternative, but it compounds into loyalty.

Personalization and adaptive interfaces

Personalization isn’t new, but it’s become more ambitious. Beyond recommending content, products are starting to adapt their interfaces—surfacing different actions, density, or guidance depending on who’s using them and what they’re trying to do.

Done well, this reduces cognitive load: the product meets before they have to dig. Done poorly, it’s disorienting—an interface that rearranges itself erodes the spatial memory users rely on.

The discipline that separates the two is research. Personalization decisions grounded in real tend to feel helpful; personalization driven purely by engagement metrics tends to feel invasive. Teams that pair adaptive design with continuous user feedback see without the creep factor.

Anticipatory design

Anticipatory design—reducing friction by predicting what the user needs next—has matured from concept to expectation. Smart defaults, pre-filled forms, proactive notifications at the right moment: these patterns now read as baseline competence rather than delight.

The technique still rests on the same foundation it always did: behavioral data and a clear model of user intent. What’s changed is that AI makes prediction cheaper and more granular, which makes restraint more important. Anticipating a need the user actually has feels like service; anticipating a need they don’t feels like surveillance.

A useful test: would the user thank you for this if they noticed it? If the honest answer is no, it probably shouldn’t ship.

Micro-interactions

Micro-interactions—the small animations and signals that confirm an action happened—remain one of the most durable tools in UX design. Loading indicators, swipe feedback, an animated checkmark after a successful purchase: they communicate state without adding text.

The four classic components still apply:

Triggers—the action that starts the micro-interaction, like a tap or a page event
Rules—the guidelines that make sure it fires at the right time, in the right way
Feedback—the confirmation itself, like a sound when an email sends or a checkmark on success
Loops and modes—how long it runs, tuned to the task it supports

What’s new is their role in AI-driven products. When a system is thinking, generating, or acting on your behalf, micro-interactions are how it communicates progress and builds confidence. As always, restraint matters—too many animations make a product feel busy rather than responsive.

Multimodal and spatial interfaces

Interaction is spreading beyond the screen. Voice, gesture, gaze, and spatial computing are converging into interfaces where users move between speaking, touching, and looking without switching modes.

The hype cycle around headsets has cooled, but the underlying shift is real: augmented reality is finding practical, applications in training, retail, and field work, and voice has become a serious input method now that conversational AI can actually hold context.

For most designers, the near-term implication isn’t learning 3D modeling. It’s designing experiences that hold up when the interface is partly invisible—where the “screen” might be a conversation, and consistency has to live in the system’s behavior rather than its layout.

Sustainable design

Eco-conscious design has settled into a quieter, more credible form. The conversation has moved past green branding toward measurable choices: efficient front-end code, lighter media, dark-on-OLED energy savings, and awareness of the carbon cost of heavy AI features.

That last one is the newer tension. AI-powered experiences carry a real compute footprint, and teams are starting to weigh whether a feature needs a large model or whether something smaller—or nothing—would serve the user just as well.

Sustainability won’t top most product roadmaps on its own. But as a tiebreaker between design approaches, and as a signal of craft, it’s increasingly part of how mature teams make decisions.

Stay on top of UX trends with Dovetail

The common thread through all ten trends is that the production side of design is getting cheaper and faster, while the judgment side—knowing what users actually need—is getting more valuable.

That judgment comes from . Whether you’re evaluating an AI-generated prototype, deciding how far to push personalization, or designing an agent’s handoff moments, the teams that win are the ones with a steady stream of real user evidence behind their choices.

Use this list as a jumping-off point, pick the one or two trends most relevant to your product, and ground your next move in what your users tell you—not just what the industry is talking about.