Google Stitch: Has “Vibe Designing” changed the Future of UI/UX?

What if you could create a fully functional app purely from instruction? And we don’t just mean through written commands – but what if there was a possibility of creating an industry-standard app dashboard using voice prompts too? 

Google Labs has made these features possible with its very own AI-powered design platform – Google Stitch. Launched in 2025 at Google’s annual developer conference, Google I/O, this piece of technology is now available across 212 countries, making it a noteworthy tool not just for designers and developers, but also for those without much background or experience in either field. 

The idea is relatively simple. You tell Stitch what you want to create – for instance, a mobile app. Then you tell it what colours it should incorporate, if there’s a theme or aesthetic you prefer, even a core feeling tied to a business objective. It considers this input just as any AI-tool would, and produces a functional prototype in an instant. Something Google calls “vibe designing”. It works as a means to bring first design ideas to life. 

Does this mean the way we approach UI/UX has now changed? This new software has caught the attention of designers, developers and tech enthusiasts alike across the $3.2 billion UI industry. Very soon, we might find ourselves using AI-driven user interfaces on our browsers or as apps installed on our phones. But has Stitch really surpassed all other UI technologies in the market? 

What is Google Stitch and how does it even work?

Stitch is a software that allows individuals to design and create industry-standard, high fidelity user interfaces for web and mobile applications. Stitch itself is powered by the Gemini 2.5 Pro/Flash from where it sources its data. Say a fashion brand like Uniqlo for example, wanted to create its very own mobile app for its customers (assuming its current one does not already exist!). They may prompt Stitch to create a mobile app UI for a minimalistic, Japanese-inspired fashion brand. 

There are three ways they could input this information. One way is through text prompts by typing instructions straight into Stitch just as you would with other AI chatbots such as Gemini. Second, they can add voice inputs which is a Stitch feature of its own. For example, a developer may follow up with a voice prompt such as:

“Stick to warmer tones”

or

“Add a separate Memberships page with a loyalty card attached.”

By speaking to the software itself, users can take control of their creative flow with almost immediate results as they continue to prompt and refine their ideas further. The third type of input includes imagery. Users may reference existing apps or upload screenshots from which Stitch’s AI technology can use to recognise patterns and replicate them too. As a result of each prompt, Stitch can generate up to five different screens simultaneously in one project – all interconnected and tied together with a mapped out user journey and interactive app flow. 

That’s not all Stitch does. Its Agent Manager feature allows you to work on multiple ideas at once while monitoring project progress from start to finish. There’s also existing templates to work off of making it highly accessible. Its ‘Stitch and Play’ feature allows you to visualize functioning prototypes in real-time by ‘stitching’ different app screens together to form a logical flow or user experience. 

However, at the height of Stitch’s buzz is its newly introduced file type – Design.md. 

What is Design.md?

One of the more unique components featured in Stitch is probably its new markdown file type – Design.md. Think of this as an instruction manual that contains everything you need to know about what an app should look like. It is an AI-readable document that stores information from visual mood, colour palettes and typography to styling, spacing and layout.

This can also be exported or imported for use in other softwares to replicate a specific style or feel. It acts almost like a brand style guide and can be implemented across various other Stitch projects to ensure brand consistency. The file in itself includes both design properties and the code to reflect this, but presents itself in a format that is both agent- friendly (as in it can be read by machines such as AI) and human-friendly too. 

The more interesting aspect of this new feature is that rather than developers traditionally having to write code based on a non-tangible idea or concept, a Design.md file generates both design elements alongside its respective, production-level code all at once. It does this in the form of React code accompanied by Tailwind CSS (a styling tool for websites) and HTML to create interactive user interfaces. This helps with better accuracy and alignment with the project idea itself, rather than working on design and code as separate tasks. 

Who is the competition?

There are a wide range of design tools on the market, with dominating softwares such as Adobe, Canva, v0 by Vercel and more notably – Figma. Since the introduction of Stitch, there has been concern as to whether it could compete with existing UI/UX softwares in the field. In fact, MindStudio predicts that the gap between Stitch and Figma is likely to align over the next two years. 

However, this does not necessarily mean making Figma, among other UI design tools, totally obsolete. 

Compared to Stitch, Figma has been the favoured design tool since around 2018, according to MindStudio. It ultimately changed the design landscape by becoming the first professional tool that could operate on a browser while also allowing for real-time, collaborative design. It incorporates the use of variants, properties and nested overrides that allows for flexible, customizable features including clickable buttons, varying text and more precise editing such as within existing components like cards on a page.

It also offers a developer-specific view (DevMode) alongside an online collaborative design space called FigJam. Figma also offers thousands of community plugins for its users, with access to icons, animations, etc. Overall, it is a flexible, professional-grade tool that has now also incorporated its own AI features including text-to-design generation and building prototype flows for visualization using AI.

Design Tool	$ per month
Google Stitch	Free
Figma (Professional Plan)	$15
Figma (Organizational Plan)	$75
Adobe CC	$54.99
v0 by Vercel (Premium)	$30
v0 by Vercel (Business)	$100
Lovable (Pro)	$25
Lovable (Business)	$50
Banani (Pro)	$30

Is Stitch really the better tool?

Like any software, Stitch has its pros and cons. Although, it has notably caused a stir in the current UI/UX landscape, particularly after Figma shares fell more than 4% since its announcement according to TechInsider. Despite this, the reality in which Stitch will replace other established developer tools completely is highly unlikely, as there are still a few implications to consider. 

The Pros of Stitch

The strongest benefit of Stitch at present may just be its price factor. It is currently free to use, and offers up to 350 generations per month plus 200 generations with their premium service using Gemini 2.5 Pro. NxCode predicts that Stitch may evolve into creating paid plans that will undercut Figma’s by approximately 30-50%. In comparison to its competitors who charge between $15-100 a month, Stitch still holds the price advantage. 

In terms of performance, Stitch is great for efficiency. It offers instant designs and prototypes which are inevitably quicker as compared to being developed from scratch. By prompting it with a final idea or outcome, Stitch works backwards to produce an end-product. It does this by analyzing prompts and generating production-ready code simultaneously, which ensures better alignment and consistency.

This is much different to traditional design tools that usually require visual assets to be translated into code first. An article by Medium points out that this results in a higher chance of producing better UIs that actually fit the brief. Stitch is also highly compatible with existing, industry-grade tools. It has comprehensive code export capabilities allowing it to generate functional React code and Tailwind-native code (for Tailwind CSS) with the option of being integrated with software such as Flutter, Google Cloud Run or App Engine.

More importantly, Stitch has made app creation all the more accessible. You wouldn’t necessarily need an extensive background as a developer or a designer to produce a web app, per say. It becomes easier to approach just as any other generative AI tool or AI chatbot. Once you feed your ideas into the system, it produces a tangible, workable result. 

Although, limitations still exist. 

Why Stitch isn’t exactly the perfect substitute

Creating digital products such as web or mobile apps require large amounts of testing, human ideation and a level of professionalism too. Stitch appears to be lacking in these areas, with concerns around its design capabilities limited to more generic, repeated outputs mainly due to how AI processes its information, that is, through pattern recognition. You may observe similar e-commerce layouts or login dashboards across its generative results as an example. Moreover, how you prompt Stitch matters greatly too. It has been observed that Stitch as a tool can lack depth and precision, as it may struggle to handle more complex projects where detailed designs or original concepts are required.

A strong understanding of software development and design concepts is still strongly recommended, including basic coding and programming, that will allow for quick-fixes in an app’s code for example, rather than having to prompt an AI-tool to amend specific parts of a project. In this case, users may take more time to prompt Stitch as a result of its inability to make precise changes or understand the depth behind original ideas, which is a longer process as compared to going in and fixing the code themselves. 

This is much unlike Figma, which offers advanced prototyping features and utilizes component and design systems to build its UIs on a much bigger scale. Another concern lies in potential copyright infringement, whereby design data produced by Stitch may have implications in terms of its originality. For organizations, this is a significant consideration to make as to avoid any legal risks that may arise from the use of AI-generated content in their digital products. In actual accessibility for the product user, UIs produced by Stitch do not necessarily comply with accessibility standards or guidelines, such as how developers may consider colour contrasts, navigation and more that comply with legal accessibility obligations. 

Most importantly, Stitch is relatively new. It has only stepped foot into the market in the past year. As time progresses, we cannot be sure of the product’s continuity as well as its integration into existing softwares and business processes, especially without Service Level Agreement compliance. 

What does this mean for Designers and Developers?

Regardless if Stitch is here to stay, we should remain adaptive and open to new technologies even if that means we aren’t onboarding them completely. AI-generated UI has become a growing trend, and therefore UI/UX designers are strongly encouraged to stay up-to-date with new industry tools and technologies. Consider experimenting with Google Stitch software as a means to upskill or even simplify your own work processes. This has been proven to make a difference as reported by Software Engineer, Sanjay Nelagadde, who attempted to do just this. He shares on Medium that kickstarting the development process with Stitch can improve workflow by making “the path from idea to implementation much shorter.”

Stitch could be a useful tool to help with creating a tangible visualization of an app, for instance, before you even start creating it. Sanjay reports that bridging this gap can help to omit the initial vagueness one might experience at the very start of a new project, and Stitch could potentially help map out a direction forward with a clearer starting point. From Sanjay’s experience, he estimates that Stitch can help to propel the first 20% of the UI process at the very least.

Likewise, consider testing Stitch’s code export quality, how it performs, how accessible it is and how well it adheres to professional standards to identify if this can realistically be implemented in your own workflows. Organisations may start to test if Stitch offers a strong integration into existing systems, whether it can accurately generate content that fits your brand, and how it could help fast-track projects through its multi-screen prototype generation feature. Understand how Stitch outcomes can be refined or complemented with other industry tools, such as Codex, Cursor, Claude Code, React and more. 

TechInsider predicts that at least 50% of UI prototypes will be AI generated by the end of 2027, and that we may see new roles such as AI Design Director emerge in product teams. This points to a clear rise in AI-adoption in the UI scene, making it almost necessary to think strategically. 

How we can use Stitch

While Stitch offers some unique features from voice prompts to its Design.md file format, at present it seems unlikely that it would surpass existing design tools purely due to its lack of capacity, ethical implications as well as its debatable quality. Rather, Stitch comes as a useful tool to complement existing workflows in the UI/UX space. Or, it acts as a new avenue for creativity more suited for non-designers or individual use. It’s great as a quick tool to generate prototypes fast for visualization rather than for implementation.

It also makes a good test for an AI-powered UI design tool at no cost. However, on a larger scale, Stitch still falls short. It is not particularly suitable for more professional projects that require high levels of detail as well as collaboration from various stakeholders. Google Labs may introduce further upgrades to the technology within the year, but until then, it may be worth giving Stitch a try. 

Interested to learn more about UI/UX and Digital Design? Our course introduces you to industry-grade design tools such as Figma, Adobe Photoshop and After Effects alongside core UI/UX principles that allow you to explore web and mobile design to build accessible, user-friendly digital products.

Click  here to get started.