Design, powered by AI

My theme at the beginning of this year is “Leading with curiosity with childlike wonder” and a great part of this is exploring and learning new things. In the spirit of New Year’s resolutions, this initially meant working out and eating right but I began to look deeper into what the future holds for the Designer and nothing screams future quite like AI.

I recently did a talk at Creativeverse centered around demystifying AI, looking into new tools and resources as well as easy ways to integrate AI into the design workflow. There’s so much to learn so I’m writing to invite you on this journey and let you in on what I’ve learned so far.

Understanding AI in Design

The landscape of design has made quite a few leaps in my lifetime alone. Here’s an outline of those changes:

  • Print Design Era (Before 1990s):
    • Focus on physical media (books, magazines, posters).
    • Manual and analog techniques (hand-drawing, typesetting).
  • Digital Design Era (1990s - 2010s): Emergence of digital tools (Photoshop, Illustrator).
    • Web design and user interface (UI) design.
    • Shift to digital media (websites, apps, digital marketing).

The digital era alone required a reframing of how we approached design. After we got over the initial fright of it potentially making the traditional designer obsolete, it became the new standard and we cannot imagine the world we now live in without it.

Though the AI-Enhanced Design Era began as early as the 2010s, with the introduction of AI tools and models like Midjourney and ChatGPT4, we are moving faster than ever in ways that will change how we think about design in the same way the Digital Design era has. AI introduces:

  • Automation of repetitive tasks and data-driven design insights.
  • Generative design, predictive analytics, and personalized experiences.

What is AI exactly?

There is a lot of anxiety swirling about when it comes to the implications of AI so naturally, the beginning of my journey consisted of educating myself at a high level on what AI and what it’s not.

The term Artificial Intelligence refers to the simulation of the ways humans think, speak, and perform tasks. There are many kinds of technologies that have different ways and approaches to doing any of this.

As mentioned earlier, AI isn’t exactly new and has existed for a long time but in weaker forms. Think of the last time you called customer service and the friendly automated voice asked, “Hey there, I can understand and speak in full sentences, what are you calling about today?” or when you typed in the chat on a website and it had a preset list of questions you could ask it. The concept of AI isn’t new as you likely have experienced but the resources we have now are and the pace by which AI is evolving is increasing rapidly more advanced by the day.

For example, Machine Learning: This is basically a bunch of code that enables computers to learn from data and improve its performance over time without being explicitly programmed to improve. For example, inside a subset of Machine Learning lies Natural Language Processing (NLP). NLP can take text and subsequently, summarize its language, translate it, and analyze its overall sentiment. Powered by an NLP, a primitive chatbot like the one mentioned earlier when combined with the recent advancements in generative AI, can now help solve more nuanced and complex problems than ever before. For instance, it could assist in deciphering legal documents for precise legal advice or in analyzing vast amounts of medical research data to recommend personalized treatment plans.

A computer that’s programmed to interpret and analyze visual information from images or videos might be the same system used for facial recognition or autonomous vehicles. You definitely want the machines behind such critical services to continue to improve at performing tasks without humans having to tell it to be better at its job.

The goal of AI is to perform in ways that make completing everyday tasks easier and more efficiently.

What AI is NOT

For one, AI is not sentient 😉

As mentioned before, it can only simulate human intelligence but it does not have a consciousness, emotions, or self-awareness. AI requires substantial human input for development like the data it learns from that only humans can produce. It still requires our ongoing maintenance, guidance, oversight, and intervention.

AI is not foolproof

Where humans would adapt, AI can only simulate doing so and because machines do not inherently possess infinite intelligence, it is still prone to error and unexpected results. If you’ve seen a hand in an AI portrait with 7 fingers on one hand, you already know what I mean.

AI is not bias-free

This one is really simple. Remembering that humans created the data that AI trains on, models can inherit bias present inside that information.

AI is not inherently creative

It can provide tools for inspiration but it doesn’t possess human intuition, judgment or the empathy required for true creativity.

The role of AI in Design right now

  • Think of AI as a powerful “creative collaborator” that can enhance and streamline the creative process.
  • Regardless of your educational background, socioeconomic status, or even skill level, AI democratizes and lowers the barrier to entry, allowing for more voices to sit at the creative table. Diverse perspectives can drive innovation by allowing for the discovery of solutions that serve more people.

Generating ideas with AI

Creative blocks are an unfortunate plight of the designer and sometimes we simply run out of ideas. Thankfully machines don’t get fatigued in the ways that we do. The benefit of using AI is that it can generate unique and unexpected concepts by connecting diverse ideas in ways you might not have thought of. Here are a few beginner tools to help generate ideas.

Google’s ImageFX

Many of us are familiar with Generative AI platforms like Midjourney and Dall-E that can take a prompt and generate an image. These AI generated images can be starting points for new creations. To add to the Generative Image toolset especially if you are new to this space, here’s Google’s ImageFX.

A screeshot of Google's ImageFX interface

Palette Generator by Canva

Generating a color palette is a great starting point for many designers. With this tool you can generate a random color palette from a demo image or you can upload your own.

A screenshot of Canva's landing page for its Color Palette Generator


This powerful tool allows you to generate web or mobile user interface ideas from a simple prompt.

A screenshot of Galileo's landing page

Streamlining design workflows with AI

How AI can assist in automating repetitive tasks

There are often tasks associated with the designer workflow that aren’t exactly creative. AI can take them off your plate giving you more space for the cool stuff. Imagine if you calculated how much time it took to prepare an image, align a document, or write captions for social media posts when sharing your portfolio. AI can help put that time back in your day for the introspection and inspiration that your work requires.

As a Product Designer, a large part of my work is gathering insights through user interviews or testing of designs. This often requires me to capture and read through lots of information to get a general idea of what design decisions are the most important to make based on the feedback I’ve gotten. With AI, I can now analyze the data gathered from feedback, allowing me to come to design conclusions faster than I ever could.

If you’ve ever extracted an individual with curly or wispy hair from a background of a photograph then you know how much time you can save by using a tool that does it both quickly and well. The tool isn’t limited for use on photos of humans but logos, products, and animals too! Even better that they have plugins for Figma, Photoshop, and more.

A screenshot of the landing page for


Buffer is already a leading social media management platform and they just made it simpler to do so by integrating AI into their solution and making content for posts and comments less time consuming.

A screenshot of the landing page for Buffer's AI Assistant

Challenges and ethical considerations

Relying on AI in the creative process has great benefits, many of which are highlighted above. However, anything not done in moderation can have its drawbacks. Here are a few things artists must keep in mind when adding AI to their process:

Remembering your unique voice

There’s so much beauty to be found in your unique perspective and lens by which you view the world. AI shouldn’t dominate the creative process so much that individuality gets lost.

Keeping curiosity at the forefront

Exploration, play, and curiosity are innate discovery tools we as humans use to learn and these instinctual characteristics are what gives space to innovation, critical thinking, and problem solving. Dependency on AI can reduce the amount of times we lean on these instincts ultimately leading to hindrances in an artist’s own skill development.

Authorship and ownership

How much of a work truly ours if AI makes up most of the composition or project? At what point should we disclose AI usage in the creation of our work? It is important that we keep our ears and eyes open for best practices around the use of AI from product to product and project to project.

Companies like TikTok have already begun asking creators to label AI generated content to reduce the spread of misinformation. Meta is exploring ways to detect and automatically apply labels to content posts on Facebook, Instagram, and Threads.

Place to learn more about AI tools

The future of AI in Design

What will drive innovation as it relates to AI will always be solving problems. AI can never replace human empathy, human intuition, and our sense of capturing the world as we see it. Here are ways I believe AI will continue to transform the design process as we know it:

  • Increased Collaboration: Many of the tools we use in AI are 1:1. Imagine exploring any of the tools mentioned above in a group setting where multiple people have access to modify a prompt together.
  • Deeper Understanding of our users, even faster.
  • Generative interfaces: Experiences that are completely unique to each user (accessibility on another level). Imagine knowing a doctor is researching anatomy vs. a middle schooler and how that could change the results of a prompt.

If you are interested in implementing Machine Learning or AI into your web application or software, let’s chat about how thoughtbot can help!