AI Generative Image Potential: The Future of Visual Marketing

• With the help of GANs (Generative adversarial networks), AI can easily turn our imagination into images
• Combining AI generative images with other techniques helps provide an immersive experience

An increasing number of businesses apply AI to their marketing solutions, such as AI technology-based chatbots and operational procedures. Rapid technology advancement in this aspect has made more AI-based applications available on the shelves for marketers to pick in accordance to their operational needs. To understand and to embrace these available applications is the only way to take full advantage of its potential. AI Generative Images is one example.

Why explore AI Generative Images?

AI generative images do not only revolutionize the art business but also provide personalised experience to audiences. With the help of generative models, particularly GANs (Generative adversarial Networks), AI can easily turn our imagination into art and create beautiful artworks and new patterns based on simple text or image instructions, providing digital marketers an opportunity to create innovative and immersive marketing experience.

To create a personalised AI digital image, all you need to do is to imagine the scene you want to create, then input in text format to let AI to generate related images. With such easy implementation, finding and testing the right text is crucial. The following is my experiment on AI generative images from text to image.

In the first image, I use “metaverse, futurist, greg rutowski, artstation, dream, unreal engine, fairy tale, social media, universe” as text input. The text instruction I was using is my envision of the future metaverse.

After generating the above image, I am interested in how AI generates “Google in the metaverse”. Therefore, I amended the text input to “Google in the metaverse, nft search engine, social media, universe, thomas kinkate, greg rutkowski, futuristic, cyberpunk, trending on artstation”

To make the output image brighter and to carry a dreamy touch, I amended the text input by adding such keywords as “dream” and “fairy tale” to make the group of text to become “metaverse, futurist, thomas kinkate, greg rutowski, trending on artstation, dream, unreal engine, fairy tale, universe”. Below is what the engine generated.

This trial and error process can be repeated to fine tune the outcome. Although the process of finding the right text is tedious, we will have a better understanding of how AI works in the experiment of mixing human inputs and AI.

AI Generative Images Opportunities in Marketing

Once we understand how to generate images using simple text with AI, the next step, to combine AI generative images with other techniques and procedure automation, would be relatively simple. It can be used for videos, product launches, or exhibitions to build the story, give audiences an immersive experience and increase engagement.

For example, BMW used AI generative art for a fresh form of advertising campaigns. According to BMW, it leveraged “StyleGAN” artificial intelligence software to cross-reference over 50,000 images of artwork spanning 900 years of art history and a curated set of 50 works from renowned and emerging contemporary artists BMW has worked with over the past 50 years.

“For 50 years, BMW has supported the arts and culture through numerous initiatives as a way to engage and interact with consumers around the world in an authentic way. As we continue these efforts into 2021 and look for new and creative ways to engage audiences, we shift to a virtual setting where we are combining centuries-old art and the latest AI technology to create something completely new and exciting.” said Uwe Dreher, vice president of marketing, BMW of North America. BMW lies deep in its history and constitutes an important link between the past, present, and future by tapping AI, combining innovation and art, and creative technologies to engage with audiences.

Bulgari also created a massive art installation through AI. According to Bulgari, it was designed using algorithms to reproduce an image of nature generated by analyzing some 200 million pictures of nature and 160 million pictures of flowers, plus 120,000 images of snakes.

The future of marketing will increasingly need to be personalised, in real-time, with more technological applications, and be data-supported. Marketers cannot just focus on traditional marketing mix and data-supported marketing. Blending Art, design, science, and technology in your marketing strategies helps maximize marketing effectiveness, such as drawing awareness, to maintain a trendy and updated image while making practical use of the technologies.

Final Thoughts

As digital marketers, we constantly look for trends and patterns. Apart from AI generative images, some AI generative media, such as AI generative 3d animation, video, or music, are also worth exploring. However, we cannot just focus on the opportunities but other aspects such as ease of applications and implementation details. For example, I used a graphics processing unit (GPU) instead of CPU for AI generative images to shorten the rendering time. Therefore, brands also need to pay attention to technical specs and technology limitations.

Next Steps

Media is more complex and moves at a pace faster than ever. The days of marketing without involving the tech are over. Marketers now need to embrace new technology and bridge the gap between traditional marketing and technology. Technology never stops growing, and digital marketers never stop learning. Take as much information as we can to achieve our goal of delivering connected, sustainable and innovative marketing strategies.

For those who have no ideas on Python nor any programming language but would like to try the aforementioned experiment for inputting text to generate AI images. I recommend a free online tool Wombo.

About the author

Karen Lau is a Web3, Metaverse and NFT enthusiast. She has a passionate interest in applying the latest technology, such as augmented reality (AR), artificial intelligence (AI), NLP, and business intelligence (BI), to digital marketing for long-term growth. She successfully developed and evolved the company’s digital activities across online platforms.

Connect on Linkedln