Generative AI creates detailed images from text prompts. You may be interested in using it within your own work . . . but are its outputs high-quality enough?
Shutterstock recently launched its very own generative AI tool. Now users can create original and unique images by typing a description of their desired image into the tool's text field.
This new technology is changing creative thinking and processes. After all, we know that 45% of Shutterstock customers are already using generative AI. As this technology continues to boom, numbers of users are expected to grow, as well.
Are you curious about trying generative AI for the first time? Or are you a seasoned user, looking to incorporate it more into your work? Wherever you fall on that spectrum, you may find yourself asking several questions that boil down to one: Is AI-generated content high-quality enough?
While generative technology isn't always perfect (you've surely heard of funky AI-generated hands), its quality is continuously improving. You can already incorporate it into your own creative processes. Use it to inspire new ideas or craft an image that's perfect for your next project.
Here, we'll detail the quality of Shutterstock's own generative AI tool and the imagery it produces to help you create with confidence.
Despite its triumphs, you're probably aware of some of AI's shortcomings. Faces sometimes look distorted. Hands may be in odd shapes or have tangled fingers. This is because the technology is young and evolving!
While you might occasionally find strange images being generated, remember that generative AI is learning all the time. It is continuously getting better. Look at the funky hands as a growing pain that only crops up every once in a while. It will soon be a thing of the past.
The good news is that you can easily edit your text prompts and generate images again for better outputs. Chances are, better-looking results are just a few keystrokes and clicks away.
Images generated by Shutterstock's AI tool are of great quality. Comics, illustrations, paintings, and AI-generated 3D assets are among some of its most excellent outputs.
Photography, though, has a little ways to go in order to look realistic. When it comes to what you want to generate, we see that scenery, abstracts, and still life images result in the best outputs.
It's easy to arrive at such great quality images. With other tools, you need to spend a long time crafting very specific text prompts to get their models to generate images of high quality.
Shutterstock's intelligent style picker does this work for you—with little text input. Users can focus on specifying the contents of their desired image, and our tool will make sure their results are in the right style and of high quality. Ultimately, these excellent outputs make it easier to explore more creative ideas.
So, are these images good enough to use right now?
It depends on your project. If you're looking for creative inspiration, yes, generative AI is an excellent tool to embrace. You can also use its image outputs in a variety of smaller-scale graphic design projects like social posts or flyers.
Another way Shutterstock is creating a quality generative AI experience for customers is by making our tool as accessible as possible. We want customers to understand the technology up front and start generating images right away! Therefore, we've incorporated the easy-to-use design approach that informs all Shutterstock products.
In order to create brand new imagery, simply visit Shutterstock's AI image generator page. Type a text description of what you want to see. In just a few seconds, you'll receive four unique images in a wide range of styles.
If you'd like to get more specific with your image outputs, simply edit your text or choose styles to apply with the style picker. A new batch of high-quality images will materialize in seconds.
By making the Generate tool easy-to-use, we're hoping to deliver the best possible image results to our customers.
Pursuing an Ethical Approach to Generative AI
In order to bring generative AI to the masses, Shutterstock has partnered with OpenAI, the creators of the popular DALL-E 2. This collaboration introduces OpenAI's seamless image generation capabilities to Shutterstock audiences worldwide.
We're the first to support a responsible AI-generation model that pays artists for their contributions, making us a trusted partner for generating and licensing the visuals you need to uplevel your brand. Also, we have thoughtfully built in mitigations against the biases that may be inherent in some of our datasets, and we are continuing to explore ways to fairly depict underrepresented groups.
"Our tools are built on an ethical approach and on a library of assets that represents the diverse world we live in, and we ensure that the artists whose works contributed to the development of these models are recognized and rewarded," says Paul Hennessy, Shutterstock CEO.
Contributors whose work was used to train these models will be fairly compensated through Shutterstock's Contributor Fund, and they will also be compensated whenever new generative content that uses their IP is created and licensed by customers.
All this is to say that, while the outputs of Shutterstock's generative tool are of stellar quality, there are many things happening in the background to further uphold this quality as well. By developing a generative AI tool that's easy to use and capable of producing excellent image outputs, we're ensuring that our customers can continue to create with confidence.