Design software companies are redesigning artificial intelligence

But as developers take advantage of new, generative AI-enabled tools that turn text into images and videos, software makers selling these products say the AI ​​itself needs some redesign.

“Artificial intelligence promises to do everything for you and magically read your mind, but the reality is that you still have to interact with it,” says Cameron Adams, co-founder and chief product officer of design platform Canva. “It’s an art and a skill to get something out of your head and explain it. And it’s not something many of us are equipped to do.

As generative AI tools like ChatGPT gained popularity in late 2022, companies like Canva, Adobe, and Figma are adding more AI tools to make creativity more productive. Designing more assets has become easier and cheaper – at a time when the industry is being asked to generate significantly more images, text and videos for different audience segments, geographies and platforms such as social and online.

But many frictions remain. Creative professionals are using artificial intelligence to help them generate all of these assets, but often these tools still integrate poorly with the documents or presentations in which these materials need to appear. Recently, design software vendors have focused on seamlessly integrating these tools into their product portfolios. “We don’t look at it as a single feature,” says Noah Levin, vice president of product design at Figma. “We see it more as a technology that can be applied across the product in different ways.”

Australian company Canva is adding AI-enabled features, including text generation tools powered by OpenAI algorithms; text-to-image tool; and AI background remover. Since the launch of its AI-powered Visual Suite, Canva says it has been focused on making the entire AI design process as consistent as possible.

“We’re seeing more and more people incorporating this into their work processes,” Adams says. An increased focus on AI offerings has helped Canva add 90 million new monthly active users, and the company’s AI products have been used 5 billion times to date.

Adams says that in 2023, the boom in generative AI has created a lot of buzz and niche tools for creative professionals, but many have struggled to integrate these new tools into their work. He notes that AI text windows are not particularly intuitive for users.

To make this easier, Canva has created tooltips in tools like Magic Media that help users work with AI to generate more precise creative assets. For example, a creator can enter a prompt like “man eating delicious pepperoni pizza” and choose a style direction offered by Canva, from dreamy to watercolor to anime, as well as aspect ratios including square, landscape, and portrait.

“The responsibility that we have as product developers is that we also have to make AI accessible,” Adams says. “We had to integrate it into our products and people’s workflows in a way that made sense.”

Democratizing the creative process is crucial for design-focused software companies, especially as the creative process has become more collaborative and concepts are discussed by a larger group of stakeholders rather than just designers who used to sit in the studio.

“Better concepts emerge when more people collaborate,” says Figma’s Levin.

Figma’s strategy is to both lower the level of design to give more customers the tools to participate in the creative process, and raise the ceiling with more sophisticated offerings for experts.

“We want to attract more people into the space who haven’t designed before and who I think can benefit a lot from learning visual communication,” says Levin. “But I also want the experts, the many millions of people whose needs we can meet, to feel that they can do their jobs better thanks to artificial intelligence.”

Adobe says that no matter how AI technology evolves, people need to stay up to date when designing creative assets. “I don’t believe these models are creative,” says Ely Greenfield, Adobe’s chief technology officer. “I think they’re production assistants.”

Of all the negative headlines generated by AI, few industries have faced as much criticism as image generators. Google had to temporarily disable Gemini’s image-making capabilities after it was discovered that the tool perpetuated racial and gender stereotypes. AI image generator Midjourney recently revealed that it is blocking users from creating fake political images ahead of the 2024 US presidential election.

“We put a lot of work into making sure we can’t generate unintended harm and bias in the images we create,” Greenfield says.

Adobe Firefly relies entirely on licensed content, which helps reduce bias that can creep into a company’s data set. While there have been complaints about content generated in Firefly, Adobe says it has put in place feedback mechanisms so that users can raise any concerns so that they can be resolved. Since Adobe launched Firefly a year ago, users have created more than 6.5 billion images using the generative AI tool.

As for why image generators are often at the center of ethical questions about artificial intelligence, Greenfield argues that when the data-driven aspect of work becomes automated, humans are not inclined to bemoan the repetitive tasks they have ceded to machines. However, because people value art, there is a much greater emotional risk in technology encroaching on the way art is created.

“I think something that will potentially disrupt artists’ ability to create art and make a living affects different parts of our brain and heart,” Greenfield says.

Sign up for the Fortune Next to Lead newsletter to receive weekly strategies for reaching the corner office. Register for free ahead of the launch on June 24, 2024.