Leonardo.ai is launching a new image upscaling tool as higher-resolution output for AI-generated images grows in demand, particularly in commercial spaces. Leonardo.ai was acquired by Canva in 2024 as part of the design platform’s broader push into generative AI, following a period of rapid growth for the Sydney-based startup. The company raised $47 million prior to the deal and built a large user base across gaming, marketing and creative industries. The integration has continued to evolve, with Canva moving to more closely align Leonardo’s technology with its own AI roadmap last month, while maintaining a smaller standalone product team. Leonardo’s Pro Upscaler is designed specifically for AI images rather than traditional photography, with output sizes of up to 105 megapixels. That brings AI visuals closer to the resolution typically required for large-format print and billboard advertising. Related Article Block Placeholder Article ID: 334886 Canva is coming for your SaaS stack Tegan Jones Higher resolution addresses one part of that equation, but it doesn’t necessarily resolve broader questions around image quality, consistency and how outputs hold up across different formats. Smarter business news. Straight to your inbox. For startup founders, small businesses and leaders. Build sharper instincts and better strategy by learning from Australia’s smartest business minds. Sign up for free. * indicates required Email Address * By continuing, you agree to our Terms & Conditions and Privacy Policy. Leonardo.ai co-founder and chief product officer, Jachin Bhasme, told SmartCompany that existing upscaling tools often struggle when applied to AI-generated images. “AI images have a completely different fingerprint from real photography,” Bhasme said. “Instead of natural textures, lens behaviour, and film grain, you’re dealing with synthetic noise, partially resolved detail, and artefacts. “When traditional upscalers hit those images, they misread them. You get over-sharpening, invented detail, and textures that break in unpredictable ways.” Bhasme said Leonardo’s approach was to better account for what he described as “AI-native artefacts”, rather than attempting to process images as if they were photographs. “We’re not trying to make AI images look like photos; we’re trying to respect the original intent,” Bhasme said.  “The most important part is that the creator is ultimately in control.” The launch comes as AI image tools move further into production workflows, particularly among agencies and marketing teams. In practice, Bhasme said some of the technical barriers have eased, but new challenges are emerging for teams using the tools at scale. Related Article Block Placeholder Article ID: 332306 Canva refutes job loss reports amid Leonardo.ai integration Simon Thomsen “One of the main challenges is consistency — keeping outputs on-brand across an entire campaign. Another challenge is curation. When you can generate thousands of options, applying judgment and taste to choose what’s going to work can become the bottleneck,” Bhasme said. While higher-resolution outputs open up additional use cases, Bhasme said the impact on traditional creative production is likely to be gradual. “It’s more of a shift than a replacement. There will always be a place for traditional photography, especially where authenticity, talent, or real-world context really matters,” Bhasme said. Instead, he said AI is increasingly being used earlier in the creative process to explore and iterate ideas, with traditional production used more selectively. On the commercial side, Bhasme framed the tool as part of a broader effort to keep creators within a single workflow, rather than moving between multiple platforms. “We saw users generating in Leonardo and then leaving to try to finish the job somewhere else, with little success,” he said.  “Our goal was to fix that problem and close the loop so people can go from generation to production without having to go somewhere else.” He also said that approach is partly a response to the pace of change across the AI tooling ecosystem. “There’s a huge amount of noise in this space right now,” Bhasme said.  “If we can deliver the best outcomes within a single workflow, that’s a better experience — but we don’t want to win on convenience alone, we want creators to stay because the output is genuinely better.” Bhasme also argued that when viewed across an entire production pipeline, the cost of high-resolution outputs is becoming less of a barrier. “If you can go from generation to production-ready output in one place, you remove a lot of time and tooling overhead,” he said.  “It becomes less about the cost of a single step and more about the total cost of getting something into market.” The move toward larger-scale use cases also brings unresolved questions around copyright, training data and brand risk into sharper focus, particularly as AI-generated content becomes more visible. “At Leonardo, we focus on giving creators high levels of control, quality, and flexibility so they can create confidently at any scale,” Bhasme said, adding that creators remain responsible for ensuring outputs meet legal and brand requirements. Stay in the know Never miss a story: sign up to SmartCompany’s free daily newsletter and find our best stories on LinkedIn.