This week, Camera Intelligence revealed the latest generative AI features in its Caira camera: adding Nano Banana video editing to its genAI photography. It was the latest reminder that the imaging industry, like the rest of the world, is doubling down on AI in its products – including the generative kind.
No hate, no shade. But it reminded me of a conversation I had with Canon earlier this year, when I asked the Japanese leadership about the subject of AI – and they were crystal clear about what is and what is not off limits when it comes to artificial intelligence.
"AI is going to keep on playing an important role for any shooting experiences for our products," said Tetsushi Hibi, advisory director of Canon's Img Optical Business Unit 1, behind closed doors at the CP+ 2026 show in Japan.
"So this is Canon’s stance: we are going to use AI not to generate imagery, but to capture scenes as close to the reality of shooting as possible – noise reduction, color correction, distortion correction, aberration – so those are the functions we are aiming to put it onto in our products."
We've already seen this in practice, particularly with the Canon EOS R1 and R5 Mark II – both of which use Canon's neural network image processing (NnIP) for features like 400% in-camera upscaling and 2-stop noise reduction. And of course, the many algorithmic subject detection and tracking options.
It is heartening to know, however, that generative AI imaging is not on the table for Canon – particularly with the company having rolled out its Authenticity Imaging System in May, offering content compliance to help verify the authenticity of images (and, by proxy, policing image manipulation and AI embelishment).
Where Canon definitely is implementing AI, however, is in the design phase.
"We have started our attempt to use AI for design," notes Hibi, which is born out by a recent company report that AI will be used in the early prototyping phase to simulate things like drop tests and assess thermal performance.
"We have no choice but to use AI in the design department," adds Go Tokura, executive vice president and head of Canon's Imaging Group. He's also sure to point out that the fruits of the technology will still be distinctly 'Canon', rather than resulting in homogenized products that are the same as other brands.
"When it comes to design, it is said that the AI can bring up the best possible solution. And you may think, if that’s the case, every single company will come up with the same new products.
"We have our own accumulated technology – not just designing capabilities, but also production technology as well, and materials. So even if we use AI for design, AI will consider all these unique capabilities to come up with the best solution for Canon.
"So even if the solution is generated by AI, that could be a good solution that will be able to differentiate our product from the others."
You might also like…
Take a look at the best Canon cameras along with the best Canon lenses for DSLRs and the best Canon RF lenses for mirrorless. To see how Canon is employing AI in its top-tier cameras, check out my Canon EOS R1 review and my Canon EOS R5 Mark II review.
