"AI equals IA - imagination amplification. It allows me to take my creativity to places I couldn’t otherwise reach."
- Daniel Ambrosi
In late 2024 we published the second annual Cultural Comms Luxury Insights Report on Art, AI and the Future of Creativity, in association with ArtTactic. This report drew upon the knowledge and expertise of leading commentators from across the art, luxury and media landscapes, and can be read here.
Over the next few months we will be sharing the full contributions from these commentators, beginning with Daniel Ambrosi: a California-based artist recognised as one of the founding creators of the emerging AI art movement, and noted for the nuanced balance he achieves in human-AI hybrid art.
Read on for his insights.
The intersection of art and artificial intelligence is one of the most fascinating and controversial developments in the creative world today. While some fear AI’s potential to disrupt or even replace human creativity, others see it as a powerful tool for amplification and innovation. In this interview, conducted in the summer of 2024, Daniel discusses how AI is transforming the artistic process; its impact on the fine art world; and the evolving relationship between technology and authenticity in art.
Balancing nuance and fear
"Part of the problem with fears around AI is the lack of nuance," Daniel Ambrosi explains. "Yes, there are aspects to be cautious of, but the conversation is often painted with too broad a brush. For example, I use an AI system that wasn’t created to mimic or use other people’s art, so I’m not subject to those copyright concerns. That said, it requires a conversation and an open mind to understand that.
Reflecting on historical parallels, he adds: "The same concerns arose with the dawn of the internet, smartphones, and photography. People thought these technologies would end traditional art forms, but they didn’t. Photography, for instance, deeply influenced art movements - it became part of the continuum. AI is just the next evolution in that journey."
Do you think AI signals the death of art or the beginning of something new?
"It’s just a progression. AI equals IA – imagination amplification. It allows me to take my imagination to places I couldn't physically reach. Even if I had the skills, I wouldn’t have the time. My grand-scale works are so intricate, it would take a lifetime to paint them.
"AI is an enabler. The only kind of art I've ever done - since I entered the computer graphics lab at Cornell in 1981 - has been digital. Working in the world of bits instead of atoms is a lot less messy. I've always been open to new technologies while being aware of their downsides. Every tool mankind has ever invented, starting with the hand axe, has been a double-edged sword. Creatively, AI has opened a whole world for me.
"Thirteen years ago, I had a computational photography breakthrough that enabled me to create high-resolution, vibrant, immersive panoramas. They looked hyper-real, like pre-Raphaelite paintings, but they weren’t just photographs. I had some success licensing my images for trade shows and events, like Toyota’s booth at the LA Auto Show, and placements in healthcare environments. But it wasn’t until AI came into the mix that the fine art world really took notice. I haven’t found anyone else doing grand-scale AI-augmented landscape photography like I’m doing. It’s opened up new creative and career opportunities for me."
Why do you think AI made the fine art world take notice compared to digital art, which has been around for years?
"In my case, the images were surprising and compelling. Nvidia commissioned me to create three large lightboxes for their GPU technology conference in 2016. From a distance, they looked like nice photographs, and people wondered why they were there. But when they got close, they saw something different, and their reaction was intense – they were curious and fascinated. The detail in the images surprised people because it wasn’t what they expected to see. This taught me that seeing is a creative act. People see what they expect until something challenges that, and once they see it, they can’t unsee it. This was a big eye-opener for me and made my work compelling. I think that’s partly why the fine art world took notice.
"Under the mentorship of incredible people like those at Robilant + Voena, applying my technique to Capability Brown's landscapes was brilliant. In the 1700s, Brown reshaped wealthy English estates to resemble landscape paintings that aristocrats loved. I come along hundreds of years later, capture those landscapes photographically, and turn them back into paintings. It’s a poetic narrative, and narrative is important in the fine art world."
You mentioned using a different kind of AI. Can you elaborate?
"The AI I use was part of a computer vision system Google was developing for image recognition, designed to classify photos, not create art. It was supposed to recognise images, but the neural network started producing strange results. The neural network had 84 layers, each layer getting closer to the right answer. But during training, one of the engineers, Alex Mordvintsev, developed a way to reverse the process and visualize what the AI was seeing at different layers. What he found was that the AI was hallucinating, showing signs of pareidolia, a trait where we see things that aren’t really there, like seeing shapes in clouds.
This diagnostic hack, called Deep Dream, went viral in 2015. People all over the world started creating psychedelic, dreamlike images. I saw an opportunity to use Deep Dream in a subtle way to push my landscapes into a dreamscape. But Deep Dream couldn’t handle the giant images I was creating, so I enlisted engineers to spend months modifying it for my purposes. Now, I’m the only one with access to this super-scale AI, which operates on an Amazon Cloud virtual server, and it creates these incredible works.
There’s a concern that AI will take away the authenticity of art, but you’re using AI authentically to create art.
It’s an opposite commentary to what people might expect. It’s just a tool, but it feels like a partner. It surprises and delights me consistently. It’s a clean example of a human-AI hybrid art form, about 50-50. I fully prepare my panoramic photograph, get it to exactly where I want it, and then hand it over to Deep Dream to see how the AI sees the world.
Daniel Ambrosi is a California-based visual artist specializing in digital and AI-augmented art. With degrees in architecture and 3D graphics from Cornell University, Ambrosi has explored digital art for over 40 years. In 2015, he collaborated with Joseph Smarr (Google) and Chris Lamb (NVIDIA) to enhance Google’s DeepDream technology, creating his immersive Dreamscape series. Inspired by grand-format European landscape painters and the Hudson River School, his work has been exhibited across the U.S. and Europe. A Lumen Prize finalist in 2019, Ambrosi’s art is in the Museum of Contemporary Digital Art’s collection.
