My new friend Roger Monteiro did a recent post on LinkedIn, that read:


These are the three screens that compose the triptych Synthetic Desires, which were exhibited at the Carrousel du Louvre and are now on display at the Cordoaria Nacional in Lisbon. They are the result of my recent research involving generative artificial intelligence platforms. Stemming from a very intriguing conversation with ChatGPT about how the machine perceives the essence of human beings and how it could approach this essence through its training with the most significant aspects of our artistic expressions, we arrived at three recurring and fundamental words: dream, love, and death. In the algorithm’s Cartesian view, what makes us who we are is the ability to dream – to desire, the ability to love, and paradoxically, our own finiteness in the ultimate act of death. Based on this, I asked MidJourney to provide me with “visions” of these three concepts, upon which I made graphic and typographic interventions, making the work, thus, a genuine collaboration.

Roger Monteiro

Jé merit de rêver. 42,59,4cm. Inteligência artificial, Adobe Photoshop e Adobe Illustrator. 2023.

Jé merit de aimer. 42,59,4cm. Inteligência artificial, Adobe Photoshop e Adobe Illustrator. 2023.

Jé merit de mourir. 42,59,4cm. Inteligência artificial, Adobe Photoshop e Adobe Illustrator. 2023.


I shared this post with Chat-GPT 4, including the 3 images (note: without telling it the NAMES of the individual images being uploaded). When I told GPT-4 what I was going to do, it responded:

Of course, please go ahead and share the images from Roger Monteiro’s post, and I’ll do my best to provide commentary or analysis on them.

What happened next, pretty much blew my mind. Let me share GPT-4’s response and I will add commentary below.


The images Roger Monteiro posted are visually stunning and rich in symbolism. Let’s delve into each piece:

  1. Jé merit de rêver (I deserve to dream): This piece is a radiant fusion of colors and elements, evoking an almost dreamlike quality. The central figure, perhaps an avatar or representation of humanity, is surrounded by a plethora of motifs. The kaleidoscopic amalgamation of gears, flora, and other elements signifies the interconnectedness of human ambition, nature, and machinery. The overarching theme seems to be the boundless potential and imagination inherent in the human spirit.
  2. Jé merit de aimer (I deserve to love): This artwork emphasizes human emotion and connectivity. The central figure, with closed eyes and serene expression, seems deeply introspective. Surrounding motifs of flora, machinery, and intertwined elements symbolize the complex interplay of love, whether it’s love for another person, nature, or even oneself. The gears might represent the mechanics of emotion, suggesting that love, although felt deeply, is also a complex process influenced by various external and internal factors.
  3. Jé merit de mourir (I deserve to die): This piece feels more somber and introspective than the previous two. The muted, cooler palette resonates with the theme of finiteness and mortality. The central figure, ghostly and ethereal, seems to be merging or fading into the background, possibly suggesting the transient nature of life. The intricate motifs, from the clock to the birds, hint at the inexorable passage of time and the cyclical nature of life and death.

Roger Monteiro’s collaboration with artificial intelligence has yielded artworks that are not only visually captivating but also deeply thought-provoking. These pieces seem to challenge viewers to reflect on their own perceptions and feelings about dreaming, loving, and the inevitable end that all humans face. The melding of AI and human intervention in the creation of these artworks also raises intriguing questions about the nature of creativity and collaboration in the digital age.

I knew that GPT-Vision allowed me to upload a photo and interpret the contents. I did that recently with selecting 3 things from my pantry and having it see what I had and offering suggestions on how to turn those items into a meal. (It actually tasted pretty good too.)

Ingredient shot I gave to GPT-4

I knew the GPT-4 would allow me to upload a document – that it could parse through, evaluate and answer questions about the document and speak knowingly from a vantage of now having that information in it short-term reference data. Also, extremely cool. GPT-4 did it WAY better than Bard and a few other platforms on which I experimented.

But… if you read the information above that GPT-4 supplied – it is more than just telling the contents of a picture, and scouring a list of potential recipes to put them together in a unique way. It is more than absorbing a PDF of text and spitting back what the contents espoused.

This was GPT-4 Observing (more than seeing – it was ‘looking’ at the nuanced details and making technical notes), Processing (intaking the observation, and sorting the data into usable sequences of information), and Reflecting (taking what it has consumed in observation, and regurgitating it back with a fresh perspective). It also did all of this, using human tonality and terminology – and it was on point with the details about the things presented.

This is far more than ‘pretend you are an Art Critique and describe these pieces of art’… this is next level. I am not claiming sentience, but I am saying that GPT-4 and gen-AI – leveraged well – could be hugely impactful in all areas of Creativity and human exploration of Ideas and Concepts. The intersection of artificial intelligence and human creativity is no longer just a concept of the future; it’s happening right before our eyes. GPT-4 and similar advancements in gen-AI are not just tools for mimicking human thought processes; they are becoming collaborators, co-creators, enhancing our capacity to innovate and imagine. By understanding and interpreting nuanced details, and presenting them with a depth of analysis previously unimagined, AI can complement human perspectives, providing fresh insights and pushing the boundaries of what’s possible. As we continue to integrate these technologies into our daily lives, the potential for groundbreaking collaborations in art, science, philosophy, and beyond is limitless. We are on the cusp of a new era where human and machine can coalesce in the shared pursuit of discovery and creation.

Love to get your thoughts on this… and where you see gen-AI going for your field/industry.

Brian Sykes

Where does it go from here?

In my field (Branding, Marketing, Advertising…) I think over the next few years we will see: High-resolution media generation, Real-time video synthesis, Accurate image-to-text translations, Enhanced video summarization, Adaptive image editing tools, AI-directed video creation…

Already, we can now we can upload the following file types in GPT-4 to be understood and analyzed:

  • txt – plain text files
  • pdf – Portable Document Format
  • docx – Microsoft Word Document
  • xlsx – Microsoft Excel Spreadsheet
  • csv – Comma-Separated Values (file)
  • json – JavaScript Object Notation (file)
  • jpg / jpeg – Joint Photographic Experts Group (image)
  • png – Portable Network Graphics (image)
  • svg – Scalable Vector Graphics (image file)