The revelinai Interview

A conversation with MYSELF, AI Artist revelinai, aka @revelinai on X

INTERVIEW CONDUCTED ON AUGUST 10, 2024

ALL IMAGES BY REVELINAI ARE GENERATED USING MIDJOURNEY UNLESS STATED OTHERWISE

A surprise revelinai interview. And why not?

I love the number 17. I was born on the 17th, and a few weeks ago, I realized that today, the 17th of August, I would be posting the seventeenth interview in this series.

The stars are aligned, as they say!

What better opportunity, then, to share a bit about who I am (and as I’ve been asked privately on a few occasions), answer some of my own interview questions, and properly introduce myself to the AI community on X?

So, without further ado, here is a bit about me, revelinai.

Out of Time

Without getting too personal, can you tell us a bit about yourself?:

My name is Laurent, and I’m a male graphic designer with dual French and American citizenship. I specify "male" because, while my X avatar is a woman—which has led to some amusing confusion with a few of my loyal followers—I am indeed a man (and no, I do not identify as a woman!).

I’m a true Gen-Xer and a huge fan of what I like to call "quality popular culture." I’m also an avid cinephile, and I was fortunate to grow up in one of the best eras—the 80s—shaped by fantastic films, the world of Dungeons & Dragons and other tabletop role-playing games, classic cult books, visually stunning graphic novels, great music, and beautifully crafted video games from the advent of the PlayStation 2 era and beyond (yes, I’m a gamer too).

On a personal note, I’ve been happily married for 12 years, and we have beautiful children, two cats—Max and Sprat—and a dog, Wilbur, a gorgeous American Staffordshire Terrier who’s the best dog-buddy you could hope for.

Stranger Things / The World of Dungeons & Dragons

Yours truly is in this vintage photo (hint: while I often played as a fierce warrior, I usually most enjoyed being the Dungeon Master).

Could you please tell us which country you live in?:

While I am originally French, I have lived in the United States for the past three decades. My wife and I are currently residing in the beautiful state of New Hampshire, where we've been for the past three years. Before that, we spent about ten years in and around Austin, Texas, and lived in California and Seattle even earlier. At this point, you could definitely say I know the USA better than France.

What led you to begin working with AI imagery?:

I love visuals. I love art, especially illustrations, and I used to spend a lot of time on the DeviantArt website. As a graphic designer and tech nerd, I’ve been following the progress of AI technology—and AI art in particular—for a while. In late 2022, I tried OpenAI's DALL-E and was blown away by the ability to generate images from text prompts. Then, in early 2023, I discovered Midjourney with the launch of version 5.0 and was absolutely floored by the aesthetics it could produce. I immediately subscribed, joined Discord, and the rest is history.

What AI tools do you use?:

Midjourney, 99% of the time. In my opinion, it’s still the best AI image generator out there, though I’ll admit I’m very pleasantly surprised by what can be achieved with other tools, such as Leonardo, Adobe Firefly, and now Flux/Grok-2.

I’m also very curious about Stability AI's Stable Diffusion and would like to install it locally on my Mac and experiment with it, but I haven’t gotten around to it yet.

Note that while I have a Runway Gen-3 subscription and have created accounts for almost every AI tool that exists—from Luma to Pika to Kling (as I’m sure many of you have as well)—I currently have no spare time to jump onto the AI animation train, which is a shame because I really want to.

What inspires you?:

Everything! It’s such a cliché thing to say, but it’s true. Graphics in all the mediums I mentioned earlier—as we understand them to be the visual aesthetic and design of any given artwork—are always inspiring to me, whether it’s a book cover, a particularly well-framed shot in a film, its cinematography, its poster, a face, a tone, a mood, a scene, and sometimes even a sound or a smell in real life. To this day, I’m still inspired by all these things on a daily basis.

I also love good storytelling and find it inspiring on many levels. Looking at other AI artists' work inspires me as well, of course, and bittersweet moments, whether in real life or fiction, inspire me even more.

Undeniably, I have a penchant for fantasy, science fiction, and the feminine, with its elegance and timeless beauty. So, obviously, my AI art reflects a lot of that… and I also love monsters and dragons!

The Dream of Dragons

In which other medium, if any, do you practice art?:

I went to art school in Paris and became a graphic designer afterward, so in that respect, you could simply say I’m a graphic artist—one who loves good composition, interesting visuals, and elegant aesthetics. I was an avid drawer as a teen, particularly with ink and black-and-white, so I’m pretty good at sketching overall, and I even did some storyboarding for a while. I also dabble in photography (and have been for a long time), but I would say I’m only a decently skilled amateur. I never took to painting for some reason, as I’m not very good with oils, gouaches, or colors when it comes to applying them to a real canvas, though I do enjoy watercolors and have produced decent pieces in the past.

Would you consider AI-generated art true art?:

Absolutely! It’s a heated debate at the moment, but let’s be clear: AI image generators are tools, period. When a person uses a tool to create something beautiful to look at, it is art.

Imagine when the first cameras were introduced—how must painters have felt and reacted? This magic box could suddenly capture figures, settings, and, with the click of a shutter, immortalize composed scenes, bringing them to life on paper through some obscure chemistry. Those painters, after mastering hundreds of years of visual art, must have been quite upset at first. Yet, by today’s standards, we recognize photography as an art form, do we not? (In fact, one of our daughters is actually an artist-photographer.)

I see AI art in a similar light. AI art generators are new tools that will take time to be fully accepted, but collectively, we are all working toward accelerating that acceptance, and there’s no reversing the process. We have to embrace it, adapt, or be left behind.

Art needs to be beautiful, inspiring, mesmerizing, and provoke an emotional response. But it HAS TO BE beautiful; otherwise, it’s not really art—it’s something else.

So, yes, AI-generated art is art (and so are video games, by the way, though not all—but that’s a whole other debate).

Please share one or a few of your favorite images with us.:

I’ll share 17 (of course) of my favorite images below.

Some of my favorite images

Do you title your AI-generated art? If you do, what inspires you to come up with these titles?:

Yes, absolutely! I love to title all my pieces (with only a handful of “Untitled” ones among nearly 2,500 published images). I usually come up with titles on the spot when selecting what I’m going to post on X. Sometimes, it takes me a bit longer, and I have to think about it for a minute or two, but overall, I would say that I go with the raw emotions the image conveys to me when I’m looking at it. The feel of the movements and the dynamics found in certain images can also inspire some great titles and maybe even short little stories.

When do you tend to be the most productive, and do you work in long sessions or short bursts?:

In the early days, over a year and a half ago now, I would spend long hours playing with prompts and AI art. My life is really busy at the moment, so currently, I work in short sessions, usually late at night.

Sometimes, I also create randomly throughout the day when inspiration strikes. In such cases, I’ll quickly do it on my iPhone via the Midjourney website on Safari. It always feels a bit odd, as I prefer working on a large-screen monitor, but it works surprisingly well.

Honestly, it amazes me that we can produce images on the go like this. I even did it at 30,000 feet in an airplane, flying back from Europe last June. We take this for granted, but think about it! You're up in the air, on your cell phone with free Wi-Fi and no cables connecting you to the Internet, you come up with a text prompt, and a website outputs a beautiful series of images in under a minute. How insane is that? Like I said, we take it for granted, but… wow! The things we can do! What a time to be alive!

The Meteorologist

What type of prompts do you prefer: text or AI-generated descriptions through fed images?:

I like both options, though lately, I lean more toward text-only.

I went through a period where I fed the AI a lot of artwork and photos (both mine and others) and played with the four-prompt output that Midjourney generates via the /describe command.

One thing I will say is that I was very surprised to learn that among the 16 artists I have interviewed so far, only a handful seem to experiment with “feeding” the AI some imagery and building from there. It seems everyone else always opts for 100% text prompts, and I’m not really sure why. Maybe it’s for better control of the finished piece (and some are really, really good at it). I don’t know… but for me, having the AI interpret an existing image is a step toward figuring out a style and then “imitating” that, in a flattering way, with your own subject and context. We are all standing on the shoulders of giants here, and I find it fascinating that we can create fantastic AI imagery using the foundations of pre-existing beautiful artistic pieces—whether they be photos, paintings, illustrations, or otherwise.

When prompting with text, do you write simple text-based prompts or complex ones?:

Here, too, I’ll go with both answers. Sometimes I compose very complex prompts, but I never quite get the result I’m looking for, so I end up spending many hours on something I’m not entirely satisfied with. On the other hand, I also use a lot of short prompts, sometimes down to a single word, adding style references, a personalize code, and playing with parameters. To my surprise, this almost always produces wonderful imagery; Midjourney is absolutely great for that!

Lately, my go-to technique is to use a pretty complex master prompt I’ve been refining for over a year, and then I edit the parts where I want to change the image subject, color scheme, and some of the parameters. Some will recognize my current so-called style of illustrations (medieval, dragons, etc.) as being a product of that workflow. I like it a lot, and the impressions I get on X with those images are always in the thousands, which is great!

I would add that, since the introduction of sref codes in Midjourney, I also like to “collide” styles and see what comes out. Plus, there’s now also the new personalize parameter in Midjourney, and I’ve been really pleased with the results. (To get your own personalize code, you have to rank images on the Midjourney website. I’ve ranked thousands and thousands of images; I was actually doing that a lot even before I started producing art and posting it on X.)

Do you think text-based prompts should be shared within the AI art community?:

It's really up to the artist. I love seeing what other people create, and I'm often very curious about how they achieve a certain look, but I completely understand and respect that most artists want to keep their prompts secret. I have no problem with that; in fact, I think the mystery of not knowing adds a bit to the magic of the entire process.

I do appreciate the likes of AI researchers, such as @LudovicCreator and a few others, who share prompts so that artists can try them, modify them, and experiment further on their own. I think it’s great!

What is the most unconventional method you have used to create an image?:

I love to experiment with any system, software, or tool to push their limits. I always like to think outside the box and see what a digital tool can really do, so with AI tools, I enjoy pushing the envelope and trying unusual things.

For example, I’ll use Midjourney's /describe feature to feed it an image, run the four prompts it generates, reroll a few times, pick my favorite result, and sometimes even slightly modify the prompt, continuing to refine it until I’m truly satisfied with the image. Then, I’ll feed that AI-generated image back into Midjourney, use the /describe feature again—so the AI is essentially now describing its own creation—and run the new prompts to produce even more variations. It’s a lot of fun and can create surprising results. I’ve even experimented by feeding AI art from one generator into another to see what happens (which is not quite the same as simply copying and pasting a prompt across different generators).

How many images you have generated using AI technology?:

I’m at nearly 20,000 images in Midjourney, which is really nothing when you think about it. Some artists have generated hundreds of thousands of images, and creators like @nickfloats, @blasfemiadigit, and @chetbff are well over half a million. It’s truly mind-blowing, and I’m in awe of these guys, but it does make me feel really, really small.

The Maiden and the Mentor

Please share your X (Twitter) handle with us.:

@revelinai

Tell us a bit about your X handle and X username.:

revelinai stands for "Revel in AI," and, I’d like to think, should be pronounced "revelin-eye."

The idea for my handle came from a line in Blade Runner, one of my favorite films, where Dr. Eldon Tyrell has a conversation with his progeny, the replicant Roy Batty, who wishes to have his 4-year lifespan extended. Before Roy kills his creator—his “Father”—they have the following exchange (which can be seen here, on YouTube: Blade Runner - The Tyrell Corporation):

  • Dr. Eldon Tyrell: The light that burns twice as bright burns half as long—and you have burned so very, very brightly, Roy. Look at you: you're the Prodigal Son; you're quite a prize!

  • Roy Batty: I’ve done questionable things.

  • Dr. Eldon Tyrell: Also extraordinary things. Revel in your time.

That "Revel in your time" line always stuck with me. I found it fitting that this replicant engineer—this God of Biomechanics—uses it toward Roy Batty, his creation and masterpiece, and what could only be described as the ultimate form of Artificial Intelligence.

Of course, the absolute best line of Blade Runner—and maybe one of the best lines in cinema history—is Roy Batty’s soliloquy at the end of the film:

  • Roy Batty: I've seen things you people wouldn't believe. Attack ships on fire off the shoulder of Orion. I watched C-beams glitter in the dark near the Tannhauser Gate. All those moments will be lost in time... like tears in rain... Time to die.

I love that poetic moment, that poignant and evocative reflection on memory, experience, and the inevitable passage of time, as well as the melancholy and existential contemplation it triggers. But, I digress...

The bottom line is that we live in extraordinary times, and life is short! So, yes, we have to revel in AI.

How many X followers do you currently have?:

I just passed 3,000 followers, going from zero to 3,000 in a little over six months—all organically, simply by participating and posting a lot of images. My goal was to see what could be achieved in that time span, and, well, I’m pretty stoked with the results so far! Of course, I’d love to add a zero or two to that figure, and we’ll see what happens down the road.

How many X accounts do you currently follow?:

About 750, which I think is actually way too many. To be honest, if I follow someone, I want to truly follow them—I want to see what they create, admire their art, and maybe converse with them occasionally. If I followed thousands and thousands of creators, I would have to spend all day scrolling on X, and that’s just not feasible.

How long have you been on X (Twitter)?:

I’ve only been on X since December 2023 with the @revelinai account, but I was hanging out in the AI Community for about a year before that, 'creeping silently,' with an older Twitter account I’ve had since 2011.

Valk

Is there any other online platform where we can view all of your previous work apart from X?:

Yes, right here on my website at revelinai.com! 😅

I hope I can keep updating and maintaining this site for posterity, as a sort of AI artwork portfolio, if you will. I like it—it's elegant—and perhaps further down the road, I will expand its content to become an AI artwork repository for other carefully selected AI artists. The name Revel in AI doesn’t necessarily need to apply to me alone. We shall see.

In this world of cryptocurrencies and non-fungible tokens (NFTs), are you currently selling your digital art on any platform? If so, which platform are you using?:

No, not quite yet, but I’m setting things up on both Foundation and objkt to give it a try. @andy_corah, @Alfred_Denes, and @blind_imagery have been really helpful in that regard, and I will soon proceed to put some of my most-liked art (by me and by X users) on there. I’m planning to rework each image with a very specific workflow, starting from an original Midjourney master, followed by an upscale for higher resolution (using Magnific or a tool like Topaz Labs), then making slight Photoshop edits, and finally adding a “revelinai” signature. I want those carefully selected pieces to be unique and prestigious, limited in availability, and I do not want to sell them too cheaply, either.

Are you selling tangible AI-generated artwork, such as high-quality prints like Giclée prints?:

I would love to—and was planning to—but it doesn’t seem there’s much of a market for Giclée prints anymore. I’m a sucker for real quality prints (I collect movie posters), and when a gorgeous piece of art is properly framed, it can be stunning in a room or office. That said, the cost of framing artwork is very expensive (for example, you want to go with non-reflective museum glass—trust me), and so the entire endeavor, from ink to paper to framing, quickly becomes very, very pricey.

I had an idea to do AI art challenges, here or on X, and reward winners with a signed Giclée of their choice, but I’m not sure if I will ever do it.

Could you suggest some AI artists you like that we could follow?:

I would definitely suggest you follow at least all the great 16 artists I have presented here and on X through the AI Artist Interviews series:

Also, follow these five artists whose interviews are coming next:

Beyond that, there are probably too many to list. I would love to suggest hundreds and hundreds of the people I follow (and I hope to be able to interview some of them soon), but here is a shortened list of 50+ must-follow artists, if that helps (and I apologize if I forget anyone):

And many more…

Is there anything else you would like to add or share with the AI community?:

The AI community on X is truly awesome, and it’s been a real joy becoming part of it. Everyone I've encountered so far has been incredibly friendly, positive, and supportive, and I want to thank you all.

I believe we are all part of something truly special at this moment, and it's important that we keep going, pushing, and sharing. We are living in extraordinary times, and a few years from now, we will look back at this starting point with amazement and humility, knowing that we were right there, together, at the beginning of something bigger than ourselves. We are pioneers of a new era, a new artistic chapter in the human saga—part of a life-changing momentum—and it’s a beautiful thing!

Good night

More fantastic imagery (I think) from me, revelinai

Follow me @revelinai

Wilbur, First of his Name and Kong Master

(real photo) 😉