Steal These Prompts

In this episode of the Art Marketing Podcast, we dive deep into the transformative power of AI for artists and creators. Discover how to leverage AI tools to streamline your art business, save time, and enhance your creative process. We share actionable prompts you can steal right now to elevate your marketing efforts and connect with your audience. Join us as we explore the future of AI in the art world and how you can stay ahead of the curve!

Podcast Transcribe

Patrick Shanahan: Coming up on today's edition of the Art Marketing Podcast, I want you to steal these prompts. Steal them. Specifically though, some thoughts on AI broadly, uh, where we're headed next, and then again, some prompts you can steal right now to level up your art business. Okay, so hot off the heels of an episode a couple back that I did on prompting—and thank you for all the messages and comments by the way. I loved all that feedback. Um, I'm actually going to read one of them because I was so impressed by it. And this is from Jennifer on YouTube. Thank you for this, Jennifer. And I quote:

"Love, love, love this episode, Patrick. I implemented your advice this weekend and spent a few hours playing around with writing prompts. I created one for a non-artreuring work project that cut my time by 90 plus minutes and will save me so many hours going forward. I'm obsessed and my head is spinning with so many ways I can use AI to help me. Learn to prompt was the secret sauce I needed to understand how to use these tools. You rock." End quote.

So, why am I so bullish on AI? What is Jennifer going to do with an extra 90 minutes weekly? Maybe it's more than that. And this is a lot of time, and especially as you stretch it out over a year and you know you look at those gains—and are those gains waiting for you? Are they waiting for you in one aspect of your life or one aspect of your business if you just take some action on this? I think they are. I think they are for all of us.

And I realized more broadly, you know, it's time to double down on AI for some episodes. And you know, as I've marinated on things and thought about AI, I mean, I don't want the Art Marketing Podcast to have to be renamed the Art AI Marketing Podcast. But AI is going to play such a huge role. None of us can afford to ignore it. We're going to have to start getting into it.

How big is it going to be, Patrick? Right. And that's your question. How big is AI going to be?

So, quick aside—but for like the first time in history, um, the White House made a position on the issue. Trump created like a—I don't even know what you call it. It's not like an official government position, but it's not not an official government position—but he created the crypto and AI czar, right? And the administration nominated David Sacks, and this is an early PayPal guy, entrepreneur, tech investor. Um, also has a hugely popular tech podcast, and he had a post recently on X that I read and it sort of blew my mind.

So, I want to read that, make a larger point that I think will help you, will help me, will help all of us continue to navigate this AI landscape over the coming months and years. And his post was titled "The 1 Million X Boost."

And he goes on to say:

"AI will become a million times more powerful during the Trump presidency. People find this hard to believe. Let's break it down. AI models are improving at a rate of about 3 to 4 times per year. One can see this in the performance evaluations, the capabilities, the user experience and the growth of the ecosystem. Compounding at that rate yields an order of magnitude improvement—10x in 2 years and two orders of magnitude—100x in 4 years. Similarly, the chips are improving at a rate of about 3 to 4x per year. One can see that in the progress of Nvidia’s GPUs from Hopper to Blackwell, soon Rubin and Feynman. The progress is not just in logic but also in memory bandwidth, HBM, and rack infrastructure. So, similar 100x improvement in chip performance over four years. Finally, the data centers powering AI are growing. The state-of-the-art cluster like Colossus, which is what X has, has about 100,000 GPUs a year ago. Now that number is 300,000—already on its way to a million. From there 3 million—eventually 10 million will beckon for the most ambitious AI companies. Another 100x in 4 years. 100x better models running on 100x better chips running in data centers with 100x more chips. In total: an increase of a million x—one million times more AI."

That’s just staggering. Terrifying. All the things. All combined in one. And growth like that is pretty much impossible for humans to be able to wrap our heads around. Like growth like that—it's just never been coming in our lifetimes. Like that's just staggering, out-of-control growth. And that's not coming in our lifetimes—it's coming in the next few years.

If you believe what he said—even if he's being optimistic there—that is still a level of growth that like, as humans, we just can't wrap our minds around. And I would say that the sheer speed—and I mentioned this in other podcasts—the sheer speed of advancements already in AI is like nothing I’ve ever seen.

I bring this up as it's important for all of us to understand what is coming, how profoundly our lives are going to change whether we like it or not. And at least have a think on what our individual game plan should be regarding AI and starting to learn how to use it and put it to work for us.

And I actually remain quite bullish on the individual creator segment. And that’s all of you guys—all photographers, all creators, all stripes. You can even put me in there as a marketer. I think AI and AI-created stuff is going to end up facing a pretty strong backlash. And it's going to drive us humans to crave things created by other humans—not by machines. We're already starting to see some of this—to crave real stories to go along with those real things created by real humans.

And perhaps you feel that way, or perhaps you're like a glass-half-empty that the AI is going to come and get us and we're all going to die. We all need a strategy here.

So, what I want to do is tell you where we are, where the next phase has taken us, and what I think you should be doing and working on—what I’m advising anyway. And this will certainly build on my prompting episode. So, if you’ve not listened to that one, you might want to go back and start with that one a few episodes back. And then I’m going to conclude with some prompts that I want you to steal and give a shot in your LLM—large language model—of choice, that I think can really, really help you out.

Patrick Shanahan: So, where we are now, what’s coming—keep in mind is, as things stand today—and if you listen to this even nine months from now, it’s likely going to be pretty seriously out of date.

Let’s start with the LLMs—the large language models. You have ChatGPT from OpenAI. You have Gemini from Google. X, used to be Twitter, they have one called Grok. Meta has one that you’ve probably seen and nobody uses, but it’s in Instagram, it’s in Facebook. They call theirs LLaMA. You have Claude, which is from Anthropic. There’s one from France that a lot of people like called Mistral. A bunch of open-source ones that I won’t go into. And then also competitions from Chinese companies like DeepSea and such.

But broadly speaking, to keep things simple—they’re all about the same. You know, like the serious AI people have like a big scoreboard, kind of like you would see at a race or whatever, that ranks the models on this, that, and the other—like how good is it at this, how good is it at that, who is leading. What are the boards? Complete waste of time in my estimation to even look at them.

The only thing I know based on the last few months to a year is—even if your LLM of choice is the top for now, you are guaranteed one of the others will unseat it in a few weeks or months. They’re constantly changing who is in the lead and worrying about it is just a huge waste of time.

So, my advice is to pick one or a few of your favorites and just play with them. That’s on LLMs, okay? The actual chatbots, if you want to call them that—the chat windows, right? So just pick one and use it.

For those of you guys that use ArtHelper, we have a bunch of them in there already. We’re about to release an open chat that uses multiple models in there. So, if you’ve got that, you don’t even need to sign up for one. You’re going to have a bunch of them in there already. But that’s LLMs.

You know, up till now, this kind of moment in time—the last couple of months—aside from the various downstream apps that are based on the LLMs—what do I mean by that? I mean like apps that generate images, or apps that create audio, or apps that create video or that do captions for your video or transcribe documents or write, etc. Up till now, it’s mostly been all about the LLMs. They’re all just iterations of what these LLMs can do. And it seems certain for now they’ll continue to compete, continue to get better. There’s a ton of options, which is going to be great. It’s great for us as consumers.

But now things have started to change over the last couple of months, and there’s this new phase of AI—and phase of the AI agent, right? And it’s just—it’s called our agentic future. It’s called, you know, the emergence of the agent, right?

And if you’ve not heard this yet, you’re going to start hearing it all the time all over the place. Agent this, swarms of agents, chaining agents together. What does that even mean?

Okay, and really simply—you know, I’ve said this a ton, a ton of times and I love it. But the LLMs—the large language models—your Chat, your Grok, your LLaMA, your Claude—it’s just a huge, huge brain, okay? It’s just this giant brain that knows more than anyone you’ve ever talked to that we can use. Okay, great.

Now, with this future with agents, it’s all about tools connected to the brain, right?

And you know, what are some of the tools that I would use as an example? Uh, the tools could be your email. Uh, the tools could be your calendar. The tools could be your website backend. The tools can be a social media platform. Any web application that you use can be thought of as a tool.

So this notion of the agents, okay? Um, you know, it’s—I'll give you an example, because this is like, you know, the one that all of them use. Everyone loves defaulting to travel, right? And so you go into your LLM—your ChatGPT, your Grok, whatever—we’ll just say ChatGPT—and you’re like, “I am looking to plan a trip to London,” right?

That thing connects to a travel tool and it goes and gets you the flights. And then it comes back and it connects to Airbnb and it finds you an Airbnb. Let’s say you’re going to go to London. It gets you flights to London, and then another agent brain connected to the tool goes and gets you all of the Airbnbs. And then it comes back and it gives it to you in the chat window and you’re like, “Okay, that flight looks good. That Airbnb looks good. Go book it.”

And then those agents will go back to those tools, they’ll have your credit card, it will book that stuff, come back, your trip is planned—hooray—there you go, right?

But it’s just the big brain connected to a tool, okay?

Patrick Shanahan: And the long and the short of this is—you know back in the day when everybody had different cell phones and as such, everyone had different charger cables, so nobody could use anyone else’s phone? I mean, you have to be dated now. Everyone has an iPhone for the most part. But it used to be that like everyone had a different cell phone charger.

Well, what’s awesome is that potentially could have been the situation with these chat models—these large language models. Instead, they set a standard this time and they said, “Guys, can we all just have the same charger connection?”

Okay? And this new protocol is called MCP. We don’t need to go into it, but what’s going to happen in this future we’re about to live in is every single solitary third-party application you could possibly think about—and take your pick. Anything that you use in your daily life—Google Maps, maybe you’re on Outlook email, maybe you do stuff on Pinterest, maybe you’re browsing on Reddit, maybe you’re doing your shopping on Amazon or TeePublic or Etsy or just anything that you can think of that you interact with—all of that’s going to be able to be driven from the chat window itself, right?

And they’re kind of collectively being called agents that do all of that.

That is the future. That is where we’re going. There are going to be agents that can talk to agents. Agents that can be strung together to do incredibly complex things and it is going to be absolutely fantastic. Okay? It’s going to be amazing and you’re going to drive all of these agents via the chat window.

And that’s why I made that episode: learn to prompt. It’s the new language of our AI world. It’s why I’m going to be giving you prompts at the end of this episode—because I think all of us need to be learning it. All of us need to be getting comfortable communicating in this new language. So obviously, I want to share a few of those.

Before that, I want to do one final—one final learning recommendation piece of advice: don’t wait. Don’t wait. Ship. Get going. Don’t wait for things to be perfect and work the way that you want and give you the results that exactly you were looking for. Just start playing with things now. You guys—get a taste. Get comfortable with things. Start experimenting. All of you can do it regardless of your level of technical acumen, regardless if you’re old, if you’re young, if you’re skeptical about AI or not.

One of the things that I found certainly in my own practice as I’ve been doing this is—and you know, I see a ton of my customers and a ton that are not commenting on like, “Yeah, okay, I tried that AI thing and I got these results but these results were not what I wanted and they’re not perfect and I think they’re a waste of time and quite frankly I’m over it. I’m not going to touch this ever again.”

Okay. And what I found is—if you can just get past that point, you can experiment and spend some time on these LLMs and experiment with it and put in some prompts and start learning and learning what works and learning what doesn’t and not being super critical about whether or not what you got in return is perfect.

Even if there’s one thing you really want the AI to solve—and it did such a good job, it got you almost there, but it didn’t get you all the way. Don’t wait—because in three months that thing will be solved and it’ll be on to the next.

And it’s going to be the creatives that embrace this, that start playing with it, that start practicing with it, that get comfortable with it, that understand how to talk to it—that really, really win.

And I firmly, firmly believe that. Like my own personal rules—if I waste more than an hour of my time trying to get something to work and it fails, I just move on. I go move on to the next thing. Because what I found every time is after I circle back after a month or a couple of weeks, it just works perfectly. And it’s amazing because you just keep learning and getting better at prompting and how the apps work—and they all sort of work the same way.

And, you know, it’s a total sea change in the way that you think. And it’s sort of amazing. Like all of us are totally reliant on ourselves to do all of the jobs, right? Especially as creatives—because you guys are all solopreneurs. You don’t have a team to execute on these things for you. So, you’re dealing with the billing, you’re dealing with the scheduling, you’re dealing with the shipping, you’re dealing with taxes, you’re creating the art, you’re doing the marketing—all the things, right?

I can see this future that we’re moving into—and this is sort of comes up colloquially online as people are talking about it. We’re all going to move into a future where we are no longer individual musicians, right?

So to use my analogy—you’re the solopreneur. You’re creating—that’s an instrument in the band. You’re doing the marketing—that’s an instrument in the band. You’re doing the taxes—that’s an instrument in the band. You’re doing the customer service—that’s an instrument in the band.

You’re going to be able to move from having to play those individual instruments to becoming the conductor. The conductor of your organization, of your—what am I looking for? Your troupe, your group of musicians, your band, right? Your orchestra.

And as you start learning this, you see the pieces start falling into place. You get one little thing that’s working that saves you a ton of time there—kind of like that testimonial I read up top. And then you move on to the next one. And then you’ve got two and it’s like—it’s literally like that Disney movie where all the buckets are just moving and dancing. Or an orchestra, if you like buckets and mops and that Disney thing. But you’re conducting the music—and it’s a beautiful, beautiful, beautiful thing.

And so if I can just give you one takeaway: don’t wait. Show some curiosity. Take some initiative on this. Start learning to prompt a little bit. Once you do—once you get shipping—it’s going to fundamentally change everything that you do as a creative.

Patrick Shanahan: You guys wear too many hats, and I really think this is the future. So, okay— Steal these prompts! 

I’m going to put all of them in the show notes so you can just copy and paste them into the LLM of your choice—or you can copy and paste them. Some of these are already in ArtHelper, so Open Chat is where you would do them, but that’s okay. So they’re all going to be in there.

Is copying and pasting the prompts the most important part of this? No, it’s not. My goal is just to teach you how absolutely amazing it can be when you pair a giant PhD-level brain to your art business and your artistic challenges. Once you do, you will arrive at a conclusion—that is my goal for you.

Prompts are super important, don’t get me wrong. But what’s more important is making use of this giant brain to help you evaluate the quality of your thinking and all of the difficult tasks of life. And you know, it’s sort of like—you know, this uber successful, crazy talented businessperson that’s had a massive amount of success, is extremely wise, had a great education, all of these things—and they agreed to go out to coffee with you to talk about your business, right?

And you’re like, “Oh my gosh, this is amazing.” Like, you know, people would die to get an appointment with this guy. He’s the family friend, or somehow I know him, and he’s going to let me sit down with him or her—whatever the case. What would you do?

You would spend so much time getting your questions ready. You would be all nervous. You would sit down for coffee and be like, “This is a once-in-a-lifetime opportunity for this person—for me to have access to their wisdom and experience and ask all these questions.”

The chatbot is that person. The LLM is that person. And you can ask them questions at any point in time, for any reason. You can have coffee with them as many times as you want. All you have to do is ask. All you have to do is ask.

And when you think about it that way, it’s like—wow. Wow, wow, wow.

So what did I do? I put together just—it’s been really helpful for me in the past to hear and see how other people prompt. To kind of get me thinking, get me marinating. So you can steal these prompts verbatim. You can go and run with them. But again, my idea—and they’re going to be all down in the show notes, you can copy them—is that the idea is you’re like, “Oh my gosh, based on what I now heard, I didn’t know I could go and ask like that. I want to go and ask.”

So let’s just kind of hop around some art-related tasks:

PR / Press Releases

Prompt: “Write a 300-word press release for local media about my solo show opening June 15th, AP Style. Include pull quotes and boilerplate.”

Boom. You’ve got a sample press release going on.

Art Critic

Prompt: “Act as an art critic who knows my portfolio. (See link).” 

And in there, put the link to your website. If you don’t have a website, put the link to your Instagram. Somewhere else your art is up.

“Identify the three aesthetics through lines most visible in my work. And then suggest wording for an elevator pitch and three hashtags I should use.”

That one’s giving me some crazy good results.

Email Marketing

Prompt: “Outline a four-email onboarding sequence for new collectors focused on story, care instructions, referral incentive, and preview access. Include subject lines and send intervals.”

Now you have an email sequence you can go put into your email.

Interior Design

Prompt: “Scan Pinterest Trends and Google Trends for the last 90 days and tell me which interior design color palettes are spiking. Suggest an art series concept aligned to each.” Need ideas? Throw that one in.

Okay, let’s stay on interior design.

Prompt: “Pretend you’re a 35-year-old interior designer in [insert the name of your city, e.g. Austin], sourcing statement pieces. Draft an email explaining why my [painting/prints/work], titled [series name], solves their pain points using no more than 150 words.” Now you have an email you can send to any and all interior designers. Again, include your website in there. Include a couple of images. Upload a couple of images. If you’re in ArtHelper, use a collection, right?

Nurture Collectors

Prompt: “Give me a DM template (direct message or email or SMS) and five short interview questions to send past buyers so they’ll film phone-friendly testimonial clips. Include polite opt-out language.”

How many of you have collectors that have filmed where your work is hanging on their walls and sent it to you as marketing collateral? All you gotta do is put that one in and it will give you the language and you can just tune it up and send it to all of them.

Publicist

Prompt: “You’re a publicist. Draft a personalized pitch email plus one-page PDF outline to approach [insert gallery name] on representing my [insert series name]. Emphasize recent sales figures, press mentions, and how my pricing tiers fit their collector profile.”

Obviously, if you wanted to do that one, you’d get your sales figures together. You’d get any and all press mentions you have together. And then you’d get some of your pricing tiers and throw that in there. All of a sudden, you could start reaching out to publicists. All of a sudden, you could send that email directly to galleries.

Limited Editions 

Prompt: “I want to release a new 24x36 in print at a $750 pricing tier. Assume my audience size is [insert number of email subscribers]. My average open rate is [insert rate]. My conversion percentage is [guess 3%]. Each buyer buys one. What edition size maximizes revenue while maintaining scarcity? Show your math and give a drop schedule (teaser, early bird, public).”

Pricing

Prompt: “You’re a pricing strategist for fine art e-commerce. Build a price ladder for my three tiers:

• 0–$100 (merch, small prints), 
• $100–$1000 (books, limited edition prints),
• $1000+ (originals and large limited editions). 

For each tier, list:

1. A) 3 product concepts, 
2. B) perceived value boosters (like artist note or calendar upsell),
3. C) psychological price points to test,
4. D) recommended gross margin percentage.”

We teach you to do all that at Art Storefronts and on this podcast, but it’s nice to be able to put it all into an LLM and have it spit it out.

Commissions 

Prompt: “Act as an operations manager. I can complete X number of commissions per month at Y. My hourly cost all-in is Z.” You insert these values. “Build a Google Sheet structure that tells me:

1. Break-even price per commission,
2. How many monthly commissions I need to hit,
3. A desired income [insert your number], and warn when I’m over capacity.”

I’ve heard so many of you over the years struggle with commission structures.

Those were a random smattering of prompts for lack of a better term. But when you start to hear those, you’re like, “Oh my gosh—I can ask it those kinds of detailed questions?”

And the answer is YES. You can. And even better, the answers sometimes are absolutely mind-bogglingly good. Like incredibly good.

Now, in ArtHelper, we have a lot of these things in there for you already. So you don’t have to think about these prompts—you can just hit buttons, which is nice.

But overall, learning prompts is so, so powerful. And you know what no one tells you? It’s pretty good if you just ask it to create a prompt for itself.

You’re in ChatGPT:

“I’m trying to figure out what new directions to go into. Here’s what I’ve created so far.”

Perhaps you upload a couple of images.

“Can you tell me, based on my current collection, some new ideas and some new directions I should go in? Please give me a prompt for how I would ask the LLM that.”

Or:

“Here is the nature of my art business.” Give it some details.

“How many people on your email list, how many sales you’ve made, how many collectors you have, how many art shows you’re doing. Can you come up with some marketing strategies for me that I’ve never thought of before? Help me with a prompt for that.”

So you can ask it to come up with its own prompts that are about the nature of an art business.

And it’s—you guys—it’s just amazing. It’s amazing.

So: Steal these prompts and get prompting! And when you do, they give you ideas for new prompts and new directions, right?

And you should also sign up for ArtCult. We’ve been iterating like crazy in there, and you can now post both carousels and reels directly to Instagram from within the app, which is really, really cool. So if you haven’t tried that, it’s free. You guys can go try it: arthelper.com.

But moreover, the AI agents are coming very soon to ArtHelper. If you’re super technical, maybe you’re already playing with them.

But the number one thing in the world that all artists, all photographers, all creatives want is the work done for them. And I’m here to tell you—for the perhaps first time in the history of the world—this is about to be possible for artists of all sizes, regardless of their connections, of their luck, of the amount of time they have, of how much they’ve sold, of where they’re from, who they are.

What a future, you guys—that will be. What a future.

So thanks for listening, and as always, have a great day.