Hey Creative Friends!
In this episode, I'm sharing some advanced AI prompting tricks that'll help you break free from those bland, generic AI responses we're all tired of seeing. Remember those basic prompting principles from last week? Well, now we're taking the training wheels off! I'll show you how to get AI to surprise you with genuinely creative ideas instead of the same old predictable stuff everyone else is getting.
🎁 Freebie link to my advanced prompt templates (copy/paste ready - no email signup required) - HERE
What You'll Learn (The Good Stuff)- It's like working with a senior creative, not an intern - Stop micromanaging AI and start setting up creative sandboxes where it can surprise you with brilliance
- Make AI show its work - With "chain-of-thought" prompting, AI has to explain its thinking step-by-step
- Get multiple options instead of just one - Why settle for a single idea when you can explore several directions at once?
- Add creative constraints - Just like how haikus or sonnets force poets to get creative, limiting what AI can use pushes it to find fresh solutions (this one's great for breaking out of creative ruts!)
- Reverse-engineer better prompts - Describe what you're dreaming of and ask AI to create the prompt that would make it happen (I stumbled on this technique a year ago and it's been a lifesaver)
- Cast AI in detailed character roles - Instead of just saying "be a copywriter," create a vivid persona with specific experience and quirks for more nuanced feedback
- Mix and match techniques - Combine these approaches for exponentially better results (though your creative judgment is what makes it all work)
The AI Tools:
- ChatGPT with that "deep research" button (great for when you need AI to really think things through)
- Claude (my go-to for reverse prompting... and everything else writing related, for that matter)
- Bing Image Creator (what I was using for image generation before I found a better workflow)
The Advanced Techniques:
- Chain-of-Thought Method - Like asking "how would you solve this?" instead of "what's the answer?"
- Comparative Prompting - Gets AI to explore multiple creative directions at once
- Constraint Prompting - For when you need to break out of predictable, boring patterns
- Reverse Prompting - Tell AI to create the prompt for you
- Immersive Role Playing - Makes AI feedback actually useful and specific
- Expert Dialogue Approach - Creates a virtual creative roundtable discussion
00:00:33 - Refresher on what we covered last week
00:02:03 - Why advanced prompting is like collaborating with a creative director
00:03:44 - Chain-of-thought prompting: Getting AI to show its creative process
00:05:37 - Comparative prompting: How to get multiple creative directions
00:07:59 - Constraints-based prompting to force more creativity (just like jazz musicians)
00:09:41 - Reverse prompting: Asking AI to write prompts for you
00:11:12 - Immersive role-playing prompting
00:13:49 - Case study: Applying advanced techniques
00:17:46 - Fitting these techniques into your actual day-to-day creative work
00:19:49 - Quick recap and what to try this week
Try These This Week- Break down your thinking - Next time you need an AI concept, ask it to walk through 4-5 steps before giving the final idea. Instead of "Give me a logo concept," try "Before creating a logo concept, analyze what makes good logos work, consider symbolic approaches, explore color psychology, etc."
- Get options, not answers - For your next creative challenge, identify 2-3 completely different directions and ask AI to explore each one. The magic is in seeing multiple possibilities side-by-side!
- Add weird constraints - This is seriously fun: tell AI it CAN'T use certain words or phrases in your next content task. I tried this with a client's Instagram captions when we were stuck in a rut, and the results were so much fresher.
- Start your own prompt collection - I'm sharing all mine in the show notes, but start saving the prompts that work well for YOU. Trust me, future-you will be grateful when you're on deadline.
- Run the A/B test - Try both a basic prompt and an advanced one for the same project and see the difference.
🎁 Grab all my advanced prompt templates (copy/paste ready!) HERE
Let's Connect!00:00:00 --> 00:00:04 Are you still doing the creative tasks that AI could handle in a fraction of the time?
00:00:04 --> 00:00:10 I used to spend four hours on tasks that now take me 30 minutes with better results.
00:00:10 --> 00:00:13 That's not just efficiency, it's creative freedom.
00:00:13 --> 00:00:16 I'm Tiana and this is AI for Creative Professionals.
00:00:17 --> 00:00:32 Music.
00:00:33 --> 00:00:37 Hey there, welcome back to AI for Creative Professionals. I'm Tiana,
00:00:37 --> 00:00:40 and we're already on our third episode together.
00:00:40 --> 00:00:44 So last week, we covered quite a bit of ground, right?
00:00:44 --> 00:00:48 We talked about the seven fundamental principles of effective prompting.
00:00:48 --> 00:00:55 Being super specific, providing context, specifying format, including examples.
00:00:56 --> 00:01:02 Using role prompting, iterating, and setting system prompts for longer conversations.
00:01:03 --> 00:01:07 And then we also looked at those frameworks for different creative fields.
00:01:07 --> 00:01:12 The CATP approach for writers, the VCRT method for designers,
00:01:12 --> 00:01:16 and the PAISM framework for marketers.
00:01:16 --> 00:01:20 If you missed that episode, definitely go back and check it out because this
00:01:20 --> 00:01:23 episode is a continuation of episode number two.
00:01:23 --> 00:01:26 In this one, we'll discuss the advanced prompts.
00:01:27 --> 00:01:32 So these advanced prompts can really take your AI interactions to the next level.
00:01:32 --> 00:01:37 We will explore stuff like chain of thought prompting, comparative approaches,
00:01:37 --> 00:01:43 constraint-based creativity, and we'll build on those basic principles we established last time.
00:01:43 --> 00:01:47 And by the end of this episode, you'll have some powerful advanced prompting
00:01:47 --> 00:01:52 tools that can definitely help you solve more complex creative problems and
00:01:52 --> 00:01:53 generate more innovative ideas.
00:01:54 --> 00:01:59 And basically, you will create outputs that don't feel generic at all.
00:01:59 --> 00:02:03 It's the thing we're all trying to avoid, right? So let's dive in.
00:02:04 --> 00:02:08 Before we get into specific techniques, let's make an important mindset shift.
00:02:09 --> 00:02:13 Basic prompting is, you know, communicating clearly to get what you want.
00:02:13 --> 00:02:19 But advanced prompting is more about creating conditions for AI to surprise
00:02:19 --> 00:02:22 you with something better than what you explicitly ask for.
00:02:23 --> 00:02:27 Think about the difference between directing a junior designer versus collaborating
00:02:27 --> 00:02:29 with a senior creative director.
00:02:29 --> 00:02:32 With a junior, you're specific about exactly what you want.
00:02:33 --> 00:02:36 But with a senior, you share the problem, constraints, and goals.
00:02:36 --> 00:02:39 And then you give them space to bring their expertise to the table.
00:02:40 --> 00:02:44 And advanced prompting works kind of the same way. So you're setting up a creative
00:02:44 --> 00:02:49 space where the AI can do its best work rather than you micromanaging every little detail.
00:02:50 --> 00:02:53 So let's say you're working with a client on naming their brand.
00:02:53 --> 00:02:58 And when you use basic prompting, you'll get somewhat decent serviceable names, right?
00:02:58 --> 00:03:03 Nothing terrible, but nothing special either. But if you start using advanced
00:03:03 --> 00:03:08 techniques, you'll definitely start getting these unexpected creative options
00:03:08 --> 00:03:11 that open entirely new directions for you.
00:03:12 --> 00:03:18 So the basic prompt would be, give me some good names for a sustainable skincare brand.
00:03:19 --> 00:03:24 But the advanced prompts is what gives the AI conditions to think more deeply
00:03:24 --> 00:03:27 about the naming process itself.
00:03:27 --> 00:03:33 And as we explore these techniques, keep this shift in mind from how do I ask
00:03:33 --> 00:03:40 for exactly what I want versus how do I create conditions for emergent creativity, right?
00:03:40 --> 00:03:43 It's a subtle change, but makes a huge difference.
00:03:44 --> 00:03:50 Let's start with chain of thought prompting, and it's where you Ask the AI to
00:03:50 --> 00:03:55 walk through its reasoning step by step before giving you a final output.
00:03:55 --> 00:04:00 Think of it as the difference between asking someone, what's the answer versus
00:04:00 --> 00:04:02 how would you solve this?
00:04:02 --> 00:04:06 The second approach usually leads to more thoughtful results.
00:04:07 --> 00:04:12 So instead of starting with something like, create a logo concept for my financial
00:04:12 --> 00:04:17 wellness app, you can try something like, before creating a logo concept for
00:04:17 --> 00:04:20 my financial wellness app, could you think through?
00:04:21 --> 00:04:27 Number one, first, analyze what makes effective fintech logos work and what
00:04:27 --> 00:04:28 visual clichés to avoid.
00:04:29 --> 00:04:35 Two, then consider how to visually represent financial stability and emotional well-being together.
00:04:36 --> 00:04:41 Three, next, explore different symbolic approaches and their pros and cons for this brand.
00:04:41 --> 00:04:46 Four, consider color psychology that would convey trust, growth, and calm.
00:04:47 --> 00:04:52 Five, finally, develop three distinct directions based on this analysis.
00:04:52 --> 00:04:57 Please show your thinking for each step before presenting the final concepts.
00:04:57 --> 00:05:03 Essentially, what this prompt does is it prevents the AI from jumping to generic solutions.
00:05:03 --> 00:05:10 By forcing it to show its work, you get better results and insight into the reasoning behind them.
00:05:10 --> 00:05:14 This technique also helps avoid those AI hallucination moments,
00:05:14 --> 00:05:19 which I think all of you already experienced, you know, when AI confidently makes stuff up, right?
00:05:20 --> 00:05:25 But when you force it to show its reasoning, those logical leaps become a lot more visible.
00:05:25 --> 00:05:30 One helpful tip, break down the thinking process into clear sequential steps
00:05:30 --> 00:05:32 that build on each other.
00:05:32 --> 00:05:36 The more logical the progression, the better results you'll typically get.
00:05:37 --> 00:05:41 Another powerful technique is comparative prompting.
00:05:41 --> 00:05:46 And it's when you ask the AI to explore multiple approaches simultaneously instead
00:05:46 --> 00:05:48 of pursuing a single direction.
00:05:49 --> 00:05:53 Think of it like gathering input from a diverse team rather than just one person.
00:05:53 --> 00:05:59 And what happens there is that the richness of multiple perspectives often leads
00:05:59 --> 00:06:01 to, you know, more innovative solutions.
00:06:01 --> 00:06:06 Instead of saying, write a homepage headline for my My Productivity app,
00:06:07 --> 00:06:11 you can try something like, generate three different approaches to a homepage
00:06:11 --> 00:06:16 headline for My Productivity app that helps teams coordinate across time zones.
00:06:16 --> 00:06:20 Approach one, focus on the pain point of coordination difficulties.
00:06:20 --> 00:06:24 Two, emphasize measurable productivity benefits with metrics.
00:06:24 --> 00:06:29 Approach three, center on emotional relief and work-life balance improvements.
00:06:29 --> 00:06:34 For each approach, provide a headline, five to eight words, a sub-headline,
00:06:34 --> 00:06:40 10 to 15 words, the psychological appeal of this approach, who it might resonate
00:06:40 --> 00:06:43 with most strongly, and potential drawbacks.
00:06:43 --> 00:06:48 And what this prompt does, it gives you multiple viable options to consider
00:06:48 --> 00:06:50 rather than a single solution.
00:06:50 --> 00:06:56 Plus, you get the analysis of strengths and weaknesses to help you make more informed decisions.
00:06:56 --> 00:07:00 Let's look at another variation of this multi-perspective prompting.
00:07:01 --> 00:07:05 Let's say we need to analyze a problem from different stakeholder viewpoints.
00:07:05 --> 00:07:07 So the prompt will go something like this.
00:07:08 --> 00:07:11 Analyze our proposed website navigation from three perspectives.
00:07:11 --> 00:07:15 One, a new visitor trying to understand what we offer.
00:07:15 --> 00:07:20 Two, a potential customer evaluating if our solution fits their needs.
00:07:20 --> 00:07:24 Three, an existing customer looking for support resources.
00:07:24 --> 00:07:29 For each perspective, identify what works well, potential points of confusion,
00:07:30 --> 00:07:32 and information that might be missing.
00:07:32 --> 00:07:35 And essentially, this helps uncover
00:07:35 --> 00:07:40 blind spots that might be missed when considering only one viewpoint.
00:07:40 --> 00:07:45 But to use this effectively, you need to be specific about different approaches you want explored.
00:07:46 --> 00:07:51 Don't just ask for different ideas, quote-unquote, but provide it with clear
00:07:51 --> 00:07:53 alternatives covering a range of possibilities.
00:07:53 --> 00:07:58 The more distinct the approach, the more valuable the comparison will be.
00:07:59 --> 00:08:04 Let's talk about another technique in prompting, which is constraint-based prompting.
00:08:04 --> 00:08:08 You know, the thing is, sometimes the most creative results come from adding
00:08:08 --> 00:08:13 deliberate constraints. So this essentially is the principle behind constraint-based prompting.
00:08:13 --> 00:08:18 For example, think about how jazz musicians improvise with musical constraints,
00:08:18 --> 00:08:22 or how poets create with strict formats, like sonnets.
00:08:22 --> 00:08:26 Constraints often spark creativity rather than limiting it, right?
00:08:26 --> 00:08:31 Instead of saying something like, write a product description for my handcrafted soap...
00:08:31 --> 00:08:36 You can try. Write a product description for my handcrafted lavender soap with these constraints.
00:08:37 --> 00:08:44 Cannot use luxury, premium, artisanal, or handcrafted. Must tell a micro story in exactly 50 words.
00:08:45 --> 00:08:48 No adjectives ending in -ing or -ly.
00:08:49 --> 00:08:53 Must include a sensory description beyond scent. Should evoke dawn.
00:08:54 --> 00:08:58 Cannot mention ingredients. Must end with a two-word sentence.
00:08:58 --> 00:09:04 And you see how we listed a lot of constraints, and it basically forces AI to
00:09:04 --> 00:09:09 get creative rather than falling back on its predictable patterns or phrases.
00:09:10 --> 00:09:15 By restricting the easy options, you push it to find less obvious but more interesting approaches.
00:09:15 --> 00:09:20 I used this technique when I wanted to create captions for my client's Instagram
00:09:20 --> 00:09:25 content. And I noticed that basically we were repeating much of the same stuff,
00:09:26 --> 00:09:28 but in exact same wording, month to month.
00:09:29 --> 00:09:33 And by adding constraints like, can't use this and this word,
00:09:33 --> 00:09:39 or you must reference an unconventional thought process, we generated content
00:09:39 --> 00:09:41 that stood out from her previous posts.
00:09:41 --> 00:09:46 Another related technique is reverse prompting. It's describing the output you
00:09:46 --> 00:09:50 want and then asking what prompt would create it.
00:09:50 --> 00:09:56 This is a very powerful technique that I believe most of you are already familiar with.
00:09:56 --> 00:10:01 So the initial prompt is something like, I want social media content about sustainability
00:10:01 --> 00:10:07 that avoids sounding preachy, doesn't rely on guilt, and feels practical rather than idealistic.
00:10:08 --> 00:10:12 It should inspire small actions without making people feel bad about current
00:10:12 --> 00:10:15 habits. What prompt would generate this content?
00:10:16 --> 00:10:22 And then the AI will spill out a detailed prompt you can use or modify, obviously.
00:10:22 --> 00:10:25 But it's kind of like having a prompt engineer on demand.
00:10:25 --> 00:10:31 I actually discovered this myself about a year ago or so because I was struggling
00:10:31 --> 00:10:34 to create a prompt for image generation tools.
00:10:34 --> 00:10:39 I knew what I wanted in my head, but I found it hard to express,
00:10:39 --> 00:10:44 you know, all the details that I thought would be beneficial for the AI to produce
00:10:44 --> 00:10:47 the result or the image I wanted, right?
00:10:47 --> 00:10:55 And so I asked Claude, I need a prompt for creating an illustration for this
00:10:55 --> 00:10:57 and this brand and this and this tones, blah, blah, blah.
00:10:57 --> 00:11:03 And then I would ask, could you please create a prompt for me for the Bing image
00:11:03 --> 00:11:05 creator so that I get the best possible outcome?
00:11:06 --> 00:11:11 And what it really came up with was something I would have never thought on my own.
00:11:12 --> 00:11:17 So we touched on basic role prompting in our last episode, but let's explore
00:11:17 --> 00:11:21 the advanced version, which is immersive role-playing.
00:11:21 --> 00:11:28 This means we create a rich, detailed persona for the AI to embody throughout an interaction.
00:11:28 --> 00:11:34 Instead of just saying things like, take on a role of an experienced brand strategist,
00:11:34 --> 00:11:39 or take on a role of an expert copywriter? We would try.
00:11:40 --> 00:11:45 For this conversation, you'll embody Sarah Chen, a brand strategist with 15
00:11:45 --> 00:11:48 years of experience in luxury lifestyle brands.
00:11:48 --> 00:11:54 Sarah trained at Parsons, spent eight years at Lander, and now runs her own boutique agency.
00:11:54 --> 00:11:58 She's known for her direct feedback that cuts through fluff,
00:11:58 --> 00:12:03 deep understanding of luxury marketing psychology, and ability to identify a
00:12:03 --> 00:12:04 brand's emotional core.
00:12:04 --> 00:12:09 Her philosophy is one brand, one feeling, multiple expressions.
00:12:09 --> 00:12:13 She's skeptical of trends and focuses on timeless positioning.
00:12:13 --> 00:12:18 As Sarah, review my brand brief, identifying core strengths and weaknesses.
00:12:19 --> 00:12:23 And this detailed persona gives the AI a rich character to inhibit,
00:12:24 --> 00:12:31 which ultimately results in more nuanced and more valuable feedback than asking
00:12:31 --> 00:12:35 it to just be like a generic brand strategist, right?
00:12:35 --> 00:12:40 You can also use this in, let's say you have a client who is preparing for a
00:12:40 --> 00:12:41 difficult presentation.
00:12:41 --> 00:12:47 And you can create detailed personas representing tough audience members they
00:12:47 --> 00:12:51 might face, and then practice responding to challenging questions from them.
00:12:52 --> 00:12:55 And another related technique is dialogic prompting.
00:12:55 --> 00:12:59 It's basically creating a conversation between multiple experts to explore a
00:12:59 --> 00:13:02 problem from different angles. Here's how the prompt would go.
00:13:02 --> 00:13:07 Create a dialogue between three experts discussing our rebranding project.
00:13:07 --> 00:13:12 Expert number one, a traditional brand strategist with 20 years at Fortune 500
00:13:12 --> 00:13:14 companies who believes in evolutionary change.
00:13:15 --> 00:13:17 Expert two, a disruptive DTC founder
00:13:17 --> 00:13:21 who believes traditional branding is outdated in the social media era.
00:13:22 --> 00:13:28 Expert three, a consumer psychologist focused on how branding affects customer decision making.
00:13:28 --> 00:13:33 Have them discuss the pros and cons of a complete brand overhaul versus a subtle
00:13:33 --> 00:13:39 refresh. And what this would do, it would generate a rich exploration of viewpoints
00:13:39 --> 00:13:42 rather than a single recommendation from one person.
00:13:42 --> 00:13:46 This would be valuable for some complex strategic questions where reasonable
00:13:46 --> 00:13:48 experts might disagree.
00:13:49 --> 00:13:55 So let's pull these techniques together with a practical demonstration of a case study project.
00:13:55 --> 00:14:00 Let's say you are working with a client who runs outdoor leadership retreats,
00:14:00 --> 00:14:03 and they struggle to articulate their value proposition in a crowded market.
00:14:04 --> 00:14:08 If you go with the basic prompting, you'll get results that don't capture anything
00:14:08 --> 00:14:10 that really makes them special, right?
00:14:10 --> 00:14:15 So here's how you can approach this by using advanced techniques.
00:14:15 --> 00:14:19 First, we can use chain of thought prompting to explore the problem.
00:14:19 --> 00:14:21 Before developing value proposition
00:14:21 --> 00:14:25 options for these outdoor leadership retreats, think through 1.
00:14:25 --> 00:14:29 Analyze the market to identify common approaches and gaps. 2.
00:14:30 --> 00:14:34 Examine psychological benefits of wilderness for leadership development. 3.
00:14:34 --> 00:14:39 Consider how their approach, solo experiences plus group reflection,
00:14:39 --> 00:14:41 differs from typical team building.
00:14:42 --> 00:14:47 4. Identify the transformation participants actually experience versus expectations.
00:14:48 --> 00:14:55 5. Explore different frames for communicating value, professional advancement, personal growth, etc.
00:14:55 --> 00:14:57 Show your thinking for each step.
00:14:57 --> 00:15:05 And what this would do is it would give you a deeper analysis rather than simply asking for ideas.
00:15:05 --> 00:15:10 If you're using ChatGPT for this, you can also turn on the deep research button,
00:15:10 --> 00:15:15 but it will just take more time for the AI to work through and spill out the answers for you.
00:15:15 --> 00:15:19 You can try both ways and see the difference in results.
00:15:19 --> 00:15:23 And then when you get the answer, the next prompt would be a comparative prompt
00:15:23 --> 00:15:24 for multiple directions.
00:15:25 --> 00:15:28 You'd say something like, develop three different approaches.
00:15:28 --> 00:15:32 Approach number one, focus on professional outcomes, career achievements.
00:15:32 --> 00:15:37 Approach two, focus on personal transformations, who participants become.
00:15:37 --> 00:15:42 Approach three, focus on unique methodology, how it differs from conventional training.
00:15:42 --> 00:15:47 For each, provide a headline, supporting paragraph, target audience,
00:15:47 --> 00:15:49 and potential objections.
00:15:49 --> 00:15:56 And what this would give you is three distinct directions with analysis of strengths and weaknesses.
00:15:56 --> 00:15:59 And then we would use the constraint-based prompting.
00:15:59 --> 00:16:03 So here's the prompt. Take the personal transformation approach and refine it
00:16:03 --> 00:16:04 with these constraints.
00:16:05 --> 00:16:10 Cannot use leadership, transformation, wilderness, or retreat words.
00:16:10 --> 00:16:13 Must connect inner journey to outer results.
00:16:14 --> 00:16:19 Maximum eight words for headline. No superlatives, best, most, etc.
00:16:19 --> 00:16:23 No business jargon, understandable by a 12-year-old.
00:16:23 --> 00:16:27 So these constraints will push you away from predictable corporate language
00:16:27 --> 00:16:31 towards something distinctive. And finally, we can test out the role-playing
00:16:31 --> 00:16:33 prompt to test our value proposition.
00:16:34 --> 00:16:38 Let's say you choose one headline that AI gave you. Let's say it was,
00:16:38 --> 00:16:41 find your center lead from strength.
00:16:41 --> 00:16:45 That was the first headline that AI gave you. And so the role-playing prompt
00:16:45 --> 00:16:50 would be something like, evaluate, find your center lead from strength from these perspectives.
00:16:51 --> 00:16:54 One, skeptical Fortune 500 HR director.
00:16:55 --> 00:17:01 Two, data-driven tech executive. 3. Potential participant interested in wilderness
00:17:01 --> 00:17:04 but unsure about quote-unquote leadership. 4.
00:17:05 --> 00:17:10 Past participants six months later. Identify how each would receive it,
00:17:10 --> 00:17:13 what objections might arise, and how to address them.
00:17:13 --> 00:17:18 And this will definitely reveal weaknesses you could fix in supporting messages.
00:17:18 --> 00:17:23 So the final value proposition will be significantly stronger than with basic
00:17:23 --> 00:17:27 prompting alone. So the combination of techniques was powerful.
00:17:27 --> 00:17:31 Because chain of thought prompting forced deeper analysis,
00:17:31 --> 00:17:37 then comparative prompting gave multiple strong options, then constraints pushed
00:17:37 --> 00:17:39 beyond predictable language that
00:17:39 --> 00:17:44 is widely used, and then role-playing identified potential weaknesses.
00:17:46 --> 00:17:49 Let's talk about integrating these techniques into your workflow.
00:17:50 --> 00:17:54 First, identify where advanced prompting would add the most value.
00:17:54 --> 00:17:58 Let's say you have a creative block. Well, definitely, the constraint-based
00:17:58 --> 00:18:01 prompting will break through your mental ruts.
00:18:01 --> 00:18:07 And then when you're evaluating concepts, multi-perspective approaches will find the blind spots.
00:18:07 --> 00:18:11 And then when you need to go beyond the obvious solutions, chain of thought
00:18:11 --> 00:18:14 prompting pushes past predictable ideas.
00:18:14 --> 00:18:19 So essentially, these techniques are your collaborative tools and not a replacement
00:18:19 --> 00:18:21 for your expertise, because you
00:18:21 --> 00:18:26 still need your expertise to actually know how to direct the AI, right?
00:18:26 --> 00:18:30 You need to know what you're looking for. That's the difference between people
00:18:30 --> 00:18:36 who use AI but don't have the expertise in their particular field and you who
00:18:36 --> 00:18:42 do have the expertise and you just know how to prompt it to get the most unique possible results.
00:18:42 --> 00:18:46 Think of it like jazz improvisation. DAI would provide variations,
00:18:46 --> 00:18:50 but then you select, refine, and assemble them.
00:18:50 --> 00:18:55 And ultimately, what it comes down to is that your creative judgment still remains essential.
00:18:55 --> 00:18:59 I recommend creating a personal library of these advanced prompts.
00:18:59 --> 00:19:03 I'm going to leave the link in the show notes for these advanced prompting techniques
00:19:03 --> 00:19:09 and the examples I offered today so that you can easily copy and paste into
00:19:09 --> 00:19:10 your own prompt library.
00:19:10 --> 00:19:15 You can also keep them organized by specific challenges. Let's say you have
00:19:15 --> 00:19:20 prompts for creating blocks, prompts for pushing concepts further,
00:19:20 --> 00:19:23 prompts for evaluating different perspectives.
00:19:23 --> 00:19:26 It'll help you grab the right prompt when needed instead of,
00:19:26 --> 00:19:30 you know, just browsing through a document filled with all different prompts
00:19:30 --> 00:19:32 you gather from different sources.
00:19:32 --> 00:19:36 This advanced technique definitely takes practice. The first few times might
00:19:36 --> 00:19:41 feel a bit awkward or overly complex, but the more you use them,
00:19:41 --> 00:19:48 the more you'll figure out which ones bring the best results for your specific creative challenges.
00:19:49 --> 00:19:55 So let's recap what we covered today. We explored a few advanced prompting techniques
00:19:55 --> 00:19:59 that can really take your AI collaborations to the next level.
00:19:59 --> 00:20:03 Chain of thought prompting, which reveals reasoning behind creative directions.
00:20:03 --> 00:20:08 Than comparative approaches, which explore multiple directions simultaneously.
00:20:08 --> 00:20:14 Then there was constraint-based prompting, which forces more creative,
00:20:14 --> 00:20:15 unpredictable solutions.
00:20:16 --> 00:20:21 Then we had reverse prompting, which helps us craft better prompts for specific outputs.
00:20:21 --> 00:20:26 And then specific role-playing techniques that generate rich, nuanced perspectives.
00:20:27 --> 00:20:32 And remember, these techniques build on those basic principles we talked about
00:20:32 --> 00:20:37 in the last episode. The specificity, context, format, examples.
00:20:37 --> 00:20:40 Role prompting, iteration, and system prompting.
00:20:40 --> 00:20:45 These advanced techniques just take things up a notch.
00:20:45 --> 00:20:49 As these AI tools become more widespread, using them in these sophisticated
00:20:49 --> 00:20:53 ways is an increasingly valuable skill, in my opinion.
00:20:53 --> 00:20:58 The difference between basic and advanced prompting often separates generic
00:20:58 --> 00:21:01 outputs from truly unique and innovative work.
00:21:01 --> 00:21:06 So for this week, I would encourage you to try to experiment with one advanced
00:21:06 --> 00:21:09 technique on a current project you're working on.
00:21:09 --> 00:21:14 And then compare the results that the basic prompt would give you versus the advanced one.
00:21:14 --> 00:21:19 And then create a template for techniques that prove valuable for your own work.
00:21:20 --> 00:21:25 And then share your experiences in Facebook groups or on Instagram or wherever
00:21:25 --> 00:21:28 you're active with other creative professionals.
00:21:28 --> 00:21:33 Next week, we'll explore AI image generation tools with specific techniques
00:21:33 --> 00:21:39 you can use for ChatGPT image creator, MidJourney, and other visual AI systems.
00:21:39 --> 00:21:44 If you found this episode helpful, please subscribe and share with other creative professionals.
00:21:44 --> 00:21:49 You can find complete prompt templates and examples in the show notes of this episode.
00:21:49 --> 00:21:53 I'm Tiana, and I'll be back next week with more practical guidance on leveraging
00:21:53 --> 00:21:55 AI in your creative work.
00:21:55 --> 00:22:17