Using AI in Human-First Business — Reflections, Crowdsourced Use Cases, Research | Episode 90

Hello hello and welcome back to the Purpose & Progress Podcast.

This is an episode about AI. Because can you even talk about messaging, marketing, and especially human-first business without AI entering the conversation. (Or the Zoom room, right?)

And disclaimer, this is a VERY long one, so I’ve timestamped each main section in the shownotes in case you care more about 1 category of idea than the other!

Now, I’m a self-proclaimed complete noob when it comes to pretty much anything AI. I just don’t see the hype to be honest. The learning curve hasn’t outweighed the benefits for me yet.

And I’m not alone in having a limited experience. When I polled people on Instagram a few months back, 12% said they literally never use AI, 56% said they use it occasionally, 20% said they use it kind of a lot, and 12% said they use it daily.

But I’ve been having the most interesting conversations with values-aligned people who love and use AI, sometimes as a catch-all, sometimes in really specific cases. I share ALL their ideas later in this episode.

So with this episode, I’m going to give my general thoughts about AI, then I’m going to share some crowdsourced use cases to give you ideas for how to get the most mileage out of AI while still running a very human-first business, then some AI research I’ve been reading. Like I said, I’ll timestamp each section in the shownotes in case you care more about 1 category of idea than the other!

My Thoughts

So, starting with my main thoughts on AI. I believe that the best brands, people, foods, anything are nuanced. And nuanced doesn’t necessarily mean complicated. Or fancy. Or unrelatable.

It just means that there are layers, and that not everything is seen and felt and understood at first glance.

And more than anything, there’s intention.

While sort-of-useful free masterclasses for thousands of people or a list of somewhat insightful ChatGPT prompts can get you started, they’re really not enough for you to make anything special. Or memorable. Or true-to-you and your perfect-fit people.

I’m not squarely against AI (and I fully realize it’s here to stay), but I’ve rarely used it for myself. I’ve sort of tried it for shownotes and episode titles, but mostly I just write from my notes and memory and gut. I’ve sort of tried to be in conversation with it to brainstorm ideas, but I’ve found the responses to be pretty generic and that they repeat what I’ve already seen and heard from years of consuming online business content.

I truly believe that AI just isn’t nuanced enough to replace humans yet. And I stand by that.

In general, all this information is readily available from every direction, and it’s even easier to access at a glance (theoretically…AI hallucinations aside). So in this new era, people are looking for connection. And empathy. And transformation. And the very human things that you can provide that their run-of-the-mill or even hyper-customized AI tools can’t offer.

But just for kicks and for the sake of my newsletter (always linked in the show notes or at ashleesang.com/newsletter), I prompted ChatGPT the following:

“Summarize these testimonials about my values-aligned messaging and marketing consulting work into a few overarching and recurring themes, noting big-picture ideas, results, and experiences.”

Then I dropped in a selection of the testimonials I’ve collected over the years—I have them organized in alphabetical order in Notion, so it was essentially randomized for offer type, client’s business type, how long ago we worked together, etc.

And what it spit out was actually pretty good. And a bit surprising. But also not. I have the blog post version of that newsletter with the actual results linked in the show notes if you’re super curious.

But the results were surprising because it actually felt pretty spot-on. (Although I’d never say something like “drives meaningful results” in my copy—a bit too generic/says something without actually saying anything and just not quite my voice.)

But it’s not that surprising that the summary was pretty on-point because a lot of my clients have the same type of breakthroughs and speak the same language.

Makes sense because I’m the same practitioner with the same POV attracting people with the same types of problems and aspirations. (To an extent. Remember from the top of this episode: people and time and space are nuanced.)

It also makes sense that the big themes came together seamlessly because regardless of who I work with or the exact container, we’re always building a brand together—a shared experience and alignment with the central question, “How do you want to be known?” (I actually have a whole podcast episode about this question, number 37, linked in the show notes.)

Also, again, AI can only regurgitate what already exists, based on what it’s prompted. So while I gave it a random sample of past testimonials, I had curated the questions I ask clients and hand-selected the feedback that lit me up the most. So it’s a bit of a self-fulfilling prophecy loop.

So all that to say, I’m confident AI can only get us so far. And least now, as of mid-2025. But honestly, I’m fairly certain humans will always rely on humans. That’s the essence of what makes us human.

Ok, but when we ARE using AI in your business planning or marketing, I think it’s super important to keep in mind that ChatGPT (or Claude AI or your platform of choice) just doesn’t do nuance well enough. Not yet at least. And that it’s the prompts that matter most.

So when I work with deep-thinking, multi-passionate people and their most common feedback to me is, “I’ve never thought of that before,” I know that I’m not competing with AI to produce the MOST amount of ideas or copy or content.

Because I hold space and offer perspective and empathetically see and imagine my clients’ visions and values, often in ways they don’t even experience themselves. And definitely in ways they don’t feel equipped to articulate by themselves.

You can’t prompt your way to a great output if you don’t have the expansive mindset to imagine what could be or what’s missing or to see what you’re just too close to see.

So yes, you can totally use ChatGPT as a baseline “thought partner” or idea organizer. It’s good enough for generic, for sure.

But please don’t stop at generic.

Bring those ideas to a human. Someone with the same (or strategically different) point of view as you.

Or do the inverse: collaborate with a human to do the heavy mental and emotional lifting (like what IS your brand voice, your brand values, your ideal audience, your lived experience), then use ChatGPT to refine and repurpose. Essentially, have it stretch all the deeply human thinking and creating you’ve done.

Here’s an example. I was on a Quick Win Consulting Call once (I’ll link the sign-up below if you’re curious) with a client who was using ChatGPT as a 3rd “brain” on our call.

We were working on her ideal customer avatar and her website copy. She was dropping ideas into ChatGPT in real time. She flagged what she liked, and I was translating it to human-speak and injecting heart/empathy/her own flair as we went.

And here’s another example of AI doing some legwork, but you needing to use your own life experience and human eye to refine:

My client and long-time business friend Erika posted about this 2024 LinkedIn Rewind and I thought it was such a fun idea. (You might have seen these posts floating around, especially if you hang out in the halls of LinkedIn. I tried to link it in the show notes, but it’s just a redirect now. I did link my interview with Erika though, all about meaningful community and collaboration.)

Anyway, while that AI summary of my online presence last year missed the mark in some ways, it was sort of insightful to see what public perception might be of me. Just like human outside perspective is SO valuable, AI-generated overviews like this can be useful to get us out of the weeds and take a look at our overarching vibe, vocabulary, and themes. It can’t capture the nuance, but it can be a useful gauge of how well our intention matches with our perception. Which, as I’ve talked about before is the essence of a brand experience.

Crowdsourced Use Cases

Ok, so that’s my limited experience with and hot (maybe just warm) take on AI’s role in human-first business and marketing. Now onto the use cases if you WANT to use it more in your day-to-day.

I was in conversation with lots of smart, ambitious, values-driven business owners about how they use AI. And here’s what they said:

A graphic designer and community manager said he uses AI for “ideas and some IG captions. I never copy and paste it exactly, but sometimes I need the brainstorming as I’m a terrible writer.” This is one of the most common uses I’ve seen, especially if you aren’t confident in your ability to write.

A community leader and photographer said she uses it for “taglines. They seem to be one of my weaker points. I’ve also used it for email subject lines, event landing pages, and as a jumping off point for email marketing. And I’ve have it script my VSL for a landing page.” (VSL is a video sales letter.) So she’s filling in the gaps for what’s hard to get started for her and to pull together ideas she’s already thought through, like elements of an event or offer. She also said she was a beta tester for Bard and was an early adopter, but even with being a long-time and frequent user, she “is very careful with the way I use it because I think it can be very obvious.”

A coach and hypnotherapist said she uses to as a thought partner, “I’ll brain dump all my thoughts into it and get it to spit out a more organized version of what’s in my head.” This is personally one of the most appealing use cases for me because I’m definitely a million and one ideas at all times person and I have 600 things on my to do list at any given time, so having an external fake brain be my executive function could probably be really supportive to me.

A copywriter and sales expert said, “I’m creating something mostly to keep up with trends, not because I believe it’s the best or even good. I was in the never use camp until last week. But I’ll use it as an alternative for Google because it gives a better word association than a thesaurus.” So I thought this was really interesting to meet people where they are and then position yourself—a copywriter in this case—as an even better alternative. Also, I hadn’t thought to use AI for word replacement, but I’m an OG thesaurus.com user and have also used relatedwords.io, which I think was an AI precursor.

A comms director said “I chat with ChatGPT a lot. I use it to troubleshoot, bounce ideas around, help me structure copy. As a one-person department, it’s nice to have that instant back and forth conversation.” As a verbal processor, I can totally see that. Plus, the advantage of being in conversation with anyone or using any tool is that even if it spits out something you DON’T like, that could point you toward what you DO want. Sometimes those gut checks are useful even if they’re in the negative or backward direction.

A web designer said “I use ChatGPT to help me get information out of transcripts or lots of data. For example, I asked about business struggles in a Facebook group and got dozens of replies. I copied all the answers and put them into ChatGPT and it organized the data so I could see the patterns in the replies.” This reminds me of how Seth Godin advocates for AI to be used—a shortcut for what humans could do, but maybe not as well and with more error. And then I love the nuance of HER picking up on patterns herself. I will say, I’ve kind of tried to get AI to sum things up for me in the past—maybe data, maybe information, can’t remember—and it said I was dumping in too much information, so I find the limits to be frustrating.

A marketing strategist said “I use Claude the most, mainly because I write a lot. It’s a great brainstormer, or gets me going when I’m staring at a blank screen. I use Perplexity for research. It’s a tool. I don’t look at it as the enemy. I took AI for Anyone—check out The Human Stack with Tim Lockie. He’s doing great work in the social impact sector.” I’ve also heard that Claude is a “better writer” and the Perplexity is great because it lists its sources (so it’s theoretically less likely to have the AI hallucinations, which is when AI just makes things up). I think part of the reason I don’t use AI for brainstorming like so many people do is that having an idea is never the issue for me. I have TOO many ideas, and then needing to edit something that AI spits out would be a bigger burden then just working through my thoughts in the first place. But the more organizing element vs brainstorming element is still something I could consider. Anyway, I have Tim’s AI for Anyone course linked in the shownotes. Can’t vouch for it personally—also sort of looks and sounds like they used AI to create the website, but don’t know that for sure.

A community builder said “If I feel like an email is too clunky, I ask it to check for grammar and readability. Usually there’s only a could tweaks.” I think the readability is actually a useful point and obviously thinks like spell check and Hemingway and other tools have existed for a long time for exactly this reason.

A DEI consultant said, “I used it for writing a follow-up email to an old lead! I love it for emails when I can’t quite figure out what I want to say in response or for outreach but never for my captions or anything like that. I don’t use it for copy because it sucks. lol I sometimes use it or outlines too. Another cool non-business thing it can do too is help with meal planning and I’ve used it for utilizing the ingredients I have at home to make a meal I wouldn’t have thought of.” I thought it was really interesting that she drew the line for email copy/content, but not more public copy and content. It seems the most important part of AI use is where those lines are for you. And then I love the day-to-day life application too. The first time I heard of this was a mom YouTube video I watch and she included the meal planning among other ideas—I’ll link it for all her ideas.

A podcast manager who was in the ”daily” camp of my poll said “The main way is to fix grammar and capitalize titles. I write a lot of content for clients. It can save some time, but I don’t think it’s that life-changing. I feel like most people use it for content creation and that’s just…well it sounds so AI.” Really interesting to me that she’s a daily user, but still very skeptical and aware of its limitations.

A web designer said she uses it for “Brainstorming ideas for blog posts in my business. For clients, I use it to help with little tasks like image names and alt text—things to speed up my process that don’t take a lot of editing. I try to balance keywords and an actual visual description.” I actually think on the SEO front, it could probably be a really useful tool. SEO is one of those things that actually works pretty well in my business and if I doubled down on it, it’d probably bring me even more leads. I sort of did a set it and forget it method, but I could 100% be optimizing these podcast episodes, my blog posts/newsletter topics, etc based on what people are actually searching for. Note to future self… And then the alt text use case is brilliant in my opinion, because that’s mindless, descriptive, non-creative copy that you really don’t need too much of a human eye or heavy hand on. But that alt text can be really supportive for both SEO and accessibility goals. I’ll link a specific alt text generator that a different marketing friend recommended.

A retail and strategy leader said “Usually just ChatGPT to edit things I wrote. But today I used Gamma to create a slide deck. It turned out ok…I still need to edit but I think it’s more of a learning curve as to how to format the text I’m putting in.” That learning curve is my main barrier—I had never even heard of Gamma to be honest, but I know tools I use every day like Notion and Canva both has AI features and I just haven’t put in the time and effort to learn how they could save me time and effort.

A business astrologist said “I use it to get very clear when I write something for my intended audience. I guess you could say I use it to “niche”. I also use it to help me whip up some ideas for highly clickable subject lines, especially if nothing is flowing to me. I’m not one to manipulate my readers per se, I just want to make sure my emails are always opened.” I think that efficacy vs empathy spectrum is at the heart of the ethical marketing question.

A social media strategist said “Generally for content brainstorming when I’m feeling a bit stuck or low on creative energy for a specific post or project. I give it really detailed instructions (including brand voice, target audience, business/marketing/CTA goals, samples of brand wording, etc) so it generates some really decent proposals which I then adapt quite a bit further and make more human and even more personalized. It’s definitely better at brainstorming and critiquing/analyzing than writing. It should remain a tool we use occasionally and with intention, not a crutch/shortcut we become overly reliant on. Still slightly terrifying to see how powerful it is.” I really liked the nuance that it’s PART of her process, not her entire process. Also interesting to me that she fears it and still sees a place for it in her business. And then as a values-aligned brand messaging strategist, I would be remiss if I didn’t flag that she includes some of the most essential brand messaging elements in the prompt so that it can sound as much like her client as possible. I have an episode about the elements of an impactful brand, number 68, linked below, and I’ll link my Messages That Matter Intensive in the show notes below if you want to learn more about how I do values-aligned brand messaging strategy with multi-passionate deep thinkers.

A content creator said “I use it for everything, even my website copy. I write a draft and then I edit just for polish and tone. It’s never 100% perfect, but it always comes up a little better than my draft.” So this is like the example I gave at the top of the episode, either use it first and edit next, or write first and use it to tune-up. Either works, depending on your own strengths.

An editor said “When I get stuck, I use it for idea generation. Not gonna lie, I’ve also used it for therapy. An actual therapist is way better, but sometimes if you’re specific enough, the advice can be decent, especially if I clarify what angle of response/practice I want.” She also said it can be a good stop gap in between sessions or when you’re between therapists because of insurance issues, etc. This is the most human use case I’ve heard of I think, and yet, sometimes when we just want something mirrored back to us or we need a reframe so that our own brains and emotions can run with it, I could see this being really supportive.

A marketing strategist said “I use it for everything. ADHD business owner’s best friend. I use it for repurposing, outlining, creating show notes for podcast episodes, so many things.” I just watched a video by Louise Henry about ways to repurpose using AI—he content is always super short and sweet, I’ll link it. And then I’ve seen many times over how AI is super supportive to the ADHD brain. I’ll link a Reel I watched with ADHD-prompts that help with executive function, like “Break this overwhelming task into tiny manageable tasks for me.” And I’m sure there are 100 more around the interwebs. I see Threads with prompt ideas daily, both for ADHDers and in general. I’ll link a Thread about 8 underused prompts that will improve your outputs based on style, POV, audience, etc. I found these ideas to be especially useful for marketing and positioning purposes.

A communications researcher said she uses Boardy — I’ll have it linked — to grow her network and act as a referral partner and warm lead generator. I think you physically call in and it connects you? You’ll have to look into it, but she thoroughly enjoyed it.

Sarah Cook is a fellow Wash U alumna and a super intentional copy and content writer. She created Brand Vibe AI Starter Pack (affiliate link in the show notes) with the intention of turning ChatGPT into your brand copywriter in one afternoon. It’s basically a course that walks you through how to create your custom Chat GPT. I’ve started it but personally REALLY struggle with online course formats (hello, ADHD brain), so I haven’t actually set mine up yet. But I’m curious. And while there are environmental concerns (in addition to the ethical ones) around AI, Sarah is putting 1% of your Brand Vibe AI purchase toward carbon offsets and water restoration projects through Terrapass. Love that she’s thoughtfully engaging in how to make your personal energy as a business owner more sustainable while thinking of our role in a bigger sustainability movement.

Thought Pieces Around AI Use

And actually, speaking of environmental concerns, definitely do your own research because they’re very real and might be significant enough to put you squarely in the do not use camp. I’ll link an article by one of my favorite environmental content creators.

There are about a bajillion (made that up—maybe AI can fact check it for me) articles on AI use, so I’ll just reference a few here to cover some overarching thoughts.

First, I feel like one of the biggest and funniest questions is whether or not we need to be polite to AI. Most people in my orbit seem to be, I think because I surround myself with super considerate humans. But I’ve erred on the side of purposefully NOT using please and thank you because it’s decidedly not human in my brain and I don’t want to be conflating anything subconsciously. I’ll link a Tech Radar article, but the gist is that it’s not necessary to use manners, but AI does mirror your style, so it might affect your experience depending on how polite you are.

Another idea I want to unpack a little is from a techxplore article about the differences between human and AI-generated text. A client sent it to me.

Some standout ideas from it for me were: “I care that my doctor’s notes are accurate. I don’t really care if they’re in the voice of my doctor. But if I’m writing a job application letter where I want to stand out, that matters a great deal. As instructors, writers and communicators, we need to be aware of LLMs’ idiosyncrasies and shortcomings.”

And then along those lines, Gaby from way back in episode 18 reposted about the use of AI in education. I’ll have Gaby’s interview and the LinkedIn post linked in the show notes. In this video, the speaker basically said to assume students are using AI, then adapt how we’re testing for knowledge and encouraging critical thinking. She likened the integration of AI to the introduction of the calculator in math. I’ve also seen the argument that Photoshop doesn’t automatically replace good photographers and designers. That techxplore article actually says LLMs aren’t quite like a calculator in that a calculator can’t (often) give completely fake answers like AI can but I thought the analogy was useful.

Bringing these ideas back to the world of human-first business, we want every touchpoint to feel like a human experience. Or I guess that’s for you to decide. But I’m personally all for injecting as much personality and a glimpse or a healthy dose of your values in everything you do, say, write, create. It makes you more trustworthy, memorable, and referable.

Ok, another AI resource that came across my radar is Google’s Career Dreamer, which is a free AI tool that supposedly accurately and playfully finds your dream job based on your skills and interests. It suggests career paths and assists with resumes and cover letters using Gemini AI. This caught my eye because job searching is a lot of capturing your essence and positioning yourself, so that’s messaging 101. I’ve also worked with a handful of job seekers recently with their online presence, career pivots, CVs, etc via my public About You Audits and my private 1-on-1 audits. The next public one is in August. I always share via my newsletter and I hope to have a podcast episode about it ahead of time. I have the private audits linked below if you don’t want to wait for my 2 cents!

And then like so many of the people I quoted above said, AI often sounds like AI. Writing and marketing pro Ann Handley often gives commentary about its use in our writing. She beautifully explained in her Total Annarchy newsletter—love a good pun, linked below: “Generative AI burps out alllll the content we could possibly need. Our differentiator is only this: our taste, perspectives, experiences, and the ability to tell true stories with a bit of inspired joy.”

In a separate newsletter, she wrote a related idea: “Your FROM Line matters more than your SUBJECT Line. This is the metaphor for Marketing in 2024. Our direct relationships with people (customers, prospects) matter more than ever. It means: How do we sign our work? Our warm, relatable human voices need to articulate who we are clearly and consistently. It means we need to inspire people to seek us out vs. hope/pray/pay Google to surface us for them.”

Ann is one of the most famous marketers out there (love than she’s a woman in a sea of male examples!). In #184 of her newsletter, she talked about the AI heist as she called it. Basically, how AI creators are committing IP piracy, grabbing any and every piece of thought leadership they can get their little bot hands on, often without crediting and almost never compensating the original creators that are doing all the heavy lifting. She wrote “They actively chose to steal the books, instead of going through the proper and legal channels, because the legal way was too slow.”

And last quote from her because while I said earlier in this episode that you can use AI in any point in your process, she makes a strong case for a specific order: “We have it backwards. The first draft is the thinking draft. That’s where you need to be fully present. On board. Just you and your glorious ideas. Otherwise, you undermine yourself. You shortchange your growth and creativity. You put the ROBOTs first. You put yourself second. The first draft is the thinking draft. (Said twice for emphasis.) Protect the part of you that truly makes your content, your writing, your voice your own. Why? Because otherwise and over time, your skills will atrophy. You will doubt yourself more. You will build a dependency.”

I also mentioned prolific writer and marketer Seth Godin. I don’t have any specific references for this episode, but I recommend his daily newsletter called Seth’s Blog. I’ll have it linked in the show notes. Yes, it’s daily, but it’s almost always super short and digestible and empathetic.

Wrap Up

All that to say: I’m not ALWAYS anti-AI. You might even try an exercise like the testimonial summary I did if you need a confidence boost, want to synthesize something more quickly, or just need to get outside of your own head and heart for a second.

And bonus, slightly unrelated tip because I touched out it briefly: capture testimonials! Use an official form, or just pay attention when you get organic feedback. (Some of my most moving feedback is totally off the cuff, like in a comment of a working doc during a month of Messaging Momentum or at the end of the reveal call for a messaging strategy intensive. Both those containers are linked in the show notes if you want to see how I work with clients.)

I’ll reiterate though, no matter how you use AI, make sure you run whatever results you get past a living, breathing, feeling human. Whether that’s your own brain and heart and values, or someone who can give you outside perspective.

Also, absolutely input your voice and values and mission into AI as better prompts and to create better outputs.

There’s a bunch of generic copy and internet-speak out there anyway. So with AI becoming more and more ubiquitous, everything you see online is going to become more similar and mushy and vague. Plus, in all the other cases where AI isn’t showing up for you and as you—sales calls, networking calls, workshops, client work, etc, although I’m sure there’s an AI tool for all of the above—you’re still going to have to show your chops. And your heart. And your POV. And it’ll be super jarring if the real person showing up is completely different from the AI-generated website copy or social post or welcome video or email or whatever. That’s why it’s important to invest in your messaging—so that you can stand up for something in order to stand out, as I often say.

And come to me if you want to feel (as ChatGPT so generously summarized for me) “energized,” “excited,” “refreshed,” and “motivated” about your marketing. And your messaging. And the way you show up and make decisions day in and day out. That’s the type of energy I want for you. And it genuinely lights me up to be part of the process (and progress) too!

If you have momentum in your business…

Or moving pieces that don’t seem to come together cohesively…

Or a great idea you want to bring to life…

But you have this feeling that you need to bring out the nuance in order to make things really click—for you, for your perfect-fit people, for the brand you envision—then let’s chat.

Be sure to check out all the resources in the show notes below or at ashleesang.com/episode90. They’re extra EXTRA juicy this week!

And come back next time (in 2 weeks!) because we’ll be in conversation with another ambitious and inspirational woman entrepreneur.