“You have a ton of work to do. Your site is messy, too much text and the UX is a nightmare. You have to redo it completely.”

So many harsh words with no context and zero softening (did I mention unsolicited?). He came at me - hard, and I felt the sucker punch me. How rude?! I thought, and called him a keyboard warrior in my head. That, and a web developer who is butt-hurt by the existence of AI and that people like me might not need him as much nowadays. But on reflection, it was me, I was the problem and he was spot on.

I’d just soft launched the first version my site (it is here if you want to take a stab too 🤣 but be kind, this is the fresh version!). I spent many consecutive weekends, late nights, very early mornings and parts of my soul to get my music out. Disclaimer: Of course it is AI-assisted (have you met me?), but I was moving fast in a 90-day push and convenience felt like a really good strategy to my exhausted self.

Here’s the part that slapped me in the face: I run on one argument: humans who stay human win. I help other women with this and I still didn’t hear it happen to my own words. I did not catch it before convenience opened the door and let it slither in.

I went back and reread my own homepage like a stranger would. And the stranger was right. It was polished and it was correct. But something was missing and it sounded like every other “AI” founder site built this quarter and nothing like me, even though it wasn’t 100% AI generated. AI still crept in there and did is thing connecting invisible dots.

There’s a framework out of Stanford and BetterUp’s 2025 research that names this precisely: Pilots vs. Passengers. Pilots use AI with high agency. They direct it, they interrogate it, they stay at the controls. Passengers hand over the wheel and hope the destination is fine. Their research found Pilots are 3.6x more productive. Not because they use AI less but because they never stop flying it. Not for a second.

For three weeks, on my own website, I was a being a blind Passenger. Blinded by passion and too deep into the many things I had to say, that I let AI eat my work, all because of overwhelm and self-imposed pressure.

I caught it because someone I’ve never met was paying closer attention to my voice than I was.

That’s exactly the mud (read: it is not mud) one can easily step in. It is not that AI writes badly. It writes fluently, persuasively and it is trained to write like me when I want it to. But it will optimize your voice into the average of every voice it’s trained on and eat itself into agreement with you, unless you’re actively flying it.

The market is already exhausted by this. Consumer excitement about AI has dropped from 50% to 19% in two years. People can feel the sameness before they can name it. We can all spot AI slop a mile away and none of us are here for it.

If you’ve built anything with any input at all from AI recently; a bio, a deck, a website, go reread it like a stranger would.

Would they recognize you in it, do you?

XO Esmarie

Catch me in the wild talking about your Authenticity vs AI, here:

I used to be a proper recluse and social clam. Many people are surprised by this when I say it. It is generally only when I get hooked on a really good convo that I am my extrovert Self, but this is not the norm. At least it never used to be.

Weak ties get credit for being an opportunity (as a rule), but no one prices the conversion tax to access them. A massive 20M-person Science study led by Sinan Aral confirmed weak ties were the most helpful for career advancement, particularly in digital sectors, but weak ties still cost conversion effort. Nobody prices that in. Opportunity and energy are different currencies.

Burnout isn’t caused by talking to too many people, it’s caused by illiquid relationships. You can be completely drained out after 3 conversations, or be on fire and fully wired up on 30. I have learned that Isolating “to protect energy” may itself be the fatigue risk, not the fix, and then I found some research to prove what I already knew to be true:

A Frontiers study on healthcare workers found both moderate and high emotional labor were associated with significantly lower odds of high fatigue, compared to minimal emotional labor engagement which suggests that under-engaging, not over-engaging, may be the bigger fatigue risk.

It’s not about how many people you talk to. it’s about translation cost. Every conversation with someone outside your representational system/frame requires you to code-switch: translate your thinking into their language, manage their misunderstanding, do emotional labor to be understood. That’s the actual energy tax, not headcount. Talking to aligned people costs near-zero translation and this is why Authenticity will always win, and leaning into what feels like alignment when it comes to our network and connections, is fuel.

My conclusion, thus, is this: what you feel when trying to display what’s expected takes real psychological effort, while deep acting (genuine alignment) feels far less draining because there’s no gap between what you feel and what you express. Talking to aligned people equals ease of connection, and even better, it is free. Talking to misaligned people equals surface level labour (aka small talk!), and this is taxing. Do you Curate your connections or are you the human version of a 24hr dollar store?

“In a tech world, I bet on People.” xo Esmarie

If you are into this kind of thing, and want to learn how to practically translate your Authenticity to your relationship with AI, I will be running a workshop on exactly that, here:

📅 28 August 2026
📍 Innovation City Cape Town
🎟 https://www.radi.co.za/events/digital-divas-summit-2026

LLMs are qualitatively different from calculators or search engines. They don’t just retrieve or automate procedures, they can perform integrative reasoning, which is the cognitive process that builds lasting knowledge and identity. When you outsource integrative reasoning, you outsource the process that makes you you.

“The Hollowed Mind” state of dependency where frictionless access to AI-generated answers enables users to systematically bypass the effortful cognitive processes essential for deep thinking. The content feels smooth. The mind behind it has been bypassed.

This leads you straight into a trap: “The Sovereignty Trap” aka a psychological mechanism where AI’s authoritative competence tempts users to cede their own intellectual judgment, mistaking access to information for genuine ability.

Simply put, the outsourcing of thought to AI is a pre-cognitive layer that shapes what reaches human awareness at all. It is nothing short of cognitive surrender, and it’s eating your humanity.

There is a better way to AI (you will agree it is right up there in the verb department with Google now!), without abandoning yourself or your Claude subscription. I teach the HUMAN AI framework and share my Human Edge content systems here (it is free):

(This image has nothing to do with it, just a reminder to look up from your phone every now and again!)

“In a tech world, I bet on people” - xo Esmarie

#AIPoweredWomen #humanity #Aitools

Projects that did define success metrics upfront succeeded 54% of the time. Projects that didn’t: 12%.

Industry consensus generally is “it’s just an alignment problem.”, alignment on success criteria requires that the criteria exist somewhere readily accessible. Also mostly because if it is not accessible, how do you consider its use case, or AI automation readiness?

Most of it doesn’t. It lives undocumented in the judgment of the people doing the work, the minds of your most lethal strategists and the memory of your most talented analysts. This data is the blood in the veins of your organisation that nobody bothers to pulse check until it lands up in the ER and the specialist pulls up in a Bentley. Because emergency, and AI is going to bury this one.

You can’t align people around criteria that don’t exist anywhere except inside someone’s head.

I deeply believe that leaders and decisionmakers aren’t failing to agree, they’re being asked to agree on something nobody has ever written down and they are flying completely blind. They are at the airport and that car is looking very, very comfortable right now.

The alignment gap is a symptom. Undocumented tacit knowledge is the disease that is going to cost you money, and then die anyway.

Berkeley’s California Management Review made this point plainly this year: most companies treat AI as a data and technology challenge. The ones actually making progress have figured out the real differentiator isn’t the data or the model. It’s the tacit knowledge sitting in the judgment of their people, and whether that judgment has ever been made visible.

In a conversation this week with an operations and people strategy leader and all I could think about was this: AI cannot function without data, data means nothing without structure, and structure doesn’t exist until every process, every SOP, every piece of human know-how has actually been documented. Not stored. Documented and with the reasoning intact.

So before talking about AI, ask a harder question: “is our data ready”? Could your team write down, right now, what “this is working” actually means in language specific enough that two people would agree on it?

If not, that’s not a data problem. That’s the actual project.

I am offering a few free Saturday slots for a chat to help you find the pulse of your business, and to see where you should be, could be or definitely should not be automating right or building with now. Drop a dm (As a side rant, I do not use manychat or auto-replies. Ever. I do not run a call centre. It might take me a minute, but you will get a human interaction).

If you prefer, you can also click here for a quick assessment of what we can solve together, first:

https://tally.so/r/gDpA2J

XO

Esmarie

#AI-PoweredWomen #TheAICollective #automation #AIreadiness #dataintegrity

DAINTY by Esmarie Swanepoel

For every $1 businesses spend on software, $6 goes to services. The AI playbook is about capturing the services dollar at software margins, and Anthropic just did that successfully.

Last week, Anthropic partnered with Blackstone, Goldman Sachs and Hellman & Friedman to launch an AI-native enterprise services firm. A week before, OpenAI announced its own consulting arm. Sequoia published the thesis: the next trillion-dollar company won’t sell software, it’ll sell the work (i.e. Results). Here’s a slightly uncomfortable truth that I will invite you to consider:

𝗔𝗻𝘁𝗵𝗿𝗼𝗽𝗶𝗰 𝗲𝗻𝘁𝗲𝗿𝗶𝗻𝗴 𝘀𝗲𝗿𝘃𝗶𝗰𝗲𝘀 𝗶𝘀 𝗮 𝘄𝗮𝗿𝗻𝗶𝗻𝗴 𝘀𝗶𝗴𝗻𝗮𝗹 𝗳𝗼𝗿 𝘂𝗻𝗱𝗶𝗳𝗳𝗲𝗿𝗲𝗻𝘁𝗶𝗮𝘁𝗲𝗱 𝗰𝗼𝗻𝘀𝘂𝗹𝘁𝗮𝗻𝘁𝘀. The race to the bottom on generic AI implementation starts now. But that’s actually the smaller story. The bigger story is this: 𝗧𝗵𝗲 𝗹𝗮𝗯𝘀 𝗮𝗿𝗲 𝗯𝗲𝘁𝘁𝗶𝗻𝗴 𝗯𝗶𝗹𝗹𝗶𝗼𝗻𝘀 𝗼𝗻 𝗔𝗜 𝘀𝗲𝗿𝘃𝗶𝗰𝗲𝘀 𝗮𝘁 𝘁𝗵𝗲 𝗲𝘅𝗮𝗰𝘁 𝗺𝗼𝗺𝗲𝗻𝘁 𝘁𝗵𝗲 𝗱𝗮𝘁𝗮 𝘀𝗵𝗼𝘄𝘀 𝗺𝗼𝘀𝘁 𝗰𝗼𝗺𝗽𝗮𝗻𝗶𝗲𝘀 𝗰𝗮𝗻’𝘁 𝗲𝘃𝗲𝗻 𝗶𝗺𝗽𝗹𝗲𝗺𝗲𝗻𝘁 𝗔𝗜 𝘀𝘂𝗰𝗰𝗲𝘀𝘀𝗳𝘂𝗹𝗹𝘆, and the data is unambiguous:

• 95% of organisations are seeing zero measurable return from AI (MIT Project NANDA, July 2025,300+ AI initiatives)

• 80% of AI projects fail, that’s twice the failure rate of traditional IT projects (RAND Corporation, 2024)

• 42% of companies abandoned most AI initiatives in 2025, up from 17% the year before (S&P Global)

• Only 2% of companies are classified as “high readiness” for AI (Riviera Partners, 2025 — 1,000+ senior tech executives)

The top obstacles identified in 2025 research tell us why: data quality and readiness (43%), lack of technical maturity (43%), shortage of skills (35%).

These aren’t technology problems. They’re organisational problems What I see here, is that the bottleneck isn’t implementation, it’s readiness. Most companies calling for AI implementation aren’t ready for it. Which raises the question nobody seems to be asking: if the labs are building the delivery infrastructure, and most companies aren’t ready to receive it, who bridges that gap?

To my mind, the bridge is someone who understands that 𝗔𝗜 𝗿𝗲𝗮𝗱𝗶𝗻𝗲𝘀𝘀 𝗶𝘀𝗻’𝘁 𝗮 𝘁𝗲𝗰𝗵𝗻𝗼𝗹𝗼𝗴𝘆 𝗽𝗿𝗼𝗯𝗹𝗲𝗺. It’s a human problem. I know this because I spent years inside law firms. Nobody (and I mean nobody) is more resistant to change than a room full of lawyers. And yet, law firms are adopting AI - slowly, deliberately, and successfully. That pace isn’t a weakness. It’s the lesson. I’m trained in NLP, and the patterns of resistance I saw in law firms are identical to what I see in boardrooms today: the language of fear doesn’t change.

The firms or businesses succeeding aren’t doing it because they bought better software or slapped AI onto anything with a hairline fracture. They’re doing it because someone helped them address the human infrastructure first: the fear, the identity threat, the workflow redesign, the cultural permission to change. That’s readiness!

Most companies calling for AI implementation right now are actually asking for something else. They just don’t know the right question yet.

What they need are:

Data infrastructure that works

Teams that are trained, not just told 

Change management that treats humans as the point, not the obstacle 

AI as a tool held in human hands, not a limb replacement

If you’re a CEO or Founder right now, here’s the question worth sitting with before your next AI investment: are you buying implementation, or have you done the readiness work that makes implementation successful?

If you like this kind of content, join my conversations over on LinkedIN

xo Esmarie

A Lancet study found skill degradation in experienced professionals after just 90 days of AI dependence. MIT coined it: “metacognitive laziness.”

In addition to coining what is in my opinion a very articulate way of saying “brainrot”, MIT’s “The GenAI Divide: State of AI in Business 2025” report found that 95% of enterprise generative AI pilots fail to deliver measurable ROI or scale beyond experimentation.

As you may have seen in recent AI news, Microsoft has essentially fired AI coding tools, and is leaning back toward human labour. Uber blew its entire AI budget in a mere 4 months, bloth claiming “AI is too expensive”. I think this headline is actually a distraction. The real story is what companies are losing while they are coming to grips with their cash bleed.

The real story is this: What did your people forget how to do while the AI was doing it for them?

A Carnegie Mellon and Microsoft study found that the more people use AI, the less critical thinking they do, and that AI users produce “a less diverse set of outcomes” compared to those relying on their own abilities (although I didn’t need a study to figure that one out - I can just scroll through my LinkedIN feed riddled with AI slop and the decay is pretty obvious!)

The erosion of human relational intelligence is the real issue at the heart of AI adaptation, and I believe this is a matter of individual responsibility. As with anything on the world stage of ethical debate, it comes down to the individual to put out into the world, what it is they would like to see in the world. It comes down to integrity, even down to how we use AI to perform a task, a job, serve a client or solve a personal problem.

So what does integrity in AI use actually look like in practice? Not as a concept or thoughts as a leader, but as a workflow.

I’ve been playing with the new Claude CoWork Legal Plugin tool. I find this move by Anthropic very relative to the “AI with a human edge” cause, particularly given one of the most ethically exposed professions in the world: law. Legal practitioners are, by professional obligation, the last line of defence between AI output and real-world consequence. They cannot afford metacognitive laziness and their clients certainly cannot either.

I built this quick reference guide that covers the practical use of the Claude Cowork Legal Plugin Tool. If you did not know, this is a free tool for small (or any) firms / practitioners that uses structured AI workflows to handle contract review, NDA triage, compliance checks, risk assessment etc., without replacing the human judgment at the centre of it all. It has been built around it.

If you’re curious or cautious about AI or Claude’s capabilities, particularly the Legal Plugin, this guide is free, practical (with a focus on the Tool itself), and built on one conviction: AI is only as powerful as the human intelligence directing it.

Grab the free guide here → The Human Edge: Claude CoWork Legal Guide

And if you’re building AI around your human edge in any profession, I’d love to hear what that looks like for you. Drop it in the comments or Connect with me

Maybe it is this: The problem is not that you haven’t found the right tool. The problem is that every tool being sold to you was built for a completely different kind of mind. The AI content machine is running on a particular type of intelligence. Faster, more linear, Output-optimised and pattern-based. It does not speak a language, and it cannot read a word of what it spits out. It is recognizing patterns and connecting dots. It strips the nuance, summarises the feeling, bullet-points (or emoji spams) the complexity and bedazzles you with em-dashes. That works beautifully if your edge is speed and volume. But if your edge is something else, if what you bring to your clients, your brand, your work is a kind of precision that comes from how you read a room, or the thing you sense before someone says it, or the particular way you put words together that makes people feel seen, then you’ve probably noticed that AI, left to its own devices, flattens things. And I hate to say it, but nobody seems to be solving for it.

There is a field of study called Neuro-Linguistic Programming or NLP, that has been doing what AI cannot, for decades. It maps how human beings process information. How we filter the world through our beliefs, our memories, the language we use without knowing we’re using it. How we build rapport before a word is spoken. How we read what’s behind what someone is saying, not just what they say. AKA being a human being. When you apply NLP principles to how you communicate with AI, to how you structure a prompt, how you specify an outcome, how you interrogate the output, the results change completely.

There’s a conversation happening in every boardroom, every co-working space, and every employee’s head right now: “Is AI going to replace me?”

The honest answer is: it will replace the parts of you that were never really yours to claim as a competitive advantage in the first place. But there are three things (precise, specific, neurologically grounded things) that AI fundamentally cannot do, and this is your edge, if you know how to merge it.

I’m trained in Neuro-Linguistic Programming, and when I look at AI through that lens, something becomes very clear. AI is an extraordinary processing machine, yes, but what it is not is a human nervous system. And the gap between those two things is where your competitive edge lives. Here are the three things AI will never replicate in precise terms, so you actually understand why I say this, and more importantly, how to use them in harmony with AI:

1. Sensory Acuity

In NLP, sensory acuity refers to your trained ability to detect real-time, fine-grained, non-verbal changes in another person and to use that live data to calibrate your real time response in the moment. We’re not talking about gut feeling. We’re talking about specific, observable data points:

• Micro-shifts in someone’s breathing (from shallow chest breathing to deeper abdominal breathing, which signals a move from thinking to feeling)

• The slight colour change in skin tone across the face or neck

• A change in muscle tension around the jaw or eyes

• Postural micro-adjustments (think about the spine lengthening or collapsing, the head tilting, the shoulders dropping)

• Changes in blink rate, eye moisture, pupil dilation

• The pace and rhythm of speech slowing or accelerating mid-sentence (I am talking about those annoying people who keep pitching their sentences like questions - I see you!)

NLP calls the full pattern the BAGEL model (Body, Accessing cues, Gestures, Eye movements, Language), a framework for reading the whole-body communication system running beneath what someone is actually saying.

Why AI can’t do this

AI processes text, audio tokens, and image pixels. Even the most advanced multimodal AI can analyse a video frame, but it cannot be in the room. It cannot detect the warmth leaving someone’s face when they say “I’m fine” while their breathing flattens and their jaw tightens. It cannot notice that the energy in the room shifted three seconds before anyone spoke. More critically: sensory acuity is a feedback loop. You notice something → you adjust → you notice the response to your adjustment → you adjust again. It’s happening in milliseconds, in the space between sentences, in what is not said. AI processes input. It does not participate in a live, co-regulatory human exchange.

2. Rapport

Rapport is consistently misunderstood as “liking someone” or “getting on well.” In NLP, it is far more precise than that. Rapport is a state of responsive interaction, perhaps better described as a neurological synchrony between two nervous systems that creates the conditions for trust, influence, and real communication. It is built through matching and mirroring at the level of:

• Physiology: posture, gesture, breathing rate, blinking

• Voice: tone, pitch, tempo, volume, timbre

• Language: using someone’s own words back to them, matching their representational system (also known as mirroring - you can also mirror someone’s body language)

That last one is worth unpacking. NLP identifies that people process the world primarily through one of three sensory systems: visual (”I see what you mean”), auditory (”That sounds right to me”), or kinesthetic (”I need to grasp this”). When you match someone’s language system (their predicates) you are speaking directly to how their nervous system processes reality. That is not a metaphor. That is neurological.

The famous communication breakdown stat from Albert Mehrabian: 7% of your impact is the words you say. 38% is your tonality. 55% is your physiology. Which means 93% of human communication the part that actually lands is non-verbal.

Rapport also has a precise test: once you’ve matched, you lead. You make a subtle change such as a shift in your posture, slowing your breathing, and watch whether the other person follows. If they do, rapport is established. You are now in a state of neurological influence, not just proximity.

Why AI can’t do this

AI can generate warm, empathetic language. It can mirror your vocabulary. It can be trained to respond with emotional attunement. But it cannot co-regulate. It has no nervous system to synchronise with yours. There is no breathing to match. No posture to mirror. No voice that adjusts its tempo because it felt you tighten. When a human being experiences deep rapport with another human being, there is a literal physiological shift. Safety is created at the level of the body, not the mind.

AI produces the language of rapport (but it cannot read it). It cannot produce the experience of it.

And here is the business reality: people buy from, invest in, partner with, and fight for people they feel rapport with. Not people they feel pleasantly processed by.

3. Intuition

Intuition is often written off as mystical, the thing you can’t explain and therefore can’t trust. NLP gives it a precise definition that changes that entirely. In NLP terms, intuition is your unconscious mind’s pattern recognition system or a rapid, non-linear synthesis of sensory data, stored experience, emotional signals, and body-based knowing that arrives before conscious reasoning has caught up.

Your unconscious mind stores every experience you have ever had. It has been processing patterns across decades of human interaction, reading micro-expressions, tracking inconsistencies between what people say and how their bodies say it, noticing when something doesn’t add up. When you get a “feeling” about a situation (that tightening in your chest, that sense of wrongness, that inexplicable certainty)

NLP also maps this through the representational system hierarchy, the recognition that different people access intuitive knowing through different channels. Some feel it kinesthetically (a body sensation, a “gut hit”). Some access it auditally-digitally (an internal voice, a knowing). Some see it (a flash of image, a mental picture of the outcome). Learning how you access your intuitive knowing makes it reliable, not random.

This is what the NLP model calls behavioural flexibility. The most flexible person in a system has the most influence. And flexibility is only possible when you can hear the signal beneath the noise.

Why AI can’t do this

AI is a pattern recognition system. But it is a backward-looking one. It was trained on historical data. It extrapolates forward from what it knows (the data).

Intuition, by contrast, is forward-scanning and context-embedded. It is synthesising what is happening right now, such as the tension in this specific relationship, the misalignment in this specific pitch, the wrongness in this specific contract clause, and producing a signal that has no historical match because this situation has never existed before. Intuition is the intelligence of the novel and the now.

There is also a fundamental difference in what gets processed. AI processes what is explicit: what was said, what was written, what is measurable. Intuition processes what is implicit: what was not said, the hesitation before the answer, the word someone reached for and then didn’t use, the way the energy in a room moved when a particular name came up.

And finally, intuition involves your body. The somatic signal. The feeling in your chest or stomach or throat that carries information your cortex hasn’t articulated yet. AI has no body. It processes no somatic data. It will never know what it feels like to know something before you know it.

The Real Conversation

We have spent three years being told that AI is coming for everything, and before someone says it, these are not womens skills. They are human skills. They have been systematically devalued because they were associated with women. The AI era is repricing them, and fast. What nobody is saying clearly enough in gender-neutral language that is actually useful, not just reassuring, is this:

I’ve spent the last two years deep diving into AI and building AI products from scratch. Testing every tool that crosses my path - it is genuinely exciting to me. I have also spent that time learning about NLP and the results of combining this with AI, playing and testing the outcome has been truly profound. AI is nowhere near replacing humans, and I am on a mission to show you how to combine your humanity to get the best from AI, whilst keeping your authenticity in tact.

I have recently being appointed as Global Ambassador and City Chapter Lead (South Africa) for AI-Powered Women, the global organisation rooted at MIT and operating across 6 continents that exists for one reason: to make sure women are not just present in the AI revolution, but leading it. I’m not telling you this to impress you. I’m telling you so you know the bridge I’m offering you is one I actually built and crossed myself. I’m still on it. Still testing. Still learning. That’s the point.

This week, I’m sharing a free guide. “How to Use AI Without Losing Your Human Edge”, Here is what this looks like in practice. The same brief but two prompts. (you get my fully trained Claude Skill to simply install in your Free Claude account). Without the HUMAN Framework it may look like this: ‘Write a LinkedIn post about why AI won’t replace human connection.’ With it: I named my outcome, identified my reader as kinesthetic (she uses words like ‘feel,’ ‘grasp,’ ‘heavy’), specified the emotional register, ran an ecology check, and integrated both the optimistic and the critical version. The output came back warmer, more specific, and sounding nothing like a robot had touched it, because the intelligence directing it was mine. That is the difference.

Before you roll your eyes at another AI resource, let me tell you what makes this one different. The HUMAN Framework. A five-step NLP-informed method for prompting AI that I built and use across everything I create. The step that makes it different from every other prompt framework you’ve seen: you can identify whether your reader processes the world visually, auditorily, or kinesthetically, and you build that into the prompt before you type a single instruction (or combine them, this is your prompt, you do what you like with it). That one step is why the output comes back in your reader’s language, not in generic AI language. No other framework does this. It doesn’t replace your thinking. It helps structure it. You name the human outcome before you type a word. You identify your reader’s representational system (are they visual, auditory, kinaesthetic?) and you match it in the prompt. You interrogate the output the way you would a limiting belief; You run an ecology check. You integrate both the optimistic and the critical version before you commit to a single outcome. What follows is a Conversation, not a Command. The output that comes back sounds like you because the intelligence directing it IS you.

You don’t need to study anything, because I have included A pre-trained Claude skill, ready to install. Free. A custom AI skill built by me on these NLP frameworks combined with the best practices and prompt protocol architecture. You install it in Claude once, and it generates flawless, detailed, specific prompts for any task across your workspace when you want it to: copy, strategy, content, campaigns. And yes, I’ve covered image generation too. If you’ve ever had an AI image come back looking generic and flat, that’s a prompting problem. This skill gives you the equivalent of a film director and a photographic director in your toolkit, permanently. Direct the AI the way a director directs a set (don’t worry you will be prompted each step of the way with questions and with intention, language, and a clear vision for what the final frame should feel like).

The guide and Claude Skill is free. Not “free with a catch.” Not “free then 47 upsells.” Free. Yours. Today, right here: https://drive.google.com/drive/folders/1N0WacJaDwiFPRzTwZI0y63aNBx2pDlwg?usp=drive_link

The AI era does not belong to the people who automate the fastest. It belongs to the people who stay most deliberately themselves. That is you. If you want it to be.

xo Esmarie

Connect with me, let’s grab a coffee!

No matter which side of that equation you are sitting on, everyone is acutely aware of the same thing: that each Tribe has a King. A leader. The one who calls the shots. The one who holds the power, the knowledge (ancestral or acquired), and the golden eggs.

If you have been following any AI news, the concept of Universal Basic Income (UBI) is increasingly viewed as an essential safety net to address widespread job displacement expected to be caused by AI.

It will also not be lost on you that the ones pushing for this are figures like Elon Musk and Sam Altman — both considered modern day Tribal Kings of AI. Whilst UBI could prevent economic stagnation if a large portion of the population loses their jobs, ensuring that people still have money to buy goods.

It could foster social cohesion by mitigating the risks of increased economic disparity, which are often discussed in conjunction with rapid AI adoption, as often explored by thinkers on platforms like LinkedIn.

The real consideration, over the long-term, might not be about finding work, but about defining self-worth in a world where AI performs most of the labor.

But here is the part we are not saying out loud.

When the Tribal King feeds the village, the village does not leave. It does not question. It does not revolt. It simply becomes dependent — and dependence is the oldest form of control ever invented.

UBI, packaged as compassion, may be the most elegant power consolidation move of our generation. Think about it. You stop working. You receive an allowance — funded, in some iterations, by the very companies whose AI took your job. You spend that allowance on goods and services increasingly powered by AI. Your data, your behaviour, your purchases, your patterns — all flowing back upstream. To the Tribe Kings. Who now know everything about you, control the infrastructure you live on, and have your gratitude to boot.

The cave has Wi-Fi now. But it is still a cave.

In ancestral times, the leader who controlled the hunt controlled the tribe. Today, the leader who controls the data controls the world. And data is only valuable at scale — which means the more dependent a population becomes, the more valuable that population is. Not as citizens. As assets.

We have seen this playbook before. Platform economies extracted our attention and called it connection. Gig economies extracted our labour and called it freedom. UBI may extract our autonomy and call it security.

The wolves have always known how to dress.

So before we celebrate UBI as the great equaliser, ask yourself: who is really being protected here, and who is being kept close?

Connect with me on LinkedIN: https://www.linkedin.com/in/esmarie-swanepoel-6766a1107/

The AI gave an answer.
He said it was wrong.
The AI adjusted.
He said it was still wrong.

At no point did he realise the problem might be him.

That’s the state of AI right now. Everyone wants smarter tools. Very few want to admit THEY might be the limiting factor.

The 4D Framework for AI Fluency, developed through research collaboration between Professor Rick Dakan (Ringling College of Art and Design) and Professor Joseph Feller (University College Cork), identifies four core competencies that, when combined, can help you make the most of your AI interactions. Here they are (for free or as we say in South Africa, for MAHALA):

Direction

Most people talk to AI the way they talk to a waiter when they don’t know what they want.
“Something nice.”
“Not too heavy.”
“Surprise me.”

Then they’re disappointed when the result is average. AI is not confused. You are.

The moment you get clear, the output changes instantly. It’s uncomfortable because now there’s nowhere to hide. If it’s bad, it’s on you.

Description

Then comes the overcorrection. Suddenly people write prompts that look like legal contracts. Calm down. You’re not trying to impress the AI. You’re trying to guide it. AI feeds on data - the more, and the more precise, the happier it is and the better it performs. The best users don’t write more. They write better. They know what matters and what doesn’t. That only comes from actually understanding the problem and not decorating it with fluff. Fluff confuses it, focus on real and clear data and instructions (there are guides for free on all LLM’s on how to effectively prompt its models - there are no excuse for bad prompts anymore).

Discernment

This is where it gets dangerous. AI is very good at sounding right. Confident. Polished. Clean. So people read something that looks intelligent and assume it is.

The truth is that AI is not nearly as intelligent as we believe it is. It is just really, really good at connecting data points, discerning patterns and delivering results based on equation. I hate to break it to you, Susy, but it does not understand the words it is producing (i.e. it does not actually speak English).

If you can’t judge quality, AI will make you faster at producing things that look good and mean very little.

Development

Most people treat AI like a vending machine. Insert prompt, collect answer, move on.

The people getting real results treat it like a slightly overconfident assistant. You push it. You refine it. You say “this part is off, fix it.” or, (for the rebels out there), you abandon ship and accept that, in fact, this is not a task for AI at all. Yes, I said it out loud - AI is not the solution for everything.

The more you do that, the more you realise you’re not just improving the output. You’re improving how you think.

The uncomfortable truth is this: AI is not upgrading your intelligence but rather exposing it. Some people use it and look brilliant. Others use the same tool and look like they Googled confidence. Same system. Different operator.

So before you ask which model is best, ask something more useful.

Am I clear.
Am I precise.
Can I tell the difference between good and average.

Because if you can’t, no amount of AI will save you.

And if you can, almost any AI will.

X Esmarie
P.S If you understand the mechanisms now you will own the narrative tomorrow.

Subscribe now