Tips for Maintaining Authenticity in AI Writing

Don't outsource your writing to AI. Use it to sharpen it. If AI writes for you, it must sound like you. Not like a robot. The four parts explained: 1. Voice — what you stand for. Your principles, your point of view, the promise your writing makes. Voice answers: what do I believe, and what do I refuse to say? It sets the stance before you type a word. 2. Tone — the emotional color. Calm or fiery. Tight or loose. Professional or irreverent. Choose a spot on that grid and hold it. Consistency builds trust. 3. Style — how your words behave. Sentence length. Rhythm. Question clusters. One list per piece. Concrete examples, not vague claims. Style is the fingerprint readers recognize. 4. Structure — how the ideas flow. Start with a sharp hook. Move from problem to principle to playbook. Use clean pivots. Close with a clear action. How to make AI write like you: - Feel it first. Read three strong posts you wrote. Note rhythm, sentence length, questions, and banned words. That is your fingerprint. - Polish with a sample and a tight prompt. Feed one sample and your rules. Ask for a short post in that style. Check tone so you do not drift into beige. - Iterate fast. Generate. Edit to sound like you. Tell the model what you changed. Update the rules. Repeat. Each loop gets closer. Want my exact prompts? Comment "AI Voice" and I will send you a short workbook with the prompts (I do it manually, so give me a few minutes :) Your move.

Do you want to post more on LinkedIn but feel short on writing time—and worry that using GenAI feels inauthentic? If you've struggled with feeling that using AI is cheating, you should know you *can* create authentic, value-filled posts with ChatGPT's assistance -- in less time. In my latest newsletter, I explore these five key takeaways: ✅ Use AI for ideation—not execution. Consider AI as a savvy intern rather than a highly skilled copywriter. Yes, use ChatGPT to generate ideas or structure your thoughts. But the final draft? That should come from YOU – not AI. ✨ A Harvard Business School study found that AI acts as a “cybernetic teammate,” allowing us to produce quality work faster. ChatGPT and Claude make great brainstorming buddies. ✅ Infuse your content with unique insights. Generic listicle AI posts often fail to establish thought leadership. When you can incorporate your personal experiences, case studies, or professional opinions, you provide insight into your expertise and personality.  ✨ Sharing specific anecdotes or perspectives differentiates your content and fosters deeper connections with your audience. Think about the last thought leader you “clicked” with and how much of their writing you read. Yup. You can have fans like that, too. ✅ Craft a hook that mirrors your voice. Using a generic hook that some tech-bro uses won’t help YOU if it doesn’t feel authentic – even if their post has over 1,000 likes. The best hooks draw your readers in, introduce your topic – and sound like you. If you’re stuck., ChatGPT or Claude can give you different hook variations in seconds.  ✨ Don’t ignore your hook or outsource it to a robot. The best hooks draw you in with a statistic, personal story, or lesson learned. ✅ Address the needs of a specific audience member. You’re not writing for all one billion of LinkedIn’s members. Instead, think of writing to an audience of one – like you’re writing an email, but longer. :) ChatGPT or Claude can help you create a reader persona document if you don't have one.  ✨ Focusing on individual concerns demonstrates empathy and positions you as a problem-solver within your network. ✅ Make your content sound like you—even with AI assistance. Yes, ChatGPT can replicate your writing style. But it can’t replicate your snark, fantastic insight, or fun phrasing. I use AI for a rough draft pass and then rewrite the main points in my voice. That way, my content always sounds like me, and I give my readers exactly what they expect. ✨ Don’t take chances with your content’s tone and feel. Let ChatGPT help write the content, but make sure you edit it and make it sound like you. 💬 How have you integrated AI into your content creation process while maintaining authenticity? Share your experiences in the comments below. #SEOCopywriting #GenAI #ContentMarketing

A truly tragic part of this mad AI scramble… Some people would have developed into exceptional writers. 1-of-1 voice. Razor-sharp, original thoughts. And instead, they’ve outsourced everything to AI and now they sound just like everyone else. It’s creative abandonment at scale. What a loss. Do I use AI every day? Yes. Am I going to let it siphon away my skills, my voice, my edge? Not a chance. The alternative... If you want to become an exceptional thinker and writer, and not just copy/paste the work of exceptional thinkers and writers, here’s what I recommend: ✓ Use AI to amplify, not originate. Start with your idea. Bring in AI to sharpen, speed up, or structure. But never lose that core insight. ✓ Keep one hand in the craft. Write something from scratch every week. Even (and especially) if no one sees it. ✓ Don’t say something smart, say something real. AI knows every trick in the book. But it doesn't know your story, your worldview. ✓ Train your thinking, not just your LLM. Your brain is still the best writing engine. Don’t let it idle. Read, discuss, ideate. ✓ Protect your voice like it's your brand.  Because it is. And once it’s gone, you'll fall into the dreaded sea of sameness. With me?

Many leaders on LinkedIn sound the same. AI made it worse. If you're just blending in, you're probably losing trust. I know because I made this mistake. 8 months ago, I started using ChatGPT for LinkedIn posts. The output was fast. It was also lifeless. No engagement. I rebuilt my entire approach. I stopped treating AI like a content vending machine and started treating it like a junior writer who needed to learn my voice. Here are 8 ways to leverage AI for LinkedIn content while keeping your voice intact: 1/ Build a Voice Document Before You Write a Single Prompt Most people open ChatGPT and start typing. That's backwards. Before you ask AI to write anything, create a document that captures how you actually communicate: → 10-15 of your best-performing posts (copy the full text) → Your sentence rhythm (short and punchy? Long and flowing?) → How you start posts (questions? Bold statements? Stories?) → How you end posts (CTA style, sign-off patterns) This becomes your "brand bible." 2/ Create Custom AI Tools Trained on Your Voice Generic ChatGPT produces generic content. Custom GPTs, Claude Projects, and Gemini Gems let you bake your voice into the tool itself. Pre-load each one with your voice doc, 20 example posts, and strict instructions on tone (at a minimum). The setup takes 30 minutes. The payoff is permanent. 3/ Feed It Your Actual Writing Don't describe your voice. Demonstrate it. → Upload real emails you've sent → Paste transcripts from talks you've given → Include Slack messages that capture your casual tone AI learns from examples, not instructions. Give it the raw material. 4/ Use the 80/20 Rule: AI Drafts, You Finish AI writes the first 80%. You do the last 20%. Never publish AI output without a human pass. Not because AI is bad but because AI is generic until you make it specific. 5/ Create a "Do Not Use" List Every AI model has favorite words. They're usually the same words everyone else's AI is using. Build a banned list and enforce it. Add this list to your custom GPT instructions. Tell the AI: "Never use these words. If you're tempted to use them, find a more specific alternative." 6/ Read It Out Loud Before You Post Read your AI-assisted post out loud. If it doesn't sound like something you'd say in a conversation, rewrite it. Your voice probably has something AI doesn't: → Incomplete sentences → Casual transitions The out-loud test catches what your eyes miss. 7/ Anchor Every Post in a Specific Experience Generic AI content happens when you give generic prompts. "Write a post about leadership" produces garbage. "Write a post about the time I canceled three 1:1s in a row and lost my best employee" produces something real. The more specific your input, the more specific your output. People don't follow generic advice. They follow specific voices. Train your AI. Protect your voice. Stand out. Get a high-res pdf of the infographic: https://lnkd.in/gCEkWjjg Save this for future reference.

At this point, we can all spot AI copy a mile away. It starts with grand declarations (“In today’s digital landscape”) and ends with clunky signposts like “In conclusion.” Readers tune out the second they see those tells. If you want your writing to sound like it came from a human mind (like yours), you need to know the AI junk to avoid and what to use instead. 1. Skip the “big sweeping” intros. AI loves to start with grand statements like “In today’s digital landscape” or “Since the dawn of the internet.” But people don’t talk like that. Instead, drop readers right into the problem or question they care about. (That's good practice in general! Get to the good stuff immediately.) 2. Avoid fake urgency and clichés. Phrases like “Technology is advancing at a rapid pace” or “In this age of digital transformation” are meaningless filler. If the pace of evolution really matters, demonstrate it with data or a concrete example. 3. Skip the warm-up. Phrases like “This article will explore…” or “Here are some tips” just delay the real value. Junior writers (and AI) often fall into this "narration trap." Cut the intro. Start with the point. A clear first sentence is stronger (and more respectful of your reader’s time). 4. Stop saying “Let’s dive in." People rarely say that outside of AI copy. If you need a transition, be specific: “Now let's set this up in AWS.” 5. Watch out for formulaic comparisons. Phrases like “AI is not just a tool, it’s a revolution” or “X is not just Y, it’s Z” scream template. Reframe in plain language: “AI helps with grunt work, but you can also use it to help shape your strategy.” 6. Don’t list your entire audience. AI defaults to “Whether you’re a junior developer or a CTO…” Skip the roll call. Write to one clear audience, their challenges and experiences, and trust they’ll recognize themselves. 7. Quit being Captain Obvious. “It’s no secret that cybersecurity is important.” No kidding! No one thinks that's a secret. Instead, get specific: “Most breaches still start with stolen passwords.” 8. Anchor big ideas in real-world detail. Instead of saying “Cloud computing has fascinated businesses for decades,” show the shift: “Netflix used to mail DVDs. Now it runs on AWS.” The choice is simple: write bilge, or write helpful content that gets read. "Let's dive in!" 😁

We can all tell when you hit copy-paste on your AI output. It’s giving “meh.” If you’re using AI to write content and it sounds robotic, flat, or just… not like you, you’re doing it wrong. I lead with AI in my work every day. Not just ChatGPT, but trained agents, embedded workflows, automation, real-time co-pilots. I know what good looks like. Not because I’m lazy, because I know how to get the best out of it. Here’s the truth: AI isn’t your ghostwriter. It’s your collaborator. And the work is still yours. Want better AI content? Try this: 1. Train your tools Drop in a few posts you’ve already written. Explain your tone. What to avoid. What good looks like. 2. Voice it out Dictate into your phone or use Superwhisper. Talk like you would to a friend. Let the AI help you shape it later. 3. Structure your thoughts Use bullets or a quick outline. A solid structure gives the model something to build on instead of guessing. 4. Edit like a human Always. The model is smart — but not you smart. Tweak it so it sounds like you, not the machine. 5. Stop chasing “perfect” People don’t want perfection. They want you. Use AI to get faster, not faker. If you’re a ghostwriter or a marketer using AI for clients, same rules apply. Do. Not. Paste. Without. Thought. AI is powerful. But the real power is in how you use it. Let’s do better.

I just got YET another thank-you email 📧 from a candidate that was clearly a direct copy-paste from ChatGPT.... And that just reminded me that… …a key part of GenAI literacy is understanding when and when not to use AI-generated output: Use GenAI too sparingly, and you miss out on (potential) efficiency gains (which are huge especially for ESL folks); overuse it, and you risk losing your unique voice. I’ve chatted with a few folks on this topic and wanted to share eight tips for writing authentically while using GenAI: 1️⃣ Motivation: This one might seem odd, but I find that I do need a little inspiration and reminder not to rely on these tools too much. Sometimes, it's tempting to settle for "good enough" output from ChatGPT, but then I remember the reason I write. Is it to churn out unoriginal content, or did I start writing online to connect with others through my words? 2️⃣ Cultural References: Incorporating cultural references, idioms, and expressions relevant only to your audience and topic can make your text more much more relatable. AI is unlikely to grasp these nuances, which can help your content stand out in a sea of generic synthetic material. 3️⃣ Fine-tune the AI Model: Feed your writing into the model and ask it to mimic your style, voice, and tone. For a shortcut, you might even ask it to define your tone of voice based on your text. But remember to still check the output. 4️⃣ No Shortcuts: Avoid relying on "AI humanizers," which can produce content that still feels inauthentic. Nothing beats your own proofreading to ensure your voice remains authentic. 5️⃣ Use Active Voice: AI-generated text often defaults to passive voice, leading to weaker, less engaging content. Opt for an active voice to make your writing more direct and engaging (and clearly distinguishable from AI-generated stuff). 6️⃣ Vary Your Sentence Structure: AI often relies on patterns and may repeat similar sentence structures, leading to monotonous content. Try mixing short, snappy sentences with longer, more descriptive ones. This captures readers attention and conveys complex ideas better than the monotonous AI-generated-pattern-heavy writing. 7️⃣ Check for Formulaic Language: Be wary of predictable phrases. Not sure why but a lot of ChatGPT responses I get start with “in the realm of.” This, of course, adds no value and can make your writing appear cliché. Keep an eye out for such formulas. 8️⃣ Editing and Proofreading: It may seem obvious, but after using AI to generate text, always edit and proofread. Never let AI output go unchecked. Especially if you’re sending thank-you emails 😏 We're developing lessons on this very topic, but I wanted to share some preliminary thoughts! Just remember: People want to hear from YOU! #edtech #edtechstartup #GenAI #AIupskilling

This is the third time someone has said to me: “the em dashes make it feel like AI is talking.” And the writer in me wants to address that. I have been writing 2,000-word essays almost every week for the past six years. Em dashes are not a glitch that showed up after AI. They are part of my writing rhythm. And if I am very honest, they are part of how I use “you.” So, it is hurtful to hear - remove a part of your writing identity because people now read it as AI. And I think this is becoming a bigger problem than we are willing to admit. We are starting to use tiny syntactical markers as proof of artificiality. ● An em dash. ● A semicolon. ● A certain kind of list. ● A sentence that sounds “too polished.” And in the process, we are flattening human writing to protect ourselves from uncertainty. But AI did not invent the em dash! AI learned from human writing. And even now, most people are still feeding it through “persona”, context, examples, and past materials. Which means what you are often noticing is not “AI style” in some pure form. You are noticing recycled, mimicked, patterned human style. But if we want to keep things human, we cannot start erasing the very quirks, cadences, and syntactical preferences that made human writing feel alive in the first place. We cannot say we want authentic voice — and then ask people to remove the markers of theirs. Yes, there are real conversations to be had about disclosure, overuse, lazy writing, and how AI is reshaping trust online. I am not dismissing that. I am saying: please stop using small writing markers as lazy identifiers of who is and isn’t human. Some of us have loved our em dashes for a very long time. And if we want to keep writing human, we have to stop asking humans to sound less like themselves. #nonprofits #community #aiethics

Being real isn’t the flex. Being remembered is. Everyone says “just be authentic.” But no one explains what that means. Or worse—they weaponize it. Polished = fake. AI = inauthentic. Vulnerable = weak. Let’s clear the air and bust the real myths 👇 𝑴𝒚𝒕𝒉 1: Authenticity means full transparency Wrong. No one trusts a flood. Trust starts with curation. 𝑴𝒚𝒕𝒉 2: Being real means being unfiltered Nope. Editing shows respect. For your audience and your message. 𝑴𝒚𝒕𝒉 3: Your story is your brand Not quite. A story gets attention. Relevance builds trust. 𝑴𝒚𝒕𝒉 4: AI makes content inauthentic Only if you copy-paste. Tools don’t kill voice, blending does. 𝑴𝒚𝒕𝒉 5: Personal brand = performance Wrong again. It’s your leadership, visible and intentional. You want to stay authentic—without shrinking your presence? This is how you do it right 👇 📌 Where AI helps Structure, grammar, tone-checks Use it to shape, not to speak 📌 Where AI fails Story, stance, lived experience That’s you. Every time. 📌 What to keep Your rhythm Your word choices Your actual point of view 📌 What to cut Generic “voice of brand” polish Excessive disclaimers Second-guessing in every sentence 📌 What builds trust Real tone Clarity over performance One strong opinion—stated simply 📌 What breaks it Trying to be liked Pretending you're not trying Authenticity isn’t soft. It’s how you lead in a room you don’t control. So—what part of your content still sounds like strategy?