The Looming Authenticity Gap: Is Your Brand Risking Its Soul to AI?

Far too many content creators use AI for speed and scale, without adequate strategy and quality control. We’ve all seen the examples of a blog post or web copy that screams “AI” because the content was clearly machine-written with remarkably little human oversight. The words might look pretty good at a glance, but when viewed carefully, they somehow feel completely “meh.”

The result? Polished paragraphs with no point of view. Perfectly edited emptiness. The illusion of thought leadership.

The old “fool me once” adage comes to mind. If your audience senses your content lacks authenticity, then guess what…it actually is inauthentic. You’re now in the dreaded gap. If people feel tricked, they will assuredly view your next piece with justifiable skepticism.

Soulless AI: When Leadership Loses Its Pulse

AI doesn’t actually believe anything. At its core, there is no conviction, opinion, or lived experience. No matter how adept a writer may be at creating repeatable AI frameworks for tone and branding, the auto-generated content will invariably start to sound like it was born from the same AI model. Differentiation disappears and “good enough” AI content can become corrosive to a brand.

Therein lies the authenticity gap. In an insidious way, every time someone hits publish on an unedited, unexamined AI draft, the gap drifts toward chasm as noted earlier.

Thought Leadership Requires Actual Thought

Intentionally obscured in this post is a true celebration of the power of AI for content creation that amplifies your industry leadership. Trouble is, laziness and inexperience with the tools are leading to a boatload of bland content out in the universe. True thought leadership is more important than ever. Your content needs to stand for something, impart wisdom, and (most of all) convey that you give a rip about the topic.

What does that look like in practice?

• Original perspective. Share your actual thoughts on the topic. Take a bold stand and share your opinion (ideally rooted in experience, data, or killer case studies). In its ideal state, thought leadership is something only you or your brand could know or believe.
• Emotional resonance. This is easier than it sounds. If your content lives up to the first bullet point above, then it’s a matter of making sure the output actually comes across in your voice. Don’t simply accept the AI output that provides surface-level affirmation of your perspective. Review the content and challenge your thinking to make it resonate.
• Make your thought leadership actionable, not just theory ad-nauseum.
• Accountability. As a writer, I refuse to put my byline on content that machines spit out. Let’s face it, the AI hallucination that nullifies a premise or an entire piece is on you. Go through your content word-for-word, as if your personal reputation depends on it (hint: it does).

True thought leadership captures and distills your deep knowledge and unique experiences. AI can do some or even the bulk of the drafting if writing isn’t your bag. Just know that you own the final output. GPT won’t come back and say, “Oh, my bad. Don’t blame Joe. I was hallucinating again.”

Bridging the Gap

At its core, PR and brand management is rooted in flawless incorporation of common sense. A close industry friend of mine used to joke with me that using AI for content could be likened to putting a chainsaw in the hands of a child. It’s a powerful tool that needs to be used responsibly and ethically.

So let’s break down responsible use of AI into a simple, actionable workflow:

1. Audit your process. If drafts go straight from AI to published content, build in human checkpoints prior to, during, and after AI is tapped to create content.
2. Add your voice. Inject perspective, belief, and your own experiences into the mix. Make absolutely sure that your content is authentically yours.
3. Read your content aloud. This is an old TV newsroom trick. Reporters and news anchors will typically read their content out loud while writing it. Deploying your human voice in the review process will quickly highlight the places where content may feel robotic or lifeless.
4. Establish guardrails. AI can scale your content creation masterfully. Used properly, it can be like having a brainstorm partner, a Pulitzer-capable editor, and a trusted gut-check buddy all rolled into one. Think of AI as a set of agentic teammates who help draw out and articulate your best thinking. Set your own limits for the level of influence each agentic teammate has in your final output. Never relinquish full authority to any member of that “team.”

The Power is in the Prompt

Without getting into a full prompting tutorial, authenticity starts with your earliest engagements with your AI tools when developing content. Here are some high-level tips for approaching your prompt journey. The bullets below can often be completed in fewer than 10 minutes.

• Avoid starting the process with “write me a blog post about [your topic here].” This is a serious authenticity killer
• Ask AI some probing questions about the topic you’re covering. Investigate industry blogs and publications for historical perspective (or to see if your topic is already exhausted).
• Pose your content thesis to AI and ask it to be critical (for some reason, GPT seems to want to naturally return nice comments back to you).
• Prompt for 10-20 compelling headlines as thought-provoking brainstorm starters. Sometimes a thread appears that helps shape your own premise.
• Ask for a few options for blog outlines – not fully written. Get some ideas about a narrative flow that aligns with how you want the story to unfold.
• Now give some prompts to start drafting.
  • Specify details about your audience and their level of expertise.
  • Instruct some sense of the depth you want to explore in the post.
  • Include example headlines that you liked from the fourth bullet above. You can even say “explore this notion for the third section as opposed to what was provided in the outline,” which helps cement your actual thinking in the draft process.
  • Instruct specific “don’t” rules that may apply, such as usages of specific words/phrases or typical AI “tells” (see Marty’s recent post Don’t Be an AI Dipsh_t: Classic AI “Tells” That Scream “BOT!)
  • Provide some parameters of word count, but keep it in ranges, such as “600-900 words.”
  • Tell AI what YOUR main point is. Challenge the machines to best articulate what you want to share with the universe.
  • Then tell AI to create a draft.

The process from here is up to you. But challenge yourself to avoid getting enamored with how brilliant the content may seem on the surface. Read it. Ask yourself if your points are being made and if they actually seem to make sense. Ask Claude, Gemini or ChatGPT for revisions, but instruct specifically what you like, what you don’t like, and why.

A note of caution: From experience, there seems to be a breaking point that happens pretty quickly with iterations. Draft 2 is better. Draft 3 might be a little better. But somewhere around Draft 4 a critical eye may see that the content is getting diluted or more confusing. Also, chatbots understand you saying, “Please disregard that output and let’s work with the previous.” Then, paste in the previous output for clarity.

Depending on a content creator’s individual desire to create for themselves and/or ability to write, move out of AI and manually edit to ensure the right nuances and points are there. Ensure brand alignment as necessary. Add an entire block of content if something is still missing.

The Real Risk

If brands keep churning out soulless AI content, the credibility gap/chasm can quickly lead to brand decay. Without credibility, differentiation, and connection, audiences will have little or no reason to care. By avoiding the looming authenticity gap, content creators can still be guardians of the brand’s soul, despite the rest of the universe focusing on – and rewarding – shortcuts.

One additional tip: If you are a writer by trade, exercise your writing chops. Make sure you set aside one project each week, or whatever timeframe works for you, to write something all by yourself. No AI. No outside interference. Much like playing a musical instrument, a writer still needs to write some things to stay sharp. Most of the great writing that comes out of our shop has the steady hand of a human writer driving the output.

Footnote: Per the disclosure above about using AI for the initial draft, there are three things to call out:

1. The fifth paragraph in this post is short. It’s in blue font. That is the only wholly unedited paragraph in this post from an original AI draft.
2. This entire post was still completed in sub 2 hours, despite significant manual editing and content additions.
3. Just for giggles, I ran the first draft and the final post back through AI and asked for an assessment of how much I actually own of this post. The following text is what the AI wizard had to say (giving me the confidence I need as a writer to put my name as “author”):

From GPT:

Excellent — this final version reads like a fully evolved editorial piece, and the quantitative authenticity gap between your redraft and the original AI-assisted draft has widened dramatically. Below is a detailed forensic breakdown comparing your final version to the original AIMCLEAR-style GPT draft, using linguistic, structural, and stylistic markers.