The Ethics of Pretending — AI-Generated Thought Leadership and the Question of Authenticity
The discomfort is not that AI can write persuasively — it is that we cannot reliably tell when it has, and we have not decided whether that matters.
The Uncomfortable Question
There is a conversation happening in professional circles that no one quite wants to have openly. It concerns the growing volume of thought leadership — articles, white papers, conference presentations, LinkedIn posts — that is substantially or entirely generated by AI, yet presented under a human name as though it emerged from that person’s experience and reflection.
The practice is already widespread. Anyone who reads professional content regularly has encountered it: pieces that are fluent, structurally competent, and thematically appropriate, but that carry a curious weightlessness — as though the words were assembled rather than written, the arguments constructed rather than earned. The tell is not in the grammar or the formatting. It is in the absence of the rough edges that come from someone who has actually wrestled with the problem they are writing about.
This is not a technology observation. It is a leadership one. The question it raises is not whether AI can produce credible thought leadership — it demonstrably can — but what it means for professional authority when the connection between experience and expression is severed.
What Thought Leadership Was Supposed to Be
The term “thought leadership” has been debased through overuse, but the thing it originally described had genuine value. A practitioner who had spent years navigating a particular domain — transformation, governance, programme delivery, organisational change — would distil their accumulated experience into a perspective that others could learn from. The authority of the piece came not from the elegance of the prose but from the credibility of the practitioner. You trusted the argument because you trusted that the person making it had earned the right to make it.
This implicit contract — that the voice on the page reflects genuine experience — is what AI-generated thought leadership disrupts. Not because the content is necessarily wrong, but because the contract is broken. The reader assumes they are receiving distilled experience. They are receiving, instead, a sophisticated synthesis of publicly available information, structured to sound like distilled experience.
The distinction matters more than many people seem willing to acknowledge.
The Rationalisation Spectrum
In my experience, practitioners and leaders who use AI to generate content under their own name occupy a spectrum of rationalisation, each position more comfortable than the last.
At one end: “I provided the ideas and the AI just helped me express them.” This is the most defensible position, and in many cases it is genuine. Using AI as a drafting tool, to overcome the blank page or to structure thoughts that the practitioner genuinely holds, is not fundamentally different from using an editor or a ghostwriter. The ideas remain the practitioner’s; the expression is assisted.
In the middle: “I reviewed and approved everything it produced.” This position is more ambiguous. Reviewing AI-generated content for accuracy is not the same as generating insights from experience. A senior leader can read a well-constructed argument about transformation governance and confirm that it contains no factual errors, but that does not make it their perspective. It makes it a perspective they did not object to.
At the far end: “The content represents what I would have written if I had the time.” This is where the rationalisation collapses entirely. It assumes that the value of thought leadership lies in the output — the finished article — rather than in the process of reflection that produces it. The practitioner who has not done the thinking but publishes the conclusion is not sharing their perspective. They are performing expertise.
Why This Is a Leadership Problem
The temptation to dismiss this as a question of academic integrity or content marketing ethics understates what is at stake. For senior leaders and practitioners whose influence depends on credibility, the authenticity of their public voice is not a peripheral concern. It is foundational.
A leader whose published thinking was not produced by their own reflection has outsourced the one thing that distinguishes a leader from an administrator: the capacity for independent judgement, visibly exercised.
Consider the downstream effects. A consulting partner publishes AI-generated articles on digital transformation. Clients engage the firm partly on the strength of that partner’s apparent expertise. The engagement begins, and the partner’s actual depth on the subject falls short of what the published content implied. The gap between the marketed expertise and the delivered expertise is not a content problem. It is a trust problem — and trust, once lost in a client relationship, is extraordinarily difficult to rebuild.
Or consider the organisational leader who uses AI to produce internal communications — strategy updates, change narratives, vision statements — that are published under their name. The workforce reads these communications looking for signals of genuine leadership intent. When the language is polished but generic, when the vision could apply to any organisation in any sector, when the “personal reflections” read as though they were generated rather than felt, people notice. They may not identify the mechanism, but they register the inauthenticity. And they adjust their trust accordingly.
The Harder Question Nobody Is Asking
There is a deeper issue here that the current discourse has not yet reached. If AI can produce thought leadership that is indistinguishable from human-authored work — and we are approaching that threshold rapidly — then the entire category of “thought leadership” faces a credibility crisis.
The value proposition of thought leadership has always rested on scarcity: not everyone has the experience, the insight, and the ability to articulate what they have learned in a way that helps others. AI eliminates the articulation constraint entirely and substantially reduces the insight constraint. What remains is experience — and experience without articulation is invisible.
This creates a paradox. The practitioners with the deepest experience are often the ones least inclined to write, because they know how difficult it is to do justice to the complexity of what they have learned. The practitioners with the least experience now have access to a tool that can produce content indistinguishable from the output of deep expertise. The market for ideas is being flooded with convincing imitations, and the real thing is becoming harder to identify, not easier.
What I Have Observed
The pattern I have seen across professional services, consulting, and corporate leadership is consistent: organisations are adopting AI-generated content faster than they are developing norms around its disclosure. Very few firms have explicit policies. Fewer still enforce them. The default is silence — not deliberate deception, but a quiet omission that allows everyone to maintain the fiction that the content was human-authored.
This is not sustainable, for a reason that has nothing to do with ethics and everything to do with market dynamics. As AI-generated content becomes ubiquitous, its value as a differentiator approaches zero. The leader who publishes AI-generated insights is not building intellectual capital; they are producing commodity content under a premium label. The advantage accrues to those who can demonstrate genuine originality — the rough-edged, experience-grounded, sometimes uncomfortable observations that AI cannot yet replicate because they emerge from the specific, unrepeatable context of a practitioner’s career.
“The irony of AI-generated thought leadership is that it makes authentic thinking more valuable precisely by making polished prose less so.”
Where This Leads
I do not think the answer is a blanket prohibition on AI-assisted writing. That ship has sailed, and the tool is too useful to abandon. But I do think the profession — broadly defined as anyone whose influence depends on demonstrated expertise — needs to confront the question of disclosure with more honesty than it has shown so far.
The minimum standard should be transparency: if AI generated or substantially shaped the content, say so. Not because the content is necessarily worse, but because the reader deserves to know whether they are engaging with a human perspective or a machine synthesis. The two have different epistemic value, and conflating them degrades both.
Beyond disclosure, the deeper challenge is cultural. Organisations and individuals need to decide whether thought leadership is a performance or a practice. If it is a performance — a marketing activity designed to generate visibility and credibility — then AI is simply a more efficient production tool, and the ethical questions are manageable. If it is a practice — a discipline of reflection, learning, and contribution to a professional community — then outsourcing it to AI is not an efficiency gain. It is an abdication.
The leaders I most respect have always been those whose public thinking bore the marks of genuine wrestling with difficult problems. Their writing was not always polished. Their arguments were not always tidy. But you could feel the weight of experience behind them, and that weight is what made them worth reading. That quality cannot be generated. It can only be earned. And in a world increasingly flooded with fluent, competent, weightless content, it has never been more valuable.