In the last week of December, the LinkedIn feed fills with reflection. Gratitude. Lessons. Growth. Quiet triumphs. Carefully framed struggles. Everyone pauses. Everyone learns. Everyone emerges stronger.

And everyone sounds exactly the same.

You see the same structure, the same emotional tempo, the same approved vocabulary. “As I reflect on the past year…” “Grateful beyond words…” “Three lessons I’m taking into next year…”

It’s the professional equivalent of endless Instagram selfies of sunsets on a Greek island. Calming, and entirely interchangeable.

What we’re witnessing is not simply a seasonal cliché outbreak. It’s homogeneity rewarded by algorithms, and now intensified by large language models (LLMs) that make it effortless to produce fluent, “LinkedIn-safe” English no matter your native language.

None of this is malicious. These posts are written by thoughtful, sincere people. And yet, scroll for three minutes and you begin to feel as though you’re reading the same post, slightly reworded, by someone who may be a founder, work in product, people ops, or “building things”.

The tragedy is that LinkedIn can be so much more. It has the potential to be the internet’s most powerful learning platform: a global classroom where hard-won insights circulate between industries, countries, and career stages. 

This is the place where a single 40-word post by Wharton’s Adam Grant has you thinking for the rest of the day, and Brené Brown gives you the understanding and the confidence to be vulnerable.

It’s where Amy Edmondson shares her own compelling insights about leadership, and generously shares the work of others for you to discover and learn.

Thanks to Linkedin you get to read a deeply personal reflection from the former Dean of Darden, Robert Bruner about retirement and reprieve, and mix it up with reflections on fulfilment, trust and collaboration from Simon Sinek.

It’s the place where Mark Ritson’s pithy insights about marketing hit the mark, while Santiago Iniguez inspires you with philosophy and the link between Dante and the workplace, and IMD’s Howard Yu takes a deep dive into US and Chinese attitudes to AI.

But beyond these gems, Linkedin risks becoming a factory for platitudes.

The question isn’t why do people do this? It’s why has a platform with this much collective intelligence converged to sameness?

Let’s name the forces producing this sameness, and then talk about how to raise the bar.

Sameness is not a coincidence, it’s a coping mechanism.

When humans feel uncertain, we copy each other. Not because we’re lazy, but because imitation lowers risk.

LinkedIn is full of uncertainty. Will this hurt my reputation? Will my boss see it? Will a recruiter misread it? Will someone with a job title longer than mine decide I’m “not senior enough”?

So people reach for formats that look safe. Posts that have clearly worked before with structures that feel “LinkedIn appropriate”. Three lessons. Five reflections. A soft landing at the end.

You can’t really disagree with “growth happens outside your comfort zone”. That’s the point.

The algorithm quietly reinforces this. Posts that offend no one travel further. that feels vaguely wise but not terribly specific floats serenely to the top.

We end up optimising for approval, not insight.

The humblebrag problem – still not fixed, still everywhere

A special mention here for the humblebrag, that most persistent of professional coping strategies.

“I’m so grateful to have been promoted.”
“This year really stretched me as I stepped into leadership.”
“I didn’t expect this recognition, but…”

It’s meant to signal modesty. It often signals calculation.

Research has shown for years that humblebragging tends to backfire. People usually prefer directness. Or at least honesty. But on LinkedIn, humblebragging survives because it feels like the least dangerous way to say, “Something good happened to me, please don’t hate me.”

When everyone uses the same workaround, the workaround becomes the content.

Enter LLMs, fluency without fingerprints

Now add large language models to the mix.

For many people, especially those writing in a second or third language, LLMs are a genuine gift. They lower barriers. They make you feel so fluent. They help people participate who previously felt excluded by tone, grammar, or confidence.

That matters a lot.

But there’s a side effect we’re only beginning to notice: global professional English is being quietly standardised.

LLMs are very good at producing text that sounds “right”. Balanced. Reflective. Slightly warm, and slightly vague. They sand down rough edges. They replace unusual phrasing with familiar rhythm, and turn local idiom into international beige.

Over time, that doesn’t just smooth language, it smooths thought. This itself sounds like something ChatGPT would write, ‘It’s not X. It’s Y.”

When thousands of people ask the same tool for “a reflective LinkedIn post about lessons learned this year”, they don’t get plagiarism. They get convergence. Different lives in the same voice.

No one wants to be the person who ruins the vibe by being specific. Or worse, uncertain. Or worst of all, wrong.

So we get reflection without risk, learning without evidence, and growth without texture. And now we can generate it in 30 seconds, which is both impressive and faintly tragic.

The cost is that LinkedIn becomes soothing, not useful

A platform full of smart people should be a place where you actually learn things. Where ideas travel across industries. Where mistakes are examined properly. Where someone else’s hard-earned insight saves you six months of pain.

Instead, during peak reflection season, LinkedIn becomes a warm bath of agreeable statements. All very calm andsupportive, but you don’t feel wiser.

LLMs seem to offer the perfect product: “Give me a warm, reflective post with three lessons and a hopeful close.” Out comes a competent paragraph littered with emojis that could belong to anyone.

It’s not that people are inauthentic. It’s that the AI machine and the rewards for visibility remove the personality. 

How to sound like a human again, without becoming unbearable

If we want LinkedIn to be more than a corporate mindfulness app, we need to change what we reward in ourselves.

Here are a few practical ways to do that.

1. Stop offering lessons. Start offering evidence.

“Lessons” are cheap. Anyone can have three of them before breakfast. Evidence costs something.

Instead of, “This year taught me the importance of communication,” try, “In March, we realised our project updates were lying to us. We replaced them with a single weekly metric. Delivery time dropped by 22%.”

Only one of these teaches.

2. Write one scene, not a summary

Summaries invite clichés, and being specific is the antidote.

Describe the meeting that went wrong. The email you rewrote five times. The number that made your stomach drop. The one line of feedback that stung.

Details make you legible as a human, not a template.

3. Make a claim with edges

If no one could possibly disagree with your post, it probably isn’t doing much work.

Try making one claim you’re prepared to defend. Not aggressively, just honestly.

“I think quarterly planning is actively harmful for early-stage teams.”
“I’ve stopped asking for culture fit interviews and here’s why.”

Edges create learning, while platitudes create applause.

4. Use AI badly, on purpose

If you use an LLM, don’t ask it to “write a LinkedIn post”. That’s how you get soup (or is it slop?) 

Ask it to organise rough notes and fix grammar without changing tone. It’s a great tool to shorten something you’ve already written.

Give it constraints. Tell it you want no clichés, no ‘as I reflect,’ no ‘grateful,’ and no ‘journey.’

Then rewrite the final version yourself, because your voice is not a formatting problem. Maybe use short sentences and unexpected phrasing. If your British, sprinkle in your dry humour. Add the line that feels slightly risky but true.

And if you’re writing in a second language, use AI for clarity but actively re-inject your natural rhythm and keep your own cultural metaphors. Don’t let the model translate you out of existence.

5. End with a question that isn’t decorative

The most useful posts often end with an invitation to think, not a moral to applaud. Not, “What do you think?” That’s LinkedIn’s equivalent of a shrug.

Ask something that reveals practice, not opinion, such as “What’s a process you quietly abandoned last year because it didn’t work?”

We all might actually learn something.

A challenge for 2026 to raise the bar

We are collectively training each other to write like corporate wallpaper. And now we have machines that can generate that wallpaper instantly, in perfect English, for any speaker, anywhere.

That doesn’t mean we should reject LLMs, but perhaps we should demand more of ourselves when we use them.

LinkedIn doesn’t need more polished reflections. It needs more transferable knowledge: decisions, trade-offs, experiments, failures, and the messy reality behind outcomes. It needs posts that make a reader better at their job, not just better at nodding along.

So, before you post your next reflection, try this. Delete every sentence that could have been written by someone in your industry who has never met you.

If there’s anything left, you’re on the right track. Otherwise, we’ll keep scrolling past beautifully fluent posts that leave no mark at all.

And that would be a shame. Because somewhere underneath all that reflection, there are real stories worth telling. They’re your stories, and they’ll sound so much better if they sound like you.

Interested in this topic? You might also like this…