AI tools have dramatically raised the floor of professional output. A mediocre writer can now produce a decent first draft within seconds. A junior analyst can generate something that resembles a competent data summary without breaking a sweat. A small team of any type can produce in a single week what used to take an entire month. That’s the reality, and it is undeniable. The floor has risen for almost every knowledge worker in almost every field.

But raising the floor doesn’t automatically raise the ceiling. In many ways, it actually makes the ceiling harder to reach.

Here are some thoughts on how AI is reshaping work, what the “sameness problem” means for business, and why “judgment-based or context-driven work” may be one of your most valuable assets.

The Sameness Problem

When everyone has access to the exact same tools producing roughly the same quality of baseline output, the middle inevitably compresses. Which means that work that used to stand out because it was polished, well-structured, or competently researched no longer stands out at all, because AI can produce polished, well-structured, and competently researched output on demand.

We see this happening across industries. Marketing teams publish more content than ever before. Email inboxes are exponentially fuller. Project proposals are getting longer. Strategy reports are more thorough. And, gosh, LI feeds are exploding. Yet, the signal-to-noise ratio has gotten worse. Most of what gets produced is just adequate. It checks the boxes, but it fails to leave a lasting impact.

Across roles, the pattern is the same: the output looks complete, but the thinking behind it never really happened, or, at best, was kept to a minimum.

A marketer can now generate a polished white paper in minutes without conducting any real research. Sure, it reads well, it’s structured properly, and it looks credible at a glance, but it rarely says anything new or speaks meaningfully to the specific audience it was intended for.

An entry-level developer, or even someone with no formal coding background, can build an application using AI tools. The interface may work, and the functionality may appear sound. But beneath the surface, there may be serious security vulnerabilities, scalability issues, or hidden technical debt that only experience would have caught.

A manager can turn a few bullet points into a full performance review. The document is complete, the language is professional, and all the required sections are there. But the employee reading it may feel something is missing: real insight, real care, and any indication that their manager truly understands their work, growth, or struggles.

An account executive can generate a quick snapshot of a prospect without doing the deeper research. The summary looks helpful, but it often misses the nuance, the recent challenges, the internal dynamics, or the specific priorities that actually matter. As a result, the outreach feels generic, and the opportunity to build a real relationship is lost.

In each of these cases, the floor has risen. The baseline output looks okay. But the level of work that is thoughtful, differentiated, and based on real understanding has not moved. In fact, it has become harder to reach because the pressure to produce quickly makes it easier to skip the very thinking that leads to exceptional work.

When the floor rises for everyone, “good enough” stops being a differentiator. It actually becomes the baseline expectation.

This means the only work that actually cuts through the noise (the only work that gets remembered, acted on, or trusted) is the work that AI cannot produce on its own. And that work is getting harder to create. The people who used to dedicate their energy to deep thinking are spending more time managing the flood of adequate, AI-generated stuff.

The Value of Judgment-Based Work

I want to be clear about this. Claiming “AI cannot do creative stuff” is a bit of an overly broad, and not entirely true, argument. AI can absolutely generate creative output. It can produce writing that surprises you, imagery that resonates, and code that functions elegantly.

What AI consistently struggles with is something different. It struggles with work that is rooted in context, judgment, and nuance.

Judgment-based work requires knowing things a machine does not know. It involves understanding the unique texture of a specific client relationship. It requires knowing the organizational history behind a complex decision. It means sensing the unstated concern in a boardroom, or remembering the specific issue your customer base got burned by three years ago. It also means going beyond just summarizing what a customer said, and recognizing what they didn’t say.

The work only you can do draws on judgment built from lived experience in a specific context. It does not rely on pattern-matching across a generalized training set.

Judgment Cannot Be Prompted

The best strategic decisions do not come from having access to more information. They come from someone who knows exactly which information to ignore and why. The best communications come from someone who understands the reader in a way that simply cannot be summarized in a prompt.

That is the ceiling. Getting there requires something AI can help you clear time for, but cannot do for you. It requires accumulated, specific, hard-won judgment.

The Risk

Here is the part that should keep leaders awake at night. Judgment is built through struggle. You develop the capacity to make hard calls by actually making hard calls. You learn by drafting the complex document yourself, wrestling with the core argument, and sitting with the ambiguity until something finally clicks.

When AI removes that friction, it also removes the development opportunity that comes with it.

Think about a junior employee who relies on AI to generate every single first draft. That person is not developing a writer’s instincts. Think about a research team that outsources all synthesis to a software tool. They’re not building the crucial skill of knowing what actually matters in a sea of data.

The immediate output might look solid. But the people behind the work are not growing the way they would have if they had done the heavy lifting themselves.

This is by no means an argument against using AI. I use it every day to streamline workflows and organize thoughts. But we must recognize the difference between using AI to handle mundane tasks and using it to skip the parts of a job that are hard. The hard parts are hard for a reason. Outsourcing them often eats away at the human capacity needed to produce exceptional work.

What to do?

Here are some ideas for fostering depth and judgment within your team while still leveraging AI effectively.

Protect the Struggle

Don’t let your team outsource the hardest parts of their thinking. Encourage them to use AI for formatting, summarizing raw data, or generating basic templates. But insist that the core arguments, the strategic trade-offs, and the final synthesis come from their own minds. Let them wrestle with complex problems so they can build the mental muscles required for higher-level leadership.

Shift Your Metrics

Speed and productivity matter, no doubt. But there are situations that require more than just a boatload of adequate work. Start rewarding the people whose work is impossible to replicate. Celebrate the deep insights, the nuanced client handling, and the creative problem-solving that no tool could have generated.

Teach Context, Not Just Tools

Since judgment-based work is your competitive advantage, actively teach it. Share the history behind company decisions. Explain the subtle dynamics of key client accounts. Help your team understand the unspoken rules and cultural nuances of your industry. Give them the rich context that AI lacks so they can apply their judgment effectively.

Ask the Uncomfortable Question

Here’s a question every leader must ask themselves: If you removed AI from your team tomorrow, what would you actually lose?

If the honest answer is just speed and volume, but the judgment, the relationships, the contextual knowledge, and the creative instinct would still remain intact, then you are using AI incredibly well.

But if the honest answer is that you are not sure, or that you would lose the very depth of your team’s capability, that is worth knowing sooner rather than later. The organizations that will struggle in the next few years are the ones that adopted AI so ferociously that they optimized away the slow, uncomfortable, generative work that made them valuable in the first place. Address this now, while you still have the time to protect your team’s potential.

What about you? If you removed AI from your team today, what would you lose?