Remember the first time you heard your company was going away AI-FIRST?
Maybe it’s all those hands that feel different from others. “By Q3, every team should have integrated AI into their core workflow,” the CEO said, and the energy shifted across the room (or at Zoom). You saw a mixture of excitement and nervousness through the crowd.
Maybe you were one of the curious ones. Maybe you’ve already created a Python script that summarized customer feedback, saving your team three hours each week. Or maybe you just stay up late one night wondering what would happen if you combined a dataset with a hint of a large language model (LLM). Maybe you’re one of those people who already lets curiosity lead you somewhere unexpected.
But the announcement felt different because suddenly, what had been a quiet act of curiosity was now a line in corporate OKRs. Maybe you didn’t know it yet, but something fundamental had changed in how innovation happens within your company.
How is innovation?
The real change rarely looks like the PowerPoint version, and almost never follows the org chart.
Think about the last time you were genuinely useful at work. It wasn’t because of a vendor pitch or a strategic move, was it? More likely, someone stayed up late one night, when no one was looking, found something that cut through hours of busy work, and mentioned it at lunch the next day. “Hey, try this.” They shared it in a Slack thread and, within a week, half the team was using it.
A developer using GPT to debug code wasn’t trying to exert strategic influence. He just needed to get home before his children. An OPS manager who automated his spreadsheet did not need permission. He just needed more sleep.
This is the hidden architecture of growth. This informal network where curiosity flows like water through concrete… exploring every crack, every opening.
But look what happens when leadership takes notice. What used to be simple and organic becomes imperative. And something that once worked because it was free suddenly stops being effective the moment it’s measured.
The great reverse
It usually starts quietly. Often when a competitor announces new AI features, such as AI-powered onboarding or end-to-end support automation—40% performance gains are claimed.
The next morning, your CEO calls an emergency meeting. The room still comes. Someone clears their throat. And you can feel everyone doing mental math about their job security. “If they’re that far ahead, what does that mean for us?”
This afternoon, your company has a new priority. Your CEO says, “We need an AI strategy. Tomorrow.”
Here’s how this message typically scrolls down the org chart:
In the C-suite: “We need an AI strategy to stay competitive.”
At the VP level: “Every team needs an AI initiative.”
At the manager level: “We need a plan by Friday.”
At your level: “I just need to find something that looks like AI.”
Each translation adds to the stress while diminishing understanding. Everyone still cares, but that changes the intent of the translation. The beginning of a question worth asking becomes a script everyone blindly follows.
Ultimately, the performance of the innovation replaces the object. There is a strange pressure look Like you’re moving fast, even when you’re not sure where you’re really going.
This repeats across industries
A contestant announced that he was going first. The second publishes a case study about replacing the LLMS. And a third shares a graph showing the gains in productivity. Within days, boardrooms everywhere began echoing the same message: “We have to do this. Everyone else is already there, and we can’t fall behind.”
So the work begins. Then came task forces, town halls, strategy documents and goals. Teams are asked to take actions.
But if you’ve been through it before, you’ll know that there are often differences between companies Announce And what exactly are they? do it. Because press releases don’t mention pilots who stall, or teams quietly reverting to old routes, or even tools that are used once and abandoned. You may know someone who was on one of these teams, or you may have been one yourself.
These are not failures of technology or intent. Chat GPT works fine. And teams want to automate their tasks. These failures are organizational, and they occur when we try to simulate results without understanding what caused them in the first place.
And so while everyone innovates, it becomes almost impossible to tell who is actually doing it.
Two types of leaders
You’ve probably seen both, and it’s pretty easy to tell which one you’re dealing with.
One spends a weekend prototyping. They try something new, fail half of it, and still show up on Monday saying, “I built this thing with Claude. It crashed after two hours, but I learned a lot.”
They try to create understanding. You can tell they’ve really spent time with the AI, and struggled with hints and illusions. Instead of trying to sound definitive, they talk about what broke, what worked and what they’re still figuring out. That invite You try something new, because it feels like there’s room for learning. It looks like the same thing by contribution.
Another sends you an instruction in Slack: “Leadership wants every team using AI by the end of the quarter. Projects are due by Friday.” They enforce compliance with a decision already made. You can even hear it in their language, and how sure they are.
A curious leader creates a fast pace. The performer creates resentment.
What actually works
You probably don’t need anyone to tell you where AI works. You already know because you’ve seen it.
Customer Support: LLMS really helps with Tier 1 tickets. They understand intent, draft simple responses and route complexity. Not exactly, of course – I’m sure you’ve seen the failures – but it does matter.
Code Help: At 2 a.m., when you’re half-stuck and your AI assistant suggests exactly what you need, it feels like being an over-caffeinated junior programmer who never judges your forgotten semicolon. You save first, then hours, then minutes a day.
These small, cumulative wins compound over time. They don’t have the impressive changes the decks promised, but the kind of improvements you can count on.
But outside of these zones, things get funny. AI-driven revolutions? Fully automated forecasting? You’ve sat through those demos, and you’ve seen the excitement die down just after the pilot.
Have the architects of these AI tools failed? The technology is barely evolving, and products built on top of it are still learning how to operate.
So how can you tell if your company’s AI adoption is real? Easy just ask someone in finance or OPS. Ask what AI tools they use daily. You might get a slight pause or an apologetic smile. “Honestly? Just chat gpt.” That’s it. Not a $50K enterprise grade platform from an expensive software suite in a last quarter demo or board deck. Just a browser tab, like any college student writing an essay.
You can admit the same to yourself. Despite all the mandates and initiatives, your most powerful AI tool is probably the one everyone uses. So what does it tell us about the gap between what we’re supposed to be doing and what we’re actually doing?
How to change your company
You’ve probably discovered this yourself, even if no one has put it into words:
What do you mean by model: Remember that engineering director who screen-screened his messy, live coding session with Cursor? You learned more from watching it debug in real time than from any polished presentation, because the vulnerability travels much further than the directive.
Listen to the edges: You know who’s actually using AI effectively in your organization, and they’re not always the ones with “AI” in their title. They are curious individuals who are quietly experimenting, finding out what works through trial and error. And this knowledge is worth more than any analyst report.
Create permissions (not push): Those inclined to experiment will always find a way, and the rest will not be moved by force. The best thing you can do to feel curious is to stay curious.
We’re living in this strange moment, caught between the promise of AI and the AI that’s actually on our screens, and it’s deeply painful. The gap between product and promise is wide.
But what I’ve learned from sitting through this pain is that the companies that thrive aren’t the ones that adopt AI first, but the ones that learn through trial and error. He lived with the pain long enough to teach them something.
Where will you be from now on?
Until then, your company’s AI-first mandate motions departmental initiatives, vendor contracts and maybe some new hires with “AI” in their titles. Dashboards will be green, and the board deck will have a full slide on AI.
But in the quiet places where your real work happens, what will change meaningfully?
Maybe you’ll be like those teams who never stopped their silent experiments. Your customer feedback system may lack humans. Your documents can update themselves. Chances are, if you were building before the mandate, you will build after it ends.
This is the hidden architecture of real progress: patient, and completely disinterested in performance. It doesn’t make for LinkedIn posts, and it defies grand narratives. But it changes companies in ways that really last.
Every organization is at the same crossroads right now: looking like you’re innovating, or creating a culture that fosters true innovation.
pressure Perform The innovation is real, and it’s growing. Most companies will be involved in theater and be involved. But some believe that curiosity cannot be forced, and progress cannot be made. Because real change happens when no one is watching, in the hands of people still experiencing, still learning. This is where the future begins.
Ski Chen is the co-founder and CEO of Runway.
Read more from us Guest authors. Or, consider submitting a post of your own! See our Guidelines here.