If you listened to the AI headlines, you’d assume we’re all living in a glorious future of effortless productivity by now.
Nearly 90% of organizations are using AI in at least one business function (McKinsey). A huge number of employees are using it, too—some estimates put that figure around 75% of workers.
So naturally, you’d expect companies everywhere to be operating with breathtaking efficiency.
Instead, only about 39% of companies report seeing meaningful bottom-line impact from AI (McKinsey).
Which is a pretty strange outcome for a technology everyone supposedly can’t stop talking about.
AI adoption is high. Impact? Still kind of a shrug.
There’s a big difference between using AI and actually working better because of it.
Most people aren’t actually resisting AI. They’re doing their best. They’re trying—weaving it into their days in small, practical ways. To start a draft outline. Clean up an email. Summarize notes. Kick around some ideas when the brain fog rolls in at 3:17.
AI shows up in little moments like these.
But the promise of AI wasn’t “you’ll get slightly faster first drafts.” It was supposed to feel like a whole new way of working.
Right now, it mostly feels like a whole lot of extra work.
Talk to almost anyone in Marketing, Sales, Ops, or Customer Success, and you’ll hear a familiar story.
You start in one AI tool. Then you jump to another because it’s better at that one specific thing. Then another tool. You copy, paste, rewrite, reformat, and babysit outputs across tabs like an exhausted project manager for your own software stack.
Somewhere along the way, you lose context. Long before that, you lost your patience. Maybe even your cool.
Individually, the tools might be impressive. Collectively, it feels like putting together a backyard trampoline in the dark. A whole lot of work, and a lot less fun than everyone said it would be.
Sure, some problems are getting solved in front of our eyes. The models are improving. Response rates are faster. Costs are dropping faster than you can say “compute.”
But the harder problem—the one people actually need solved—is still hanging around like an unwanted guest:
How do I get from “I need to do this thing” to “it’s done”—without doing a shocking amount of manual stitching and fact-checking and reworking in between?
We don’t talk enough about how expensive “using AI well” becomes. Mentally, emotionally. Sometimes even physically. It’s exhausting.
You’re constantly making decisions, checking assumptions, assessing risk and reward.
Which tool should I use?
How should I phrase this prompt?
Is this good enough? Should I try again?
Is this taking too long?
Am I overcomplicating this?
Should I have just done this myself?
None of those questions are catastrophic on their own. But stacked together all day, they create significant drag.
They are, quite literally, a drag.
One study found that employees lose the equivalent of 51 workdays per year to technology friction—disconnected tools, broken workflows, systems that don’t carry context from one step to the next.
That’s why so many people stay in the shallow end of AI use cases: drafting, editing, brainstorming.
Useful? Definitely.
Transformative? Not quite.
The deeper value — the kind that genuinely changes how work moves—still feels frustratingly close, but not fully here.
That helps explain another pattern in the data: Even as adoption rises, many companies are still stuck in experimentation mode rather than scaling meaningful impact with AI (McKinsey).
AI can already do remarkable things. Wonderful things, as a long-ago explorer once said. But getting from a good intention to a great outcome still requires way too much work: too much translation, too much context switching, and too much rework. You keep restarting your engine instead of crossing the finish line.
What’s missing isn’t another feature. It’s continuity. A system that gets you, gets your context, knows what good likes like for you.
The next phase of putting AI to work won’t be defined by better models alone. It’ll be defined by whether the user experience starts to feel obvious. Whether or not it feels good. Whether the tools reduce the cost of making endless decisions instead of creating countless new ones. Whether they help you get real work done, not just endless fragments of it.
If AI disappeared from your workflow tomorrow, would your job get meaningfully harder—or would it be a relief?
The answer says a lot about how far all these miracle products really have to go.
Where does AI still feel clunky in your day-to-day work? Where are you doing more work than the tools are? That gap—that friction between what you’re trying to do and actually getting it done—that’s where the real opportunity still lives.