I’ve been working with AI long enough now that I can usually spot a bad answer on sight. But I’ll be honest — there was a stretch a few months ago where I got sloppy. I trusted a few responses too quickly. I used AI-generated content without checking references. I even shared something internally that sounded brilliant but turned out to cite documents that… didn’t exist.
That moment embarrassed me. And it woke me up.
Since then, I rebuilt my entire workflow from the ground up — everything in my world now has sources, citations, and references. Even this article.
But that experience did something else: It made me pay closer attention to how AI is changing me — not just what I produce.
And then I started noticing something in my company as well.
1. The Moment This Paradox Hit Me
I watched one of my best team members become more productive and more isolated at the exact same time.
Same person. Same role. Same month.
That shouldn’t be possible.
And yet the more I looked around, the more I saw the same pattern:
People are finishing work faster
Fewer Slack messages
Fewer “Hey, can I run something by you?” moments
Fewer little interactions that make a company feel human
It bothered me enough that I started digging into the research. And that’s when things got… uncomfortable.
2. The Productivity Numbers Are Wild — But They’re Only Half the Story
One of the most rigorous studies out there analyzed 5,172 support agents using an AI assistant. The results shocked me:
14–15% productivity lift on average³
~34% gains for novices³
What surprised me wasn’t the gains — it was who gained them. The people who struggled the most suddenly accelerated the fastest. AI leveled the playing field. But buried in that same field of research was something I didn’t expect… and honestly didn’t want to see.
3. A Pattern I Wish I Could Unsee: AI → Loneliness → Fatigue → Counterproductive Behavior
A 2025 study unpacked something I’ve been sensing but couldn’t articulate:
Working more with AI makes people lonelier.⁶ Loneliness drains emotional energy. And drained people start withdrawing, disengaging, or pushing back.
The researchers actually mapped it:
AI collaboration → loneliness → emotional fatigue → counterproductive behavior.
Once you see that chain in action, you can’t unsee it. It explained exactly what I was witnessing in real-time — faster work and fewer connections.
And if I’m honest… I felt it personally, too. There were days when I worked almost entirely through AI systems, moved a ton of work forward… and interacted with nobody.
For a guy who cares deeply about God, family, people, and balance… that’s a problem.
4. The Leadership Insight That Stopped Me Cold
Here’s the part that made me sit back in my chair:
In the same study, high leader emotional support reduced loneliness by 44.7%.⁷
Forty-four percent. That’s not a small adjustment — that’s a lever.
But here’s the punchline:
Most leaders are unaware that they’re even pulling this lever.
And AI is making the consequences louder.
Leaders aren’t just responsible for productivity anymore. They’re responsible for protecting the emotional fabric of the workplace — the part AI can’t touch.
5. A Quick Detour: The Messy Middle No One Talks About
Something else surfaced during all this research — something less philosophical and more… annoying.
Workslop. Low-quality, AI-generated junk content that looks right but is wrong in subtle ways.
A national survey found over 40% of workers have received workslop in the last month.⁸ And in my case, I created my own workslop by trusting results without verification.
I don’t think this is a separate issue. I think it’s all part of the paradox:
More efficiency
More mistakes
More isolation
More uncertainty about what to trust
Speed isn’t the problem. Speed without grounding is.And that’s why citations matter — at least to me — far more than they used to. In fact, I have a project in the works that I hope to help with this need... stay tuned for that one, soon.
6. So What Does This Mean for Me? (And Maybe for You Too)
This is the part where most articles give a neat, polished list of recommendations.
I can’t do that here. Because I’m genuinely still figuring this out.
Here’s what I am seeing so far:
AI is accelerating productivity faster than anything I’ve seen in my career.
It’s also quietly draining the human connection that keeps teams strong.
Leadership matters more right now than it has in a decade.
Belonging is becoming a competitive advantage.
Verification and citations are not optional anymore — they’re the cost of entry.
I’m experimenting in real time, trying to design workflows that deliver both truth and connection.
There’s tension here. And I don’t think that’s going away.
7. The Question I’m Sitting With
We’re entering a time where AI will continue to multiply what humans can do.
But if AI accelerates our output while eroding our relationships, what does that actually mean for human flourishing?
For families? For work cultures? For meaning?
For me personally, this paradox touches all four walls of my life — God, family, work, and myself. If I don’t acknowledge the emotional cost of AI (in myself or in my company), I’m just becoming a highly efficient version of a disconnected person. And that’s not success.
The companies that win won’t just use AI. They’ll master the balance between productivity and belonging.And I’m still trying to crack that code.
Footnotes
¹ National Bureau of Economic Research (NBER). “Workplace Adoption of Generative AI.” 2024.
² Stanford HAI. 2025 AI Index Report. 2025.
³ Brynjolfsson, Li & Raymond. “Generative AI at Work.” Quarterly Journal of Economics. 2025.
⁴ Federal Reserve Bank of St. Louis. “Measuring AI Uptake in the Workplace.” 2024.
⁵ Boston Consulting Group (BCG). “How Generative AI Helps Knowledge Workers.” 2024.
⁶ Meng, Q., Wu, T-J., Duan, W., Li, S. “Effects of Employee–AI Collaboration…” Behavioral Sciences, 2025.
⁷ Ibid.
⁸ Stanford + BetterUp “Workslop” Findings, reported via Quartz/HBR, 2024.