A conversation I had recently with a senior leader left me thinking for days. She’s been deploying AI seriously inside her organization. Not experimenting, not piloting, but restructuring real workflows and committing real resources. She’s seen what it can do. And the thing that surprised me was where the conversation kept going. Not toward what AI could do next, but toward people. What her team members were uniquely good at. Where the human relationships in her business carried weight that no system could replicate. How to make that more visible, not less.
It wasn’t defensive. Not the “but we’ll always need humans” reassurance that shows up in public panels. It was more precise than that. She was looking at AI-native companies entering her space, companies built on a different substrate, moving at a different speed. And when she thought about what keeps her organization competitive, the answer she kept arriving at wasn’t more AI. It was the parts that AI can’t touch. The depth of a client relationship built over years. The institutional knowledge that lives in her people, not in systems. The ability to read a room, to sense what a partner isn’t saying, to know when a situation calls for patience rather than a solution.
She wasn’t retreating from AI. She was doing something harder. Trying to see clearly, in the middle of rapid deployment, which parts of her organization become more valuable precisely because AI is handling everything else. And then investing there.
I don’t think she’s the only one doing this. But it’s not what you’d hear if you followed the public conversation about AI and work, which tends to oscillate between two poles: AI will replace most of what people do, or AI is just a tool and nothing fundamental changes. What she was doing was neither. It was something quieter. A recalibration happening inside the work itself, visible only to people who’ve been deep enough into deployment to have a feel for both what AI delivers and where it stops.
In Part 2, I wrote about grounded intelligence, the kind of knowing that comes from hands-on contact with AI in a specific business context, and how a sandbox team can produce it for leadership. But that conversation made me think grounded intelligence is bigger than a team’s output. It’s a way of describing something that lives in every person who has spent real time in real situations with real constraints. The judgment that comes from having been wrong before in this particular domain. The intuition built from thousands of small interactions that never made it into any dataset. The ability to know that what someone is asking for isn’t quite what they need, and to hold that gap with care rather than efficiency.
AI doesn’t have this. Not because the models aren’t good enough yet, but because this kind of intelligence is grown, not trained. It requires time, error, relationship, and the slow accumulation of context that can’t be compressed into a prompt. She seemed to understand this intuitively, even without naming it that way. What she was protecting and amplifying in her organization was precisely this. The grounded intelligence distributed across her people.
There was a second thing she said that I found even more interesting. Rather than thinking about how AI can replace roles, she was thinking about how AI can help each person on her team grow. Not grow in the sense of “become more productive.” Grow in the sense of reaching capabilities that were previously out of range.
The logic underneath is worth sitting with. When AI handles execution, the drafting, the formatting, the first-pass analysis, the scheduling, it doesn’t just free up time. It lowers the floor of what’s required to operate at a given level. A junior strategist with strong AI fluency can now produce work that used to require a mid-level practitioner. But this doesn’t make mid-level practitioners obsolete. It means the mid-level role is no longer defined by execution; it’s defined by something else. Judgment, contextual reading, the ability to know what the analysis is missing. The capability ladder isn’t losing levels. Every level is moving upward. What used to be the work of each level becomes the baseline of the level below, and what defines each level becomes more about the kind of grounded intelligence that can’t be automated.
If AI is understood primarily as a replacement engine, then every role is a countdown. People feel it, and they either resist or perform busyness to justify their existence. But if AI is understood as something that raises the floor, that gives each person access to capabilities they didn’t have before so they can spend their time on the harder, more human work, then AI becomes growth infrastructure. Not a threat to navigate, but a substrate that lets people become more of what they already were, faster.
There’s a growing unease about AI that I think comes from a real place. The fear of being made redundant, of watching a machine do in seconds what took years to learn. What I took away from that conversation isn’t a rebuttal to that fear. It’s a different frame, one that comes from practice rather than theory. In her experience, AI hasn’t made people less necessary. It’s made it clearer, sometimes for the first time, what about people was necessary all along.
For a long time, the value of human work was tangled up with the execution it produced. Reports, designs, analyses, decks. The artifacts were the proof. When AI takes over much of that production, what’s left isn’t nothing. What’s left is the part that was always there but harder to see: the thinking behind the decisions, the relationships that made collaboration possible, the judgment that knew which question to ask before anyone had thought to ask it. AI is removing the camouflage. And what’s revealed, in the organizations paying closest attention, is that the human layer was carrying more weight than anyone realized.
What’s becoming clearer, to the people doing the hardest version of this work, is what was always at the center.
About the author
Qinqin (Stella) Yang is a director, builder, and imaginist working across AI capability and creative practice. She builds workflows, systems, and creative direction that operate under real constraints, and designs speculative frameworks that expand what teams can imagine before they start making. Over the past decade she has built brand experiences, product design, and speculative projects for Tencent, IKEA SPACE10, Burberry, and the Palace Museum. She is the founder of Actual Fiction, where she develops AI-native tools that turn imagination into a working method.
This newsletter is where she thinks out loud about creativity, AI, and what it means to build toward a future that doesn’t exist yet.
For more details: qinqinyang.com