Soft Skills Are Hard Now

Why the most valuable abilities in the AI workplace aren’t technical

---

A team finishes a pitch deck. It looks sharp—ChatGPT polished it in 14 seconds.
But the proposal still stalls.

Not because the slides are wrong. Because the story is.

Let me know if this sounds familiar:

Your team keeps tweaking, rewording, reframing—each round getting them closer to… nothing.
The real blocker isn’t the work. It’s the conversation about the work.

This isn’t a one-off. It’s a signal.

As AI gets better at helping with the hard stuff—analysis, writing, summarizing—the bottleneck shifts to something squishier: how we ask questions, align decisions, bounce back, and stay human in rooms full of automation.

The tools are getting smarter. That doesn’t guarantee the teams will.

---

The Stack That Doesn’t Go Obsolete

AI is automating tasks.
Humans still have to relate, decide, and adapt.

The World Economic Forum’s latest skills forecast makes this crystal clear: six of the top ten in-demand capabilities through 2027 are human.¹
Not technical. Not new tools.
But timeless abilities like empathy, curiosity, resilience, leadership, listening, and systems thinking.

At Orion Group, we call these the “new core stack” of work.²
Not because the others don’t matter—but because these don’t go obsolete. They scale. They compound.
And most importantly, they’re the things AI can’t fake.

Yet most organizations still treat them as extras. Add-ons. Nice-to-haves.
Which is why they’re often the first thing to break under pressure.

But what if we treated these as core infrastructure?

---

The Three Most Practical to Build First

Empathy — The trust that teams run on

Empathy is often misunderstood as being nice. But it’s not about tone. It’s about tuning.

Gallup’s global panel found that high-empathy managers more than double employee engagement and reduce voluntary turnover by 37%.³
Meanwhile, Edelman’s 2024 Trust Barometer shows companies scoring in the top quartile for empathy beat others by five percentage points in total shareholder return.⁴

That’s not fluff. That’s throughput.
When teams feel heard, they stop wasting energy protecting themselves.
And they start focusing on the work.

How to build it into your week:

• At the start of a meeting, ask: “Where’s your energy today?”

• Before offering feedback, say: “What I’m hearing is…”

• When someone’s stuck, ask: “What do you need from me?” instead of offering fixes.

Then track the ripple effects.
A two-question monthly pulse—“I feel supported” and “I know what’s expected”—tells you more about team health than any status report.

---

Curiosity — The upstream source of better decisions

Curiosity isn’t about asking more questions.
It’s about asking the ones no one else thinks to ask.

Orion’s report found that curiosity is the best predictor of cross-functional problem-solving.²
And a Harvard Business School paper showed that teams high in intellectual humility filed 2.7× more patents over five years.⁵

But you don’t need to invent the next iPhone.
You just need to stop solving the wrong problems.

Curious teams don’t just challenge assumptions—they discover which ones they didn’t know they had.

Try this 15-minute team ritual:

• Everyone writes three wild “How might we…” questions about a current challenge.

• Share and group the best ones.

• Vote. The top idea becomes next week’s pilot or prototype.

This isn’t brainstorming. It’s exploration.
And the best solutions often come from the question that seemed too weird at first.

---

Resilience — Recovery as a competitive advantage

Resilience isn’t about toughness. It’s about bounce.

The WEF found that demand for resilience is up 43% since 2020.¹
Gallup reports teams trained in resilience show 29% lower absenteeism.³
And Orion flagged “change saturation” as the #1 risk in AI rollouts—because the pace isn’t slowing down.²

If your team doesn’t have space to recover, you don’t just lose energy. You lose judgment.
And eventually, people.

Here’s one way to build recovery into regular meetings:

• Name one thing that caused avoidable stress this cycle.

• Reframe it. (e.g., “confusing expectations” → “opportunity to clarify scope”)

• Pick one micro-habit to reset (a walking break, a daily shutdown ritual, no back-to-back calls).

• Track it the same way you’d track performance: visibly.

If you never check the battery, don’t be surprised when it dies.

---

Measuring What Matters

People often say soft skills are hard to measure.
But what they mean is: we’ve never bothered to measure them well.

Here’s how you can start:

These aren’t KPIs. They’re human telemetry.
They tell you whether the system is healthy.
And when it’s not, where to look first.

---

Culture Is What Happens Between the Tools

There’s one more variable that makes or breaks all of this: psychological safety.

Orion’s research is clear—teams only grow in soft skills when they’re not afraid of looking stupid.²
If people are punished for uncertainty, they won’t speak up.
If failure is taboo, learning stops cold.

This means every new AI tool rollout needs a human counterpart:

• A workshop on emotional intelligence

• A ritual where people share mistakes without shame

• A leader modeling curiosity instead of perfection

Orion suggests a mix:

• Play Labs to practise empathy

• Micro Courses to deepen curiosity

• Practice Labs to build resilience habits

The goal isn’t just to train—it’s to normalize.

---

Don’t Checklist This. Embed It.

Start with five simple moves:

• Send out a two-question empathy pulse Monday

• Try a curiosity brainstorm Friday (15 minutes)

• Add a resilience moment to your next team debrief

• Put soft-skill metrics in your next quarterly review

• Pair every new AI tool with a human-skill ritual

Soft skills aren’t side work.
They’re how the real work gets done.

---

The Closing Question

AI can write the summary.
Can your team write the insight?
The pitch?
The moment that gets remembered?

Because that’s still ours.
And for now, it’s still the most valuable part.

---

Footnotes

1. The Future of Jobs Report 2023, World Economic Forum, p. 5. https://www3.weforum.org/docs/WEF_Future_of_Jobs_2023.pdf

2. The Future of Work & Leadership in the AI Era, Orion Group, 2025, pp. 13–18, 21.

3. State of the Global Workplace 2023, Gallup. https://www.gallup.com/workplace/394922/state-of-the-global-workplace-2023.aspx

4. 2024 Edelman Trust Barometer, Edelman. https://www.edelman.com/trust/2024/trust-barometer

5. Frazier, R. & Edmondson, A., “Curiosity and Patenting Outcomes,” Harvard Business School Working Paper 22-014, 2022. https://www.hbs.edu/faculty/Publication%20Files/22-014_5ed6f71c-9f73-43b0-ada4-fda60c5828d2.pdf