🎉 What I’m Carrying Into 2026: Permission to Learn in Public

[Hannah_Gaddini]

Hi everyone,

2025 taught me that no one is “ahead” in AI—we’re all figuring it out in real time. The times when I had the biggest wins weren’t when I was the most polished or prepared. They were when I gave myself permission to just start, stayed curious about what others were trying, and let myself fail fast.

About eight months ago, I began building AI-powered workflows inside my own Asana environment as an experiment

No big strategy.
No approvals.
Just me trying to do my job better, learn my own product more deeply, and help my customers see real value.

The first workflow I built was straightforward: a five-minute form where clients described the value of Asana in their own words. Once submitted, Asana AI renamed and organized the task, pulled insights from their responses, wrote a clean value story, and multi-homed it into the relevant account plan.

This helped ensure that the value stories shared with me were in the customer’s language—not scattered across notes or sitting in my head. It was visible, reusable, and easy for my account team to build on. Several teammates across our global org adopted the workflow. It was a clear, low-risk, high-impact win.

---

Learning in Public Means Owning the Missteps

But giving yourself permission to learn in public comes with the other side of the coin: not all experiments will be good. Some will be messy. And learning in public means owning the missteps too. Sometimes the best thing you can do is take the “L,” learn, and iterate.

That’s what happened when I started building AI Teammates.

One teammate I built was called “Client Hero.” I trained it on my client’s earnings statements, my call notes, their marketing pushes, and our account team’s success goals. The idea was simple: instead of hunting for information, I’d have a living teammate that knew everything and could help me draft plans, spot opportunities, or coach me on alignment. So I invited it into our account plan project.

But “Client Hero” quickly became too eager—jumping into comment threads it wasn’t invited to. The responses weren’t wrong—but they weren’t invited, and they interrupted the flow of work.

My wake-up moment came when multiple cross-functional partners messaged me the same thing:

“Hey… can you please make your AI teammate stop?”

It was a humbling reminder that learning in public means sometimes learning publicly what not to do.

That micro-lesson reframed how I talk about AI with some of Asana’s largest global customers.

---

The New Measure of Expertise: From Mastery to Curiosity

For most of my career, credibility came from mastery—being the person who knows.

But AI forced a shift: mastery isn’t the starting condition anymore. Curiosity is.

I talk to CIOs, CAIOs, Chief Transformation Officers, executive sponsors, program leaders, engineers, and frontline operators across industries—finance, retail, tech, professional services, hospitality, logistics. And the throughline is always the same:

Everyone is building the plane while flying it.

Nobody I’ve spoken to—no matter how senior—has said, “We’ve fully figured out AI.”

What they do say is:

• “We’re testing fast.”
• “We’re learning out loud.”
• “We don’t have the full picture yet, but we can’t wait.”
• “We can’t afford to be wrong, but we definitely can’t afford to stand still.”

Part of learning out loud meant that I had to let go of the idea that I needed to be an expert before experimenting. The truth is, AI is changing too quickly for any of us to “arrive.” Expertise now is measured by the willingness to explore, to ask foundational questions, and to share learnings as you go.

Asking “naive” questions became one of the most important tools I had.

I found myself in rooms with global AI leaders—people building enterprise-wide strategies, designing governance models, standing up AI councils, deploying copilots, and setting multi-year transformation agendas. In the past, I might have tried to anticipate what they expected me to know and prepare accordingly. But in this season, I started saying:

• “Walk me through why you approached it that way.”
• “What surprised you?”
• “Where did value actually show up?”
• “What are you still unsure about?”

And, internally at Asana, when I shared my own experiments—some polished, some incomplete—it gave others permission to share theirs too. People open up when you show your unfinished work.

Learning in public is a practice. And it’s contagious.

---

A Call to Action for 2026

I recently walked a CIO through how we are building value-anchored, low-risk AI workflows—grounded use cases that utilize Asana to reduce ambiguity and increase clarity. She shared a blunt takeaway from an earlier, failed AI rollout to thousands of employees: she is now willing to replace any application that doesn’t deliver a 3x ROI from AI within a year.

It reinforced what I’m seeing everywhere: even CIOs are learning in public. Everyone is navigating pivots. Nobody has this perfectly figured out.

Her comment reinforced the deeper truth: people don’t need more AI; they need meaningful AI that measurably advances their most important work.

If there’s one message I want everyone—from ICs to CIOs—to walk away with, it’s this:

Just start. But start grounded in value. Start in public. And be willing to fail fast.

Because the people who will go furthest with AI aren’t the ones with the biggest budgets—they’re the ones willing to build, test, learn, and course-correct out loud.

[Sabine_Berger]

I love the take on teaching how to own the missteps and inviting feedback this way! One of my favourite traits of my current team is that making mistakes is not condemned and no one is scared of owning their mistakes, and then we work together to fix them. No one is scared to innovate and no one is scared to say there is an issue. Starting in public pushes this to a new level, definitely worth a try!

[Vanessa_N]

This is such a refreshing perspective, @Hannah_Gaddini! Thank you for sharing!

It also resonates a lot with what we want the Community Forum to feel like: a safe place for sharing and learning. I often chat with members who hesitate to share their experiments, points of view, or proposed solutions because they’re worried they don’t have enough experience yet or that their workflow isn’t ‘perfect’

Seeing you actually model that ‘learning in public’ by sharing your own missteps makes that advice feel real.

And plus one to @Sabine_Berger’s comment! Team culture definitely plays a huge role. A culture of learning together is essential for making people feel safe enough to experiment, learn from mistakes, and share in the first place.

[Ryan_Gordon3]

@Hannah_Gaddini

For most of my career, credibility came from mastery—being the person who knows.

But AI forced a shift: mastery isn’t the starting condition anymore. Curiosity is.

This was quite profound, and I really appreciated this perspective. We can focus so much on the what and the how that we can neglect the why behind it all. It also reminds me of Ted Lasso’s encouragement to “Be curious.” Thanks for this honest and thought-provoking write-up!