AI in Innovation: Flying the Plane While Building the Runway

BACK TO OVERVIEW PAGE

Leaders are attempting to fly the plane while simultaneously building the runway and trying to distinguish between genuine transformation and merely doing old tasks faster.

Christina Habib, Chief Insights Officer of Beauty and Wellbeing at Unilever, draws on nearly two decades of experience and a personal history of constant reinvention to unpack how organizations can move from process replacement to true transformation, build resilience in the face of instability, and preserve human judgment in an age of accelerating automation.

A personal history shapes Christina’s approach to AI instability: 28 different homes across a lifetime. "I grew up in a moving family. We were moving countries all the time for different reasons. I then immigrated. I moved my jobs, and I've mostly also lived in rentals."

This constant state of movement—changing countries, jobs, and rentals—creates a specific type of resilience necessary for the AI era. "It's adaptability. I've made every place a home, no matter how temporary it was. It was the only way to coexist and survive. It gives you resilience, for sure, but it gives you a lot of instability."

We are currently at an inflection point. When disruption arrives, the human tendency is to assimilate old habits into new ways. We use the new technology to do the same things—just cheaper, better, and faster. This is "process replacement."

The true shift happens when we move past using AI for efficiency and begin to realize it has the potential to completely change how things are done.

"It's an adoption curve. It begins with the familiar until the human becomes adapted to it. And then you begin to realize that, actually, I can leverage it to much more advantage than just process replacement."

Similarly to any innovation challenge, adoption is not uniform. It follows a bell curve. There are early adopters with a growth mindset—the "starter disciples" or seeds—who operate with less fear. These individuals prove the concept and eventually influence the rest of the organization.

However, the organizational immune system presents a significant barrier. This mechanism is driven by people wanting to know everything perfectly before acting. In the age of AI, this desire for perfection is fatal because by the time they've learned it has changed again.

Beyond the technical capabilities, judgment is often flawed by fear.

When leaders judge AI tools out of fear—fear of job replacement, fear of being exposed to high-risk decisions, fear of becoming obsolete—the judgment itself becomes biased. In fields like insights, innovation, and leadership, where psychology and subconsciousness play fundamental roles, this bias undermines decision quality from the start.

The solution requires conditioning and education. Leaders must interrogate their own motivations: Are they judging a tool critically because of its actual merit and fit? Or are they judging it because they're afraid it might replace their team, threaten their position, or expose them to risk? The first step is developing integrity in judgment—being able to separate fear-based reactions from objective assessment.

To counter individual bias, Christina employs what she calls "collective genius", or gathering diverse minds and skill sets to interrogate AI outputs and sense-check decisions together.

This doesn't require a decision by consensus, but it does require enough diversity of perspectives in the room. The group applies critical judgment to what makes sense and what doesn't, interrogating the options before deciding.

Christina applies the same principles as a consumer. For example, before putting a high-potency skincare product on her face—where there's real risk of damage—she doesn't rely on her own judgment alone. She asks someone who's tried it. She consults a doctor, a dermatologist, a friend. She seeks sources of trust, authority, and credibility.

The same applies to AI adoption: "We can't assume we know everything all the time. We need to make sure that we're asking the right questions and then arrive at a critical judgment that's well informed."

AI applications that work today might become obsolete tomorrow. Aanother application might become stronger or better. Christina describes this as "AI warfare. It's like nuclear warfare. It's just endless.” Experimentation needs to keep going, because AI doesn't stop developing.

Therefore, the real question isn't whether to keep experimenting—it's what to experiment on versus what to scale. This requires a three-stage funnel that balances continuous evolution with the need to standardize and deliver outcomes.

This occurs in a sandbox environment. The brief is never academic (e.g., "I want to adopt AI"). The brief is always: "I have a business problem to solve, and I think AI can solve it."

Once an experiment is identified as high value, it moves to scale. This is about standardizing workflow processes to deliver outcomes across the business.

This stage identifies unique competitive advantages. Since data is often a commodity ("data is a dime a dozen"), the goal is to create an "unfair advantage." This involves deep partnerships to sustain a competitive edge that coexists with the competition.

The infrastructure of big companies—from procurement to supply chain—is not built for the AI age. Transforming these legacy systems takes time, yet the market will not wait.

The reality at Unilever sometimes feels like flying the plane without having built the runway. This requires a specific type of leadership courage. But asking people to be brave is not enough.

Christina explains that calling for courage without equipping people is both noble and ineffective. She uses the example of asking someone to cross a busy street when they're worried about being hit. No amount of encouragement will remove the fear. In fact, that fear might cause them to cross so slowly and carefully that they actually do get hit.

The solution is counterintuitive: the best way not to get hit by a trolley, bicycle, or train is to move very fast. The call to courage is easier said than done. Leaders need to equip people with their ability to be brave.

Part of this is making milestones achievable and showing that leadership has their back. Christina notes that the more courageous team members come to her asking for bigger, more audacious goals. Her response is to let them pursue those goals with the explicit promise that she has their back if they fail.

Others are more cautious, saying they won't act until certain conditions are met. Christina's approach is to accept that distribution curve—understanding that people have different readiness levels while still moving forward.

A common struggle for organizations is to find concrete examples of AI making an impact, meaningfully influencing decisions, and changing the pace of those decisions. Christina mentions that this happens all the time at Unilever. Three major areas demonstrate this:

Looking forward, the hype and complexity of AI strips away to a single, grounding metaphor: Electricity.

Growing up in an age where electricity didn't exist for everyone, it is now a given. AI is the same. It is a utility. It will be present in every "home" regardless of who the provider is or what the specific tool looks like.

For innovation leaders, this shift in perspective is crucial. It moves the conversation away from the novelty of the technology and toward its utility. The competitive advantage lies in how you choose to use it.

"AI is like electricity, no matter who the electricity provider is, or what home you're living in, it will always have electricity... Be okay with changing homes, they will always have electricity."

The goal is to remain in the driver's seat, using the electricity to power human ingenuity.

BACK TO OVERVIEW PAGE