AI multiplies what a company is able to produce. However, it does not tell you which of those outputs take growth in the right direction. At Nordic Business Forum 2026, digital anthropologist Rahaf Harfoush reminded the audience that in the age of AI, competitive advantage comes from the human ability to think.

I wrote about Harfoush earlier in the blog post Digital anthropology reveals the human impact of technology (in Finnish). Back then, she spoke in the Meta Select training programme about how technology changes culture and behaviour. In her recent NBF talk, Rise of the Thinking Class: The Human Advantage in an Exponential World, the message had sharpened. The question is no longer just what technology does to people, but how it affects companies’ ability to make decisions and grow.

Why do some companies get growth from AI while others just get more output?

Technology choices are decisions about trust

Harfoush began with a familiar example: the Replika app, which lets users create virtual characters and even form romantic relationships with them. Last time, she highlighted users’ emotions: when the app was changed without telling its users, their virtual partners forgot them. This time, Harfoush reflected on Replika’s business model, in which a romantic partner can be bought for five dollars a month. She asked whether it is ethical to build a business that gamifies intimacy.

When you choose how AI serves, recommends or communicates with your customer, you are always also choosing what kind of relationship it builds with your customer.

From a growth perspective, this is crucial. According to Harfoush, we focus too much on the surface-level changes brought by technology, such as processes and efficiency, and underestimate its impact on our culture.

A momentary spike in efficiency won’t take us far if we lose our customers’ trust in the process.

AI makes the strong stronger and the weak weaker

Harfoush’s core argument was that AI amplifies deep expertise but causes weak expertise to collapse. Expertise becomes a competitive advantage that sets apart the organisations that use AI effectively from the rest. Harfoush also referred to a Microsoft study according to which strong trust in AI can reduce critical thinking.

In the Q&A session, she described the phenomenon as the Dunning–Kruger effect on steroids. More output is being produced than ever before, but the ability to evaluate it is not developing at the same pace. The problem is not laziness, but the fact that we don’t always know what we don’t know.

This is already visible in marketing. When every competitor produces content and campaign versions with the same tools, the content starts to look alike. At that point, volume is no longer enough to be a competitive advantage. Growth comes from someone being able to judge which outputs stand out and move the business towards its goals.

According to Harfoush, the responsibility lies with leadership. If an organisation rewards speed and quantity, people will use AI to maximise them. The ability to evaluate output critically may then weaken.

The same applies to young people and new employees entering the field. If all information comes from AI alone, experience and judgement do not accumulate automatically. Growing into an expert also requires mentoring and learning by doing.

Whose goals do your tools serve?

According to Harfoush’s second theme, every tool has its creators’ view of what matters built into it. Herein lies the danger that invisible perspectives guide us without our noticing.

Harfoush gave leaders two concrete tools:

1. Draw up a belief manifesto. Write down what your organisation believes about your people, your customers and your work.
2. Carry out a technology audit. Go through the tools you use and ask whether they reflect your values and culture.

In marketing, media buying, content production and customer experience increasingly run on platforms and algorithms. Do they steer towards the desired growth goals or towards the platform’s own goals? A tool that maximises engagement does not automatically build a lasting customer relationship or profitable growth.

Hypertargeting is a good example. Because technology enables ever more precise targeting, we may focus too much on existing customers and the most likely buyers alone. Even though growth often also requires reaching new customers, in other words marketing to a broader audience.

In times of uncertainty, those who learn fastest grow

The third theme was accepting duality: technology can both help and harm.

Harfoush urged marketers to adopt a researcher’s mindset that combines curiosity, safety and the ability to look at technologies from a sufficiently broad perspective. In practice, this means regularly experimenting with new tools and setting aside time in the calendar for difficult conversations too.

According to Harfoush, volatility is the new normal, and it always creates new opportunities as well. From a growth perspective, this is perhaps the most important message of the talk. When the environment is constantly changing, the number of tools is not what counts. The organisations that succeed best are those that experiment, learn and adapt faster than others.

Intelligent growth needs both AI and people

Harfoush’s message supports our own thinking on growth at Dagmar. For us, Intelligent Growth means smarter marketing and more impactful growth. It is based on combining data, technology and human expertise in a way that helps companies stand out and achieve the results they want.

For us, AI is a tool that speeds up analysis, planning and execution, but it does not make decisions. That still requires human judgement, insight and experience.

In my view, this is the most important lesson of Harfoush’s talk for business leaders. Investing in AI alone does not bring growth. Growth comes when an organisation has the ability to use AI to make better decisions.

Three questions for the leadership team

When there is more output than ever before, differentiation does not come from quantity. It comes from the ability to evaluate, prioritise and question.

I gathered three questions from Harfoush’s talk that every leadership team should consider:

• Do we measure the benefits of AI by speed or by impact?
• Who in our organisation evaluates AI output, and do they have sufficient expertise to do so?
• Do our tools support long-term growth or short-term volume?

The central idea of my previous blog post was that the most important thing is to understand what drives change. Now a follow-up question could be added: are we driving change, or is it driving us?

Harfoush ended her talk with the words of Eric Hoffer, which fit this moment exceptionally well: “In a world of change, the learners shall inherit the earth, while the learned shall find themselves perfectly suited for a world that no longer exists.”

If these questions got you thinking, get in touch. Let’s talk about how AI can be harnessed to drive sustainable growth.