People laying a puzzle together

Cases

Artificial intelligence should help us think for ourselves

When When we discuss the value of artificial intelligence, it is not enough to focus on the money we can make by making it easier for our  customers to buy and use our products. We must take responsibility for the entire value chain - starting with the principles we use to develop artificial intelligence to the impact our customers have on the world when they use our products. And we have to start, not from the beginning, but at the end.

When I was invited to give a talk at an event titled "Can AI Bring Value to Your Business?", I did not hesitate. I wanted to share my own experience with this important question.

When the host started the conference by calling this headline misleading, I pricked up my ears. Misleading? Yes, he said, of course it should not have been phrased as a question. "It goes without saying that the answer is yes".

None of the other speakers commented on the host's statement, but when it was my turn, the questions came: Is it really a given that artificial intelligence brings value to any business? Is it misleading to frame the value of artificial intelligence as a question? Or is there a point in seeing artificial intelligence as an issue we will never finish deciding on, and taking responsibility for?

Artificial intelligence cannot save the world

Talking about the value of artificial intelligence, it is not enough to focus on the money we can make because it supports customers in buying and using our products. We must take responsibility for the entire value chain. From:

The principles we use when designing artificial intelligence, to

The way our products affect our customers, to

The influence our customers have on the world when they use our products.

If our products make our customers act in a way that is harmful to them or to the world they are a part of, they can be  ‘intelligent’ and valuable to the business only in the short term. But it’s only a matter of time before new solutions are developed that are less dangerous and create more value - and then customers will choose those.

Therefore, it is smarter to start from the outcome. Instead of asking whether artificial intelligence can bring value to our business, we need to ask how we help our customers bring value to the world. That way, it is not technology that determines what influence we have on the world, but vice versa: The value we want to create determines what technologies we do, and do not, use.

Global companies are the answer to the world's problems

Let's take my own business as an example. Qvest is a small tech company that helps large companies develop in step with the world they are a part of. Global companies are influencing, and are constantly being influenced by, millions of people worldwide. Their hundreds of thousands of employees constantly register and respond to changes in the market. Companies who excel at creating meaning and direction across their various employee groups not only have a competitive advantage, they also help solve the world's most critical problems.

Our Qvest answer to the question - how we help our customers bring value to the world - is:  We do it by making it easier for large companies to create meaning and direction across hierarchies (e.g. top management and frontline employees) and key functions (e.g. IT and business). Our digital platform engages employees in different countries, departments, and functions in dialogues about the same strategic topics - at the same time.

This ensures:

  1. That top management bases its decisions on employees' experiences, and
  2. That employees feel ownership of top management's decisions and help each other translate them into concrete solutions

The value lies in employees' questions

At Qvest, the answer to the second question - how our products affect our customers - is that we get more people to ask more questions. Allow me to explain:

The digital Qvest platform is based on 20 years of research into questions. Research shows that it is when we humans ask each other questions that we:

Take a stand on what we think is important

Connect with other people

Commit to a common purpose

Therefore, Qvest is based on the questions employees ask each other when they have to decide for themselves who to ask and what to ask about within a given strategic topic. For our customers, the questions employees ask each other are data. And the questions are at least as valuable as the answers employees provide in questionnaires and interviews. Employee questions show what excites and worries people across the company about management decisions. And by looking at how employees distribute the questions among themselves, our customers can also become wiser about the informal roles and relationships that determine how and how quickly different areas develop.

Unruly data requires human control

With the last two links in the value chain in place - helping large companies unleash the strategic potential of their employees' questions - we turned our attention to artificial intelligence a few years ago.

When hundreds of employees exchange questions and answers on the same topic at the same time, large quantities of qualitative data are collected. One of our customers has almost 20,000 employees worldwide. When they recently ran a Qvest, 670 employees from 14 different business units exchanged over 2000 questions and answers. This is more than 2000 unique data points which, unlike questionnaire data, do not consist of numbers, but of different employees' different ways of expressing themselves.

While a machine can easily and quickly make graphs and network visualizations of how participants divide their questions among themselves, it requires human thinking and analysis work to understand and find patterns in what participants talk to each other about. This work is very time-consuming, and it is therefore tempting to ask if artificial intelligence can help.

The issue of artificial intelligence

And so we finally come to the question I was invited to present on: Can AI bring value to your business? For a company that attaches value to questions as much as we do at Qvest, it is important to address the question of artificial intelligence as a question. But so should all other companies. My research into questions shows that the biggest threat to the value that we create when we ask each other questions comes from technology.

Simplified, the risk with all technology, and therefore also artificial intelligence, is that we forget to ask questions. Or rather: Because technology makes it easier to ask a certain kind of question, such as "What is it?", "How does it work?" and "When can I use it (again)?", we find it harder to ask another kind of question, such as "Do I need technology for this?", "What does this technology prevent me from doing?" and "Would I be better off if I did not use technology in this context?"

The easier technology makes it for us to solve our tasks, the harder it is for us to decide what we think is important; to connect with other people; and to commit ourselves to a common purpose. Therefore, it is important that we do not develop artificial intelligence until we have asked ourselves:

What tasks do we not want to leave to artificial intelligence because we know that it creates greater value when this type of task is performed by humans?

When we discussed this issue at Qvest, we came to the conclusion that the answer to whether AI can bring value to our business is: Yes, but. We want to make it easier for our customers to navigate and prioritize the large amounts of question data they collect on the Qvest platform, but it is important that neither we, nor so-called ‘intelligent models’ make their choices and opt-outs for them. Therefore, not only have we developed a model that supports our customers' critical approach to the machine's proposals, we also have a guiding principle for our AI approach, which is:

Artificial intelligence must help our customers think for themselves.

--

A Danish version of this article was published by Tech Management.

 

Pia Lauritzen

Pia Lauritzen

Co-founder and Chief Scientific Officer at Qvest. Pia is the inventor of the Qvest method. She has a PhD in Philosophy and has spent the last 20 years researching and writing about the nature and impact of questions.

Go further. Together.

Get in touch to explore how Qvest can help get better results in your organization.

Contact us