A podcast from Aftenposten.
If you don't get this into your companies now, you're finished.
Nicolai Tangen, full of enthusiasm as always, believes Norwegians haven't understood how big artificial intelligence is and how enormously revolutionary it will become. But what if he's completely wrong? What if we're looking at a bubble that bursts right in our face, costing us thousands of billions and ruining our oil fund?
Is it like starting to work super early in the morning? Is this like midday for you?
What's the rhythm like then?
Well, we'll see if we're interesting enough.
No, I try to get going around 5:30, 6:00. Yes.
Exactly. Yes. Have you always done that?
No, it gets earlier and earlier actually. It's probably partly because you get older, but it's also because what's happening in the world right now is so incredibly exciting that if you can choose between sleeping or reading the newspaper, I'd rather read the newspaper.
But do you go to bed very early too to adjust for that?
Yes, I go to bed around 9.
But don't you feel like you're missing out on things then?
No, you don't miss anything in the evening. It's in the morning that things happen.
But is it because you live in different time zones or is it biology?
No, no, I've never been that fond of the evening actually. I think the morning is much better.
The nature is nicer, it's calmer, there's less wind on the sea. It's just lovely, and there aren't so many people. It's quiet.
Well, I get up, and if I'm somewhere with a fireplace, I light the fire, make a cup of coffee, and then I start reading newspapers and emails. And now that I work at the fund, I check what's happened with the fund overnight.
Exactly. Because that was it.
So I get the results early in the morning from what's happened in New York and Asia.
Because I've been wondering about the management of the oil fund, which is supposed to be long-term, and at the same time that website is ticking, so to speak. It's a minute-by-minute count of results, or I don't know how often it actually updates.
And I can get like that if I'm very engaged in a project or the payments for a podcast or everything, I can get into a mode where I have to check it, and then I have to catch myself sometimes and stop just checking the result. I have to work with the content. Do you know what I mean?
You have to, of course. You always have to focus on the process. Yes.
But long-term is a combination of lots of short-term.
Yes. So you have to have results.
But when you sit there reading the news, because I think many people sit and read news during the day and see the big world, and maybe when you wake up there's a pile of push notifications from various international media about what Donald Trump has said or something unstable. When you read those news, do you think consistently in a financial oil fund perspective?
Yes, but not every single article. But when I wake up, I read the Financial Times and New York Times and Wall Street Journal, plus of course I check all the Norwegian ones. And then afterwards I think about what are the implications of this? What are the consequences?
Mhm. And it's so exciting what's happening now, because you can either wake up and read these things and get a bit depressed about what's happening in the world, or you can put on the student hat and just be amazed and think it's exciting what's happening. There's a reason why these things are happening. You just try to understand the reasons behind it.
But is that something you've always done, and is it a personal interest, or if I understand this correctly, does it make your job easier or the oil fund richer, so to speak?
No, but you have to understand these things to be able to lead the oil fund, I think. And I believe more and more that there is a kind of objective truth. And then we have to look at the arguments from both sides to get closer to that truth.
And that's why I'm keen on more discussion and livelier discussions and more disagreement in the fund.
Exactly. Yes.
But what do you mean by an objective truth?
No, I just believe there is a kind of truth. I believe there is a kind of objective truth and...
But can you give an example? What did you not believe was... If you say you believe more and more that there is an objective truth, can you mention a truth that you maybe didn't quite believe before that you think you've found?
No, but it applies to everything. It applies to all discussions. There is a solution no matter what you discuss, there is a solution to that problem that is optimal. There is a perfect solution to everything. And that's what you have to get as close to as possible. And you get closest by hearing arguments from both sides. And that's that discussion. It's what they call the scientific method. And some places are better at it than others. But I won't say what we're best at in the world in the fund, but I think we're getting better. There are more outspoken people. There are more discussions. Scientists are very good at this, right? They have a view. They come to a conference, lots of people say the opposite of them. And then maybe they adjust their view after taking in more information.
But isn't that, it's a super fascinating thought because it's also a bit exhausting thought, isn't it? Because it means that you...
No, it's not an exhausting thought. I don't think it's exhausting at all. I think it's incredibly exciting to try to find this truth.
Yes, but then there are 100,000 different truths. So you have to, in a way, is it possible to optimize any process?
Yes, but right. Okay. We shouldn't talk about politics and such, but clearly, let's take what's happening in different places in the world. There's a reason why it's happening. There's a reason why people vote for one or the other. There's a reason why the great powers aren't in agreement the way they were before. But you can try to understand what's driving this.
Mhm. Do you think that as leader of the Oil Fund, which is part of the financial world, you're a driver in what's happening and not just reacting to what's happening?
No, I wouldn't say we're a driver. I'd say we're too small to be a driver. But we do contribute, of course. We own 1.5% of all companies in the world. In Europe it's 2.5 to 3%. But people look at how we vote, for example. So we have influence on about 3 percentage points of the votes. So almost 5% of the votes at important general assemblies in the world are influenced by us. So we have influence there. But otherwise I don't think we should overstate the oil fund's influence.
No, maybe not the oil fund, but the industry or the world you live in, the financial markets also drive this development and influence the world you see.
Yes, the financial world is influenced by the world and it also influences the world. There's a bit of reflexivity there. It does both.
But do you think it influences in a fundamentally good way?
Yes. Whether finance influences in a good way, I don't know. It's an important part of society because companies need capital to create things, to invest. And this money has to be traded on the stock exchange. And countries need to finance themselves. They need to take out loans. And we own a part of those loans. So we're part of the big whole.
I can't help but explain something I struggle to grasp because I'm not an economics person. I haven't stood selling flowers growing up and felt rich. But one thing, because many people experience that the world is actually going a bit wrong, very uncertain, and yet the markets seem to be doing very well. Why is that?
No, that's a very good question. The markets are doing surprisingly well given what's happening. And we do so-called stress tests that we publish. And it's our fantastic risk department that does this. And they look at different things that can happen. One is a fragmentation of the world. And that's what's happening a little now, that the great powers trade less with each other. That's actually quite negative for the world economy and should have been negative for financial markets. Then you have a number of other things that would normally be negative for the market but haven't been so far. And the question is, why isn't it? Well, I think first of all the big technology companies are doing very well and they have a profit and sales growth of almost 10%. And that drives profit growth in the stock market. So that's positive. And then I think people see what's happening with artificial intelligence and what influence it will have on society, on productivity, on inflation and those types of things. And that's very positive potentially.
But this, we can talk more about this because this is super...
We have to talk more about this because this is the most important thing happening in the world right now.
But just since you brought it up, if productivity growth goes up, won't that also mean that a huge number of companies will go under, and therefore the value shouldn't go up for everyone?
But not everyone goes up. Many are going down now.
Right. And a week ago Anthropic released something called Claude, various plugins that will make it difficult for lawyers, accounting firms, marketing firms. So those companies went down. While Anthropic's value went up, presumably.
But have I understood you correctly that you think we in Norway actually haven't understood how big AI is?
No, I think in general we haven't understood it. It's an absolutely incredible revolution that's happening now. And the reason I say that is I was at a conference here in Norway earlier this year where they talked for 6 hours about poor conditions for Norwegian businesses and 12 minutes about AI.
And I think they should have talked 6 hours about AI and 10 minutes about conditions. Because if companies use AI correctly, it will improve the competitiveness of all these businesses. And when I talk to the leadership of the big technology companies, they say it's a steadily rising curve. And we're not talking about a flat curve. We're talking about a steady increase on the curve. That there's a steady acceleration on top. So the technology curve is almost going straight up now. Like a bad coronavirus curve. It goes straight up and it accelerates because artificial intelligence is helping to create the new tools. So Anthropic says they used two weeks to create these new plugins that are extremely important for many industries. So that just goes faster and faster. You need to do less and less of the coding yourself. It's auto-generating and that means if you use these tools, it has enormous power. And we see it in the Oil Fund. We're one of the leading companies in finance when it comes to adopting these things. We use them everywhere. And we just see what it does with productivity. It's absolutely incredible. And sorry, I'm talking a lot here, but this really engages me. If you can create 20% productivity growth per year by using these tools, then in 2-3 years you're 50% ahead of competitors and they'll never catch up. Right? They're finished. If you don't get this into your companies now, you're finished.
But what's a concrete example of something that can give that effect?
Yes, we have it. We have it everywhere, right? How we trade, how we reduce trading costs, how we can better predict so we can put money in at the right time. We see it in our legal department. We see it in how we monitor companies. We see it in how we use communication and how we use our podcast preparations. We use it everywhere.
Yes, but can you imagine that now, here you can correct me, but for example the Oil Fund, which is very much more complicated, and you can elaborate, but it's about buying right, understanding the world, and selling right. It seems like something that could potentially be largely taken over by artificial intelligence. And if you follow, if you think like you talked about earlier, that there's an objective truth, then it's possibly something you can calculate your way to with enough data.
Now I think there are many who are working on exactly that. And I'm a bit humble about what can we achieve here in the Oil Fund. I don't think we can become the world's best at using AI in that way, because there are so many doing it and trading these markets. So I think more about processes and procedures and that's where we use it. But you could say Norway is actually perfect for using artificial intelligence. We have high education, we're very digitized, and I won't say we hate working, but we're quite fond of being at the cabin. And the combination of this is totally fantastic. And these technology companies say that companies that use this well use it seven times as much as other companies. And countries that use it well use it three times as much as other countries. And now we see initiatives, for example in Sweden, where they're trying to lift up the entire Swedish society. And this is something that the public and private sectors in Norway can do together. That we can really lift our country. It's really exciting.
So you think it should go into almost all processes and then optimize and just push a little. Get a little better here, solve that task a little faster, and then it sets direction and speed for us.
But where do we end up? What do we get out of it?
Well, we get better productivity, we get more done, and we increase our competitiveness against other countries.
Mhm. But does it take the best or worst out of the job, do you think?
No, it takes all the worst. You get rid of everything boring. So I have a kind of dream in the fund that we shouldn't do anything boring. Nobody should do anything boring. So this lifts you up, right? And I see in the fund that those who use this the most think it's even more fun at work. It's just so fun to be part of this. And we make sure they have the best models and the best tools to be able to do this. So it's just so fun.
But wasn't it supposed to end with that investments and stock trading is just a big computer buying and selling with each other? Why do you need people in this?
No, that's a good point. And we think we must have people in this loop. We feel that's important. We want to have control over this. We think it's best when it comes to having control over these things.
You can just feed in an ethics prompt and see if it goes just as well.
Absolutely. Absolutely. You can put in an ethics prompt and it works fine.
Hm. But what's left? How human is it now?
No, it still depends on people. But you could say we've opened something called a technology gap. And that's the gap between what technology can actually do and what the company is able to utilize. And that gap is growing. So technology becomes more and more advanced, but companies can't keep up with how to absorb this. So that technology gap is starting to get quite large. And there are some types of companies that are good at it. I think, for example, in medical research they're very on it. And some other things, but then there are other types of companies that are lagging behind.
Is that an investment question or a mindset question?
No, it's mindset. It's extremely difficult to get people to change how they work. We're creatures of habit. We hate changing anything, right? But I'm sure if you look at your life, how habitual are you? We don't need to go into details on that, but it's probably pretty bad. And so we don't like change, but I love change. And so what you can achieve here is...
Have you, because now, if there's a natural resistance to change, then one way in is to train yourself to use AI also in the rest of your life. Are you someone who uses it? Do you have an AI assistant you use?
I use it, almost all emails I send go through Claude. Especially if it's in English because I don't write English that well. So I always check it, it comes out much better, it looks more professional, right? It comes in the right place and so on.
No, I think that's very good. But could you come home after a long day, around 8 o'clock, right before you go to bed, and you've been tired, and then I'll put on a nice song, and then put on AI music. You know what, I don't even know who made the music I play. According to Spotify, which we've had on our podcast, you have two types of music. You have mood music, which is just, now I'm going to make dinner music. Or you have artist-specific, as they call it, I want to hear Sting. And I'm more the mood type, now I'm going to make dinner music or now I'm going to get up early music.
Does it matter if it's made by a human or AI?
No, it doesn't matter to me.
Yes, as long as it's good music. I mean the same about art.
But why... you have to look at the art and the music. You have to look at what is the result. Is it good music? Then I don't care who made the music.
But could you... yes, right. Because could you have bought... if you could have sold your art collection and bought just an AI...
Now you've already given away the art collection. But...
No, but clearly if it's good art, we have to look at things. We have to see them for what they are.
But isn't there something about the experience of art or music that there's a feeling behind it, or a lived life, or some kind of spirit that you don't get if it just looks nice and is made by a robot?
No, but that spirit is in the art. So if it's not all art that captures the spirit.
No. Right? I'm not that concerned about exactly which spirit it captures. Because I struggle with, I tried listening to some AI music and I think maybe it's just that it's principled, or for my own part I experience it as flatter.
Yes, it is for now, but it won't always be like that.
But what makes a great work of art for you?
No, it's that you're engaged. It's that you don't finish with it. It has to be complicated. It has to be something that makes you not finish with it so quickly. That's what we call wall power. That it has a force hanging on the wall. But just taking AI, which is interesting. Now I think we've spent quite a lot of time solving tasks a bit faster. So for example, what took two days before now takes over 5 minutes. But the whole process doesn't necessarily go any faster, right? Because it's the same meetings you have to have and the same regulations. So now it's about working with workflow. And that's quite a heavy matter to get right. But there are many ways you have to work on this. And then I think to get it into the company, the leadership has to talk about it all the time. And I don't see very many leaders talking about this continuously the way I think they should. Then you have to think about who you hire. They need to be more technological than they had to be before. And with us we have ambassadors who work on this all the time. And we have events and we have mandatory training. So you just have to push something enormously here.
Mhm. But aren't you worried that it goes, I don't know, takes over the world and pushes us off the cliff? Because I feel the AI thing is two-fold. One is that it makes things more efficient, but at some point you say it could create its own consciousness and the whole oil fund...
Absolutely. You have to do this in a responsible way. And we have strict rules and regulations for how we do this. Yes, do it responsibly.
So what do people have left if art and music are going to be made by AI and large parts of work tasks, what do we end up doing?
We end up talking to each other and getting closer to each other as humans. And I think that's important. In a world with more artificial intelligence, we also need more human intelligence and more human communication.
I think about that. But you haven't experienced AI shame, thinking it degrades the process or the email or the meeting?
If people haven't had things taken away, they haven't understood it.
So when I, for example, need documents and such, I want them to be auto-generated. I'm very keen on people using... I walk around the fund, right, on the floor and just, what are you working on now? What do you have on that screen? What do you have on that screen? Almost everyone now has their own screen next to the other screen that they use for AI to run those types of processes. And 100% of employees use these tools. And what's really cool is that now we're almost up to 70% who actually code. Who actually do this. And we have our delivery teams that program and work with the web. We have our own AI team that also works on solving more central tasks. So we're just completely on it.
