AIAW Podcast

E162 - Bridging US-SWE AI Ecosystems - Minna Sandberg

• Hyperight • Season 11 • Episode 3

Share episode

In Episode 162 of the AIAW Podcast, we sit down with Minna Sandberg, Founder & CEO of Swenode.ai, for a deep-dive into how Sweden can strengthen its position in applied AI by building meaningful bridges with Silicon Valley. We explore the mission behind Swenode.ai and its role in enabling collaboration between Swedish startups, industry leaders, and the global tech ecosystem. Minna shares reflections on the cultural and structural contrasts between Sweden and the U.S. when it comes to AI innovation, the growing importance of “Physical & Industrial AI,” and what it takes to unlock real-world value from cutting-edge technologies. We also touch on Swenode.ai’s strategic collaboration with NVIDIA and Dell Technologies, how to spot high-potential AI startups, and why diversity, ethics, and public trust will define the future of AI adoption. A thoughtful and timely conversation on building not just AI systems, but the ecosystems and values around them.

Follow us on youtube: https://www.youtube.com/@aiawpodcast

Anders Arpteg:

You saw in a newspaper something about using phones in classrooms, right, or what was it?

Minna Sandberg:

Yeah, so we saw the news and we are not allowed to use phones in classrooms and there have been a lot of debate back and forth. If this is good for innovation, is it going to harness the speed of innovation, how we use, use technology, or is it going to make it, you know, is it going to increase the creativity for kids because they get to focus and they.

Anders Arpteg:

But what was? Was specifically when there is a classroom, or was it in general in the school that you were not allowed to use a phone?

Minna Sandberg:

I think it was in school in general. Really I'm not sure, Because I can see it in classroom right?

Anders Arpteg:

I mean, if you have a teacher teaching something, then someone is distracted by using the phone and texting on social media or something. That's not a good thing, would you agree?

Minna Sandberg:

Yeah, I agree, especially in the classrooms, that I think that if you have a distraction it's hard to get in deep focus. You will get distracted by, you know, texting friends. I mean this goes back, for I had my first 3310 when I was a kid and you know I spent a lot of time, you know, texting friends all the time and it was really distracting. So I think that's a good thing. In Menlo Park, you know a lot of my friends have kids in school there. They are not allowed to have phones at school at all.

Henrik Göthberg:

In Silicon Valley. In Silicon Valley, is the general actually restricted?

Minna Sandberg:

Exactly.

Henrik Göthberg:

Even in the TIC capital, of course not.

Minna Sandberg:

Yeah, up to if you are in high school, you're allowed to have your phone, but for younger ages, up to I think at least 12 or 13, you're not allowed to have your phone at all at school.

Henrik Göthberg:

That's more strict than we have today.

Minna Sandberg:

Yeah, exactly, we are usually not that strict in Sweden.

Henrik Göthberg:

My kids strict than we have today? Yeah, exactly, we are usually not that strict in Sweden. My kids they check in their phone.

Anders Arpteg:

When they go in a classroom.

Henrik Göthberg:

The typical thing. The way it works in Värmdokommun is municipalities, more or less. Each school has its own way how they do it. My kids went too. They have a box and when they go, when they come in through the classroom, everybody all the children puts the phone in the box and then when they walk out of the classroom they get it so they can use the phone in school as like in the schoolyard or whatever, but in the classroom setting they check in their phone well, that sounds okay.

Henrik Göthberg:

I think it's quite okay and I think, when you pick this apart, is it the phone or is it social media? Is it the phone or is it the distraction? You could argue we could use the phone as a tool, where we then can use it as a calculator. A smartphone is a tool, right? I think it's the whole distraction that is so hard to control what they're doing on the phone. And so this is one angle. The other angle, I think, is obvious. We really, even in the age of AI, want the children to become more and more proficient in the social interaction.

Minna Sandberg:

Yeah.

Henrik Göthberg:

I can't see a society where we don't interact with each other. So there's there's a purpose of interacting with your peers communicate, learn, learn all the group dynamics and then and then, maybe a toolbox. We talked about this. You know that that that can amplify, or you can have your, your personal companion, and all that as, which is a great tool, but then now we're moving into ai and not only the phone use. What's your thinking about using? Yeah, sorry, yeah. So the reason your personal companion and all that, which is a great tool, but then now we're moving into AI and not only the phone use.

Anders Arpteg:

What's your thinking about using yeah, sorry.

Minna Sandberg:

Yeah. So the reason why I'm so engaged in this topic is actually I mean, I moved into Silicon Valley my first time nine years ago. We created a filter for social media feeds. We did actually machine learning, where we collected, we aggregated data from a lot of different social medias and we made the topic around what you like and what you don't like in five different categories. So it's basically a filter for your social media feed, and back then everything was in a chronologic order, so it was just you know everyone you follow, everything is posted, and now everything is filtered.

Minna Sandberg:

But back then it was a new thing and we worked a lot very closely with Facebook, now Meta, instagram and Twitter, and we were doing this together with them. And what we realized back then was that their algorithms and what they are doing is really doing everything it takes to catch your attention. And now you know I have two kids myself. I try to spend the hours to prioritize what I do when my brain is working the best, which is between 9 and 10 in the morning, and then I prioritize in the afternoon to do like less you know, heavy lifting tasks. So therefore, I'm thinking of you know kids when they are in school. They shouldn't be distracted by the phones that are built to catch their attention. It's an addiction.

Anders Arpteg:

Do you enforce some kind of screen time restrictions on your kids?

Minna Sandberg:

today. Yes, I mean my kid. She's turning four. She loves to watch, you know, Bully Bumpa games and YouTube and such, and she's not allowed to do that more than a few minutes a day.

Henrik Göthberg:

Minutes.

Minna Sandberg:

Yeah, so less than half an hour. Gosh, I'm a bad parent.

Anders Arpteg:

Yeah, I learned, learned your kids is older, right so?

Minna Sandberg:

I actually learned the hard way. So the when she was one um, she watched a lot. We saw it as a really you know easy way to put the. Uh, to put the screen up when we were out on you know a restaurant was what is the first kid you know? We wanted to enjoy life as well, so we thought, okay, just put on some screen time and it's you know who cares.

Anders Arpteg:

It's fine, she loves it yeah.

Minna Sandberg:

And then we saw, okay, she's not picking up the language, she's not learning to speak as fast, and she was really really early. And then when we started with phone time, she just stopped learning new things. And then we soon realized, okay, if and I read, you know, it was a big news article about how hard, how bad it is with screen time for, especially for kids under two years old. So, yeah, I reduced it 100, and only after a few months. So she was yeah, so we took it away 100, so no anecdotal evidence.

Henrik Göthberg:

We want real, real research, real evidence. There was a paper actually recently.

Anders Arpteg:

You don't know that, because it said actually people not kids, but people in general become stupider when they use AI, in this case. Yeah, at MIT study yeah, but it also was kind of nuanced. I mean, it depends on how you use AI. I mean, if you use AI, ai properly I think they said that in the study as well if you use it properly, it actually can increase intelligence and knowledge at least.

Minna Sandberg:

right yeah, I actually read that article. Um, it was uh, they measured their brain activity from people using a lot of gtp or people using it no ai at all, and I was so using no AI at all, and I was so I was surprised because what I see is like I interact back and forth with AI. So I feel like, okay, I'm learning and I'm creative, but the study, even though it's quite small, showed that you are actually not activating as a big part of your brain. So I actually sent it to one of the most famous doctors I know in Sweden, markus Lingman.

Anders Arpteg:

Oh yeah, I know him in Sweden, Marcus Lindman I know him.

Minna Sandberg:

Yeah, he's great.

Anders Arpteg:

The.

Henrik Göthberg:

AI suite of the year, I think.

Minna Sandberg:

Västergötland.

Henrik Göthberg:

Västergötland, region, halland, halland, halland.

Minna Sandberg:

And then he said yeah, even though it's a small study, I think they are right. So I think that is really interesting.

Anders Arpteg:

It is, but okay, so just to close this topic a bit as well. I mean, phone is one thing and I can certainly see if you use it it will really distract. You know the classroom experience and and yeah remove the learning possibilities for kids.

Anders Arpteg:

But and then you know, of course, if you use ai on the phone in the classroom, it's also would be distracting. But what about AI in general? And just to give some more background here, we have a collaboration with University of Oslo and in that university they're not even allowed to use AI to write reports or homework and things like that this is thesis.

Henrik Göthberg:

now we're talking about master thesis or something like this, right?

Anders Arpteg:

Yeah, what do you think about this? So I think we can agree that in the classroom, yeah, probably not a good idea. But what about other type of work, homework reports, et cetera?

Minna Sandberg:

I mean I see why it's so easy to just say that you're not allowed to use it at all, because then you're doing everything as you have always done it as a teacher. But I mean it's a part of everyday life, for everyone right now.

Minna Sandberg:

And I still think that you can see it's not like you will prompt something and then you will get the perfect example from the first time, and by doing this back and forth you will actually create something that is really good in end, and that is real life. When you start to work and when you're is real life. You know when you start to work and when you're uh, in in real life.

Anders Arpteg:

afterwards, after school, you will be able to exactly you should prepare you for that.

Henrik Göthberg:

You need to be prepared for what is going to be like hygiene factor exactly that's why I thought it was stupid.

Minna Sandberg:

When you know it was, a lot of people were really against that. Uh, our prime minister used a uh as a second opinion. That was funny. We can come back to that one we should yeah speak more about that.

Henrik Göthberg:

We can come back to that one.

Anders Arpteg:

But you, okay, I hope you thought the media really misportrayed it badly, right? Yeah, yeah, so weird. I would be more annoyed if our prime minister didn't use AI.

Minna Sandberg:

Exactly that would annoy me.

Henrik Göthberg:

I think that was really showing that we have to type up people in Sweden and that they don't really actually realize how. What is AI and how it's used and why you should use it, and I mean, I thought that was really, in my opinion, we should come back to this, but the debate is skewed in the wrong way, because it's not if, but it's how Exactly. It's that simple right. So of course you should use it, but how should you use it safely and all this?

Minna Sandberg:

Yeah.

Anders Arpteg:

If you don't know how to use it, of course it will be a bit problematic, but if you know how to use it, then the positive benefits is huge.

Minna Sandberg:

My assumption is that he probably I think he's quite tech interested. I guess he yeah, I heard so he's quite interested.

Anders Arpteg:

So I I guess, yeah, I heard so yeah, and actually did a nice uh proposal for a budget for ai for coming five years 500 million per year, but let's not go that there yet. But so many interesting topics and I love the coming discussions with you. But before that, very welcome here, mina. And sunday to the AI After Work podcast.

Anders Arpteg:

Thank you Founder and CEO of Sweenode AI, and I will let you describe shortly what they do. But also you have received like scholarships and prizes from Wallenberg and being a top grade student, etc. So really really appreciate you coming here and giving us your thoughts. But to start with, perhaps you can give a quick introduction who is really Mina and how did you get into the position you have today?

Minna Sandberg:

Thank you so much for having me. I'm so excited to be here. Thank you, yeah, so I started in Gothenburg as a student at Chalmers and I have always been very interested in technology. My brother he built his first computer. I have a twin brother when he was 10. And I have been traveling to Silicon Valley as a kid with my father when he was doing some exchange studies there, when he was doing some exchange studies there, and so for me it has always been very natural to think of the world as a global place. But yeah, I started at Chalmers. I founded my first machine learning startup. We got some really good traction, moved over to Silicon Valley in a small team.

Henrik Göthberg:

Friends at Chalmers or the startup team.

Minna Sandberg:

Yeah, the team was actually. It was me and my brother he's an AI engineer my twin and then his best friend, so that was the developers. And then we had some other friends as the I mean marketing scaling CEO, so we were six people.

Anders Arpteg:

Wow, what was the name of the company?

Minna Sandberg:

Picador Picador. Yeah, and what we did was basically we aggregated data from different social media feeds and we built a filter that learned what you liked and what you didn't like, and then we categorized it and prioritized the feed based on the usage. You could also do some back then cool features like switching accounts and stuff that wasn't doable in Twitter and in Instagram, so you got everything in one app and so heavy users thought that was really cool, because then you could follow so much news on twitter and you can get it all like prioritized based on the like. Uh, so we use the tinder swipe function for like and dislike but this is interesting because it's so early.

Henrik Göthberg:

I mean like now it's filtering and you, we don't even know. We know about the filter bubble and the algorithms is all over, but not then.

Minna Sandberg:

That was the cool thing. So we were invited. When we came over, we won the Chalmers Ventures startup camp and we went over to San Francisco.

Anders Arpteg:

But you've been there before, right With your dad, right?

Minna Sandberg:

That was my first time. So we were invited by Twitter to come and join and sit at their headquarters time. So we were invited by twitter to come and join and sit at their headquarters and, uh, we did everything from like talking to the developers trying to build stuff together, but also, uh, you know, played ping pong and stuff like that, so typical things. Uh started to spend time at facebook and working together with them. What we didn't realize was that they just wanted to learn so much from us Pick our brains.

Minna Sandberg:

And then our biggest input to the app was actually from Facebook. So they one day just shut us down and built it themselves.

Henrik Göthberg:

Really yeah.

Minna Sandberg:

So we learned the hard way but it was a really fun journey.

Anders Arpteg:

Been there, done that by Google actually once yeah.

Minna Sandberg:

And then I become the junior fellow from you know Wallenberg fellow at the Nordic Innovation House. So Wallenberg has an opportunity for one young person a year to be a part of their team at the Nordic Innovation House and work with kind of promoting Sweden, helping with startups and larger companies to succeed and open doors in the US. So I did that and at the same time I studied courses in deep learning near networks and at Stanford and built up like a half a year of courses that was part of my master thesis.

Henrik Göthberg:

So all this is happening before you even taken your master thesis and finished off in Chalmers.

Minna Sandberg:

Yeah, that's true, this is the last year in Chalmers.

Minna Sandberg:

Exactly so. I actually studied at the same time that I was a fellow. I also took courses at Stanford and built my own year. So I did that at the same time as a part of the studies at Shellmers, and then I also did my master thesis at Volvo Cars. They opened up an office in Silicon Valley and they asked around like who knows AI? And people said Mina knows AI and she knows startups. So they wanted to learn how to invest in startups and they failed. So they started, just like I just started, to look at why didn't they succeed. And then we we, we build a process for how to invest in.

Henrik Göthberg:

Was that your master thesis?

Minna Sandberg:

Yes, so you can. Exactly. My master thesis was about how to invest in startups, and in a successful way, and that was kind of the baseline for Volvo Tech Fund, as it's known today. So there were three people when I started and it was 75 when I ended six months later, so it was like a fast growing startup.

Anders Arpteg:

And how? What happened with SphereNode? What was the origin of that?

Minna Sandberg:

Yeah, so I mean, I've always been back and forth in Silicon Valley. So after being at Volvo, I also worked in a Norwegian startup. I worked with larger companies, swedish companies. So I've been both in different parts of the ecosystem, but always one leg in Sweden and one in Silicon Valley. So before I started at Sweenode, I actually was responsible for something called Intek, which is heavy heavy industry in Sweden.

Minna Sandberg:

How to connect that to Silicon Valley, which is heavy heavy industry in Sweden. How to connect that to Silicon Valley, which is the first strategic platform that Nordic Innovation House has from the Swedish team, and we did some really good results. So we had 12 collaborations in one year with larger companies in Sweden within mining and heavy productions. That got really good AI collaborations and three of the companies actually moved to Sweden and opened facilities here. So after that, when Vinnova decided to open a focus area on applied AI, I got the question to drive that initiative. So I'm now the director and driving that. We have been doing it for two years and the aim is to strengthen the connection between Sweden and Silicon Valley within applied AI and to build ecosystems on both sides and our main target groups is fast scaling startups and industry companies.

Henrik Göthberg:

So how is Sweenord funded?

Minna Sandberg:

So Sweenord has funding from Vinnova. It's a strategic initiative, but we also have funding. So it started with it's a three-year initiative. We had 100% first year and then it's decreasing over the years. So we also have funding from our industry companies. So we have an industry program where leaders from big companies like Saab, ericsson, elkab, boliden, and we have a collaboration with Combient, where we work very intensively with industry leaders over years. So we have, for example, petter Perduvar and Elias here in the program. They go to Silicon Valley several times. But we also work a lot here in Sweden on increasing the collaborations with the startup ecosystem and to make sure that we increase the usage of AI in a greater extent in the larger companies that we have in Sweden.

Henrik Göthberg:

Because you have your beautiful T-shirt on you the Silicon Valhalla t-shirt and we can come back to that. But it's both about the connection with the startup community in Silicon Valley. But, as I understand it, sweden is now also focusing more and more to do the same, but in Sweden.

Minna Sandberg:

Exactly so what the aim is not. What I see is that you can get a lot out of Silicon Valley, but what we? We don't want to move Sweden to Silicon Valley. That would be like the biggest failure. So it's more about I see the huge potential in Sweden right now. In January we had a live DMO with Legora and we had Anton Åsike from Lovable and we had Sana on stage and it was the first time ever where I felt like, okay, we have Silicon Valley in Stockholm. It's really happening.

Henrik Göthberg:

And do you agree with Anton's view here that it's about the generation of first level founders, of first level founders of Donny, Liek and Joel, et cetera? And now they spawned to depict. And now you know that's what's happening. Now we see the generation that one startup becoming hyperscale feeds the next generation. I think that's one and this is what's happening and what Silicon Valley is really good at.

Minna Sandberg:

Yeah, that's one part of it. I mean, what I always seen in Silicon Valley is that it's all the the ecosystem is all about all the different components. Like, we have the really good universities, uh, that bring in talent from all over the world. We have the hyperscalers that have. I mean, nvidia has 17 000 people with phd and higher within ai. So it's amazing. We have Google and others and people work around 10 to 12 months in average in Silicon Valley at one company and then they switch. So if you work at Google, if someone says like I work at Google, they're not really proud about it. It's the same as in Sweden. If you say I work at Volvo, it's not like you're proud about it.

Minna Sandberg:

It's more like I do it right now, but you know, I have a dream of doing a startup.

Minna Sandberg:

So as soon as they get a really good idea, they start something, they get investment fast and then they fail fast as well. And that's what I see in Stockholm right now that we have a. Really I mean the inspiration is really good from Sana and from Lavov and others, but we also have the closeness between some early stage funding starting to really attract people here in Stockholm, the angel investors that actually paid forward, but also the mentality of helping each other out all the time and the kind of underground networks that are growing.

Anders Arpteg:

But yeah, let's go, let's get back to the Swedish Valhalla etc and that topic. But can you just elaborate a bit more? What does really sphere node, do I?

Minna Sandberg:

mean how?

Anders Arpteg:

do they make the collaboration better than in some way?

Minna Sandberg:

yeah, so we do organize, uh, or we, we. We are very small, so we co-organize everything with people that are aiming for the same thing, but we have a lot of collaborations with big companies like NVIDIA and Google and OpenAI and others, and we try to bring startups from the Nordics, and especially from Sweden, to be a part of these initiatives. We also organized around last year we had 20 events.

Minna Sandberg:

So we do have a lot of meetings, like meetups in Silicon Valley, but also at larger conferences like NVIDIA, gtc, with a lot of EOP activities, and we actually make sure that both startups and industry companies get a place to I mean, both be shown, because we sit with the agenda together with the companies to see that, okay, how can Sweden be one of the ones that are picked to be showcased? But we also make VIP opportunities. So a lot of different. You know, startups can come and meet the best VCs in the world. They can meet 500 companies. So these type of things which could actually lead to real business, and we are tracked on six different categories. So business opportunities, how many business opportunities it leads to, how many investments, how many strategic investments for Sweden? So we have a lot of measurements and how many people we engage. So we started with around 300 people a year, last year 1,500, and now we hope for 3,000 people that are actually engaged physically in our activities, which is a lot, because we're only like two people.

Anders Arpteg:

And for a startup that's interested in getting involved in this. What should they do?

Minna Sandberg:

So they go to my webpage and fill out a very simple form. Then they take a call with me and I try to assume, in only 30 minutes, try to understand what will take the startup to the next level. So if it's connecting with, if it's a B2B company, we could connect them with opportunities like new customers. We can connect them. We have 50 early stage AI investors in Silicon Valley in our networks. We can connect them to people there but also involve them in all the different activities. We have four or five delegations in October and we work very closely with the big conferences like CS and NVIDIA GTC, so we can make sure that they are part of that ecosystem.

Anders Arpteg:

And it's a free service paid by Binova, more or less.

Minna Sandberg:

Yes, so for startups they don't pay anything. They pay their own travel expenses, which is important because you have to have some skin in the game. But besides that, it's open for anyone if you are ready for it, so that's also a good thing. If you only have an idea, you're probably not ready for the very expensive prices in Silicon Valley. But if you have some type of traction or funding, I think you could be ready for it. So yeah, Awesome.

Anders Arpteg:

Perhaps we can move into the topic we've already been discussing a number of times here, but the term Silicon Valhalla has been floating around more and more, and you even have a t-shirt with that as well.

Henrik Göthberg:

I got one too now.

Anders Arpteg:

Can you just describe what is really Silicon Valhalla?

Minna Sandberg:

I mean so Silicon Valhalla. I think it's a funny term because in Sweden I think what people are referring to is that we could see the new Silicon Valley in Sweden and in the Nordics. So I don't think Silicon Valley is only Sweden. It's actually the Nordics, but it's actually mostly in Sweden.

Henrik Göthberg:

Valhalla is the Nordic mythology.

Anders Arpteg:

Exactly Vikings kind of reference All the heroes.

Minna Sandberg:

That's what you referred to earlier. So we have a few unicorns. We were earlier the most unicorns per capita in Sweden, so that type of I think that refers to kind of the gods in Valhalla as well. But what I love with it is that we are seeing something. We have a movement in Sweden that I, at least in my nine years, have never seen before the pace, forward mentality, the investments and the really fast growing startups. And with Legora and Lovable we can actually see that AI makes it possible to be in Sweden and to scale globally. You don't have to be in Silicon Valley anymore and that is new and we have never seen any investments large investments in companies, without moving the headquarters before.

Minna Sandberg:

So that's really the movement.

Anders Arpteg:

And I remember a quote from Anton Åsike in Lovable. He said, once we got the question about you know, why didn't you go to San Francisco and launch your company there? And he said, yeah, I probably would have received more money there, but I I don't want to do it, I want to stay in Sweden and Europe. And this would be just to hear your thinking. You know, should a startup this would be interesting to hear your thinking Should a startup focus on Sweden and Europe, or should they aim to move to San Francisco or collaborate with San Francisco or Silicon Valley? He said basically that I'm maybe paraphrasing, but he said that if you were to move to San Francisco, of course you would be more or less nobody and it's really hard to compete in the high competition that do exist there. But in Sweden and Europe you're a champ, right, he's a star and he really is right.

Anders Arpteg:

And in that sense, it's actually easier and better and you can actually grow and have more traction if you actually do focus on Sweden and Europe. Would you agree with that?

Minna Sandberg:

I mean, I totally agree, he is the big fish in the small bowl here and in Silicon Valley, as you say, you know the competition is really hard. It's so hard to hire really good engineers and he managed to build a culture around really young people that are open to work, even though it's, you know, vacation time. You know they have hackathons in the weekends, evenings he has. So he managed to really bring the best people in europe to have some silicon valley mentality also back to sweden then yes, I mean a few things.

Minna Sandberg:

That is really special with lovable is that they have a lot of angel investors, like a lot of people, exactly so a lot of people. Um, that is typically the silicon valley way, like bringing people on board, doing it together, pay it forward, do it together as movement, so that I really love that and this really, I think this movement pay.

Henrik Göthberg:

I heard now a couple of times. I was down listening to Anton today. I was thinking that would make sense for this podcast in Kungsregården. Yes, and they talked about that. I think that was even the title. You know, the ending title was and Giving Back, pay it Forward, right. Ending title was and giving back, paying forward, right. I think this is also a big portion of how we need to understand Silicon Valley, or at least I have taken this community thinking, or that we are growing as a community and some like you will. You know we put in some money, sebastian puts in some money, et cetera, et cetera. But it's not only that. It's the communities helping with the network helping, and this is quite cool. What is happening? But when did that start? When did that emerge? Because you say it's different in the nine years. So what happened here?

Minna Sandberg:

In the AI era at least. I mean we have that with Spotify and others. Early, the movement was really. You know the innovation is. We are strong in innovation in Stockholm and we've been known for being one of the most innovative parts of the world for a very long time, yeah, and we recently ended up, I think, in number two in Europe, exactly.

Minna Sandberg:

So we are still like top in the world. We have a very high digitalization. I mean we use digital tools, we're always the try it out market for any product, so we have everything and it's really a nice way of you know. It's so nice to live in Sweden.

Anders Arpteg:

But we are lagging behind in AI usage, I believe, though, compared to other countries in Europe and other places, so in some ways, it's not perfect, right.

Minna Sandberg:

No, it's not, and I mean I think. So what I see the perfect like Silicon Valley Silicon Valley is that we we see a movement right now that a lot of people never traveled to Silicon Valley like really good founders from Stockholm are starting to travel back and forth a lot. So there is also like going to Silicon Valley, meet the best investors, mentors, inspiration and also like other professionals and then bring that back to Sweden. We have almost a hundred founders from the Nordics coming the first week of October to Silicon Valley, together with investors. So it's a lot of movement back and forth right now. And that is what is the cool thing with Silicon Valley we take the best from Silicon Valley but we build here, the build mentality here I think is important.

Anders Arpteg:

Yes, and I guess also if we get the portion of the capital that do exist in Silicon Valley and get that to Valhalla, that could be a really big thing and good for Valhalla right.

Minna Sandberg:

Yes, yeah, definitely. We just discussed, before I came here today, the founder's house. You know mentality on the same street over there. If we have, you know, we should have like a swedish founder's house in san francisco and we should make this bridge even closer because there is so much to gain from having the strong pit forward network of swedes even stronger there in San Francisco and take the best cherry pick but build here. We have so much talent from the universities that we should take advantage of.

Anders Arpteg:

I think it's a really big strength that we have in Sweden and Europe and Sweden especially, I think so lots of talent.

Minna Sandberg:

That's one of the biggest differences. I see, though, that in Silicon Valley, if I tell anyone from NVIDIA, meta, whatever, that I know the best person in the field within AI, agi or you know health or whatever from Stanford, I say, okay, the best person, the best professor, I can set up a meeting, and they will clear their calendar in seconds. Used to be there. So there's really a thing in silicon valley where there's nothing that is has a higher priority than meeting the best researchers, and apple and google and others they sponsor billions of dollars to have a very close connection with Stanford, and Stanford also have. They have a business model where everyone all the researchers need to spend 30% of their time on consulting. That, you know, brings back to Stanford.

Henrik Göthberg:

This is a topic in its own the difference in the academic system and those things of collaborations.

Minna Sandberg:

Yeah.

Henrik Göthberg:

Where Stanford is maybe not unique, but they are very, very good at this, and Sweden, it maybe doesn't work exactly the same. So we can also learn. I'm saying Silicon Valley, how can we get the Swedish universities as part of it? I mean, like, if you really talk about the ecosystem, all these thoughts need to connect if you want to understand the magic of silicon valley, right?

Anders Arpteg:

yes, so I think it's a really good point and I think you know we can learn a lot from silicon valley here in valhalla in how universities work. We have john bosch. You know from john, he's trying to do something like this, but to get both the universities to do consulting, as you say, but also to make the industry and the companies do more academic research, which all these kind of tech giants do to an extreme. Yeah, it's something that we can do more.

Henrik Göthberg:

I love your anecdotes. There's nothing more than if you can talk with the top-minded researcher. Clear a calendar.

Minna Sandberg:

Yeah.

Henrik Göthberg:

I mean put that in the Swedish context.

Minna Sandberg:

Does it really work like that? No, no.

Henrik Göthberg:

Call McKinsey.

Goran Cvetanovski:

Call Gartner.

Minna Sandberg:

Exactly.

Henrik Göthberg:

Why the fuck do you call Gartner when you can go to Jan Bors? Yeah, in reality you know.

Minna Sandberg:

Yeah, true, the truth is both ways, because if you talk to a professor in Sweden, they're like no, I don't want to work with industry. It's not like yeah right, it's not. It's not the academic, uh, exactly so so some pride of being doing research without any connection to to real world.

Henrik Göthberg:

Uh, examples I'm doing foundational research, like basic research is uh, the most, the most beautiful thing, you could do as a researcher in sw, but it's totally the opposite.

Minna Sandberg:

We work very closely with Stanford and they have studies paid by Apple, by Google, and we work together with them. But in Sweden that is something different.

Anders Arpteg:

It would be amazing if we could learn from that what a rabbit hole we have here.

Henrik Göthberg:

Let's not go into this one.

Minna Sandberg:

But we should continue.

Henrik Göthberg:

I mean like there are so many angles here, should we move on? Yeah, sure, okay. I mean like, so what is this with the investment in collaborations with NVIDIA and Dell Technologies, and how you? You know something that we sort of picked up on, but you know, could you clarify this sounds very cool, very interesting.

Minna Sandberg:

Yeah, so that was one of the companies from our industry network yes, uh. So iron is uh iron yeah, so that is the. It's one of the companies in our network that we work very closely with. They are opened a factory in a data center or a factory in uh, lean, lead, lead shopping actually not lean shopping, that's the funny part or AI Factory in Lidköping actually Not Lidköping. That's the funny part it's Lidköping.

Henrik Göthberg:

And that was big. It is made in normal news. Yeah, it's big news in Sweden, of course it was big news because it's a big amount.

Minna Sandberg:

It's a big amount and they work very closely with companies like NVIDIA and Dell. They co-invested and they also have some really interesting collaborations with OpenAI that will be announced soon, I hope. So some really cool things. But the niches for them is that they are building safe clouds. So they really, instead of doing on-prem, they should do safe. You know safe clouds within Sweden. That's their strength, so they are part of our industry network, meaning that we travel to Silicon Valley back and forth for a long time. They have a lot of customers in the US, but now they could be the safe alternative for Swedish industry as well.

Anders Arpteg:

They build on top of the American clouds then, but have a safe version of that, or how does it work?

Minna Sandberg:

No, they have the data centers in Sweden and instead of installing on-prem at facilities, they have their data center there in Lidköping, and they will open it in other places as well.

Henrik Göthberg:

They have one in the us as well yeah, so together with the nvidia and dell, you know if you look at the hardware exactly and then the software and the.

Anders Arpteg:

You know the, the top stack that you need to have on top of the hardware in the data center I mean, they do, um, work very.

Minna Sandberg:

They work with anyone, but they do work. Of course, the software stacks that nvidia have they are in close collaboration with them they can use. But they do work with the software stacks that they have. They are in close collaboration with them, they can use them. But their niche is that they want to be very price competitive. So you could get much more using their software and hardware than any competitor.

Anders Arpteg:

But is the cloud service mainly for AI or is it more general kind of cloud service for software Mainly for AI, or is it more general kind of cloud service?

Minna Sandberg:

for software, mainly for AI, and one big thing is defense, of course, but there is other areas.

Anders Arpteg:

But sometimes, you know, I get a bit annoyed with this and I'm sorry to say that. But we've seen a lot of investments from Europe and EU Commission and also Swedish investments in AI specific and and also research-specific data centers or HPC clusters, where they say, yeah, you can train your model here, but you can't do anything else, you can't serve it here, you can't use the inference here.

Anders Arpteg:

No, you can't do the inference here, then you have to move to some other place, and if you need hardware for other things than AI, then do something else, and I think that's not really how a company should work. I mean, they need to have everything integrated and it's not like you train once and then never do it again. You have more of an agile iterative loop here, where you know you use it in production and then continuously improve. And how could you switch between a cluster that is only for training and only for AI when you have the cluster otherwise for the rest of the software and for inference purposes elsewhere? To me it doesn't make sense.

Minna Sandberg:

I think that comes from that we are an industry country In Silicon Valley. It's like software born Software is from Silicon Valley, so they would never even think of that in that sense right exactly but we are in, we are a production country. We are. We are producing a lot of really strong things and we should be proud of that. But you can't think of software as you think of building a car or you know producing something, because it's not like three or eight years of a straight line of it's not a waterfall, exactly.

Minna Sandberg:

That's not how it works, but we try to do it in the same manner, and that's why it's not efficient or it's not even working.

Henrik Göthberg:

But is Iron's business idea a little bit different here? Because you're referring to the HPC setup a lot and the academic setup which is literally we can prove it works, but then you need to figure out how to operate it. We don't do that. Or is a iron different here, like because I can see how they now trying to be do something where they want to become the production environment?

Minna Sandberg:

yeah, they want to have exactly. So they want to offer, uh, compute as a service. So they want to offer compute as a service. So they want to be. You can use it whenever you want. That's the big idea, the big announcement they did on Almedalen. I mean, I don't work, I'm not a part of the Iron team, but I think what they are trying to do is, um, doing something different, which is always, you know, it's always good. How, if, if it's going to work, we'll see.

Henrik Göthberg:

Yeah, but and now circle back to what is swia nodes, uh contribution in this, or what you know so is we're not talking about them, but we're talking about you. So what did you do? It's a good example.

Minna Sandberg:

So what we do with them is we bring them closer to the Silicon Valley ecosystem, and also together with the leading industries in Sweden. So they are part of this movement. We travel together to Silicon Valley many times a year. They were actually at NVIDIA GTC. They got to meet Jensen and had some oysters with him at the party. These type of opportunities.

Henrik Göthberg:

So we promote them in the global setting, which is a big deal. It has been a journey, and being part of this community and Silicon Valley life you want it has facilitated that, this journey. You know it's not the journey, but it's facilitating the journey, yeah, is that a way to?

Minna Sandberg:

yeah, exactly, they got actually I don't know how much I'm allowed to to to say, but they got some really good businesses, with a few of the biggest companies we have in sweden from this. Uh, so so working actually back to Silicon Valley, learning from Silicon Valley but actually applying it in Sweden.

Henrik Göthberg:

So it's not only about the facilitation as such, but it's actually creating an arena where matchmaking can happen for real.

Minna Sandberg:

Yes, yes, and we see a lot of the companies that are part of our AI Connect program, which is for scale-ups and startups. They have I follow up frequently because we have you know, we're measured on KPIs and we see that when they something happens when you travel together, right? So when we take both the bigger AI leaders and CTOs and CEOs from companies and then we bring the startups and we do things for example, the bigger AI leaders and CTOs and CEOs from companies and then we bring the startups and we do things For example, we brought around 35 startups and 25 startups larger companies to Paris before summer. With them, as a part of we built a Swedish event at the GTC conference in Paris. We saw that right away, businesses started to happen between the startups and the bigger companies, and we do that in Silicon Valley and so bringing them together in Sweden and abroad, that creates the opportunities. That's the ecosystem thinking here.

Henrik Göthberg:

Great.

Anders Arpteg:

Awesome, and I think you said something interesting before, which is you know, of course you want to increase, with SphereNode, the collaboration between Silicon Valley and Silicon Valhalla, but in some sense you don't want to make Valhalla the valley either. So perhaps you can just elaborate a bit more what are really the differences and what should we potentially try to bring to Valhalla and what should we not, so to speak?

Minna Sandberg:

Yeah, I mean everything is not better in Silicon Valley. I think we have some really good things in Sweden. We have a lot of talent coming from universities that I hope to bring closer to the AI ecosystem of builders Lärarundatagen. That's unique that you can actually own your own IP. That's rare. That is not happening in the US. Stanford owns 2% in all the companies that come from Stanford, which means that they are the biggest fund in the world.

Anders Arpteg:

So people employed at universities actually can own their own conventions companies that will come from Stanford, which means that they are the biggest fund in the world. So people employed at universities actually can own their own conventions more or less in Sweden, but that's not usually the case in Bali.

Minna Sandberg:

No, I think it's unique for Sweden. So Sectra, for example, in Linköping. They are a huge company. They built it as researchers on paid time and now they own it themselves. So that is unique and it's actually really an opportunity that we should take advantage of better in Sweden. So deep tech really, I mean also the cost of living. It's extremely expensive in Silicon Valley. I mean, a two bedroom is 100,000 Swedish crowns, so it makes sense to be here with the kindergarten, with school, uh, for free and everything. So it's really a good way of living and it's not, um you, you have a lot of opportunities here, um so, but I what I think is the best is to to have the possibility to take advantage of the network there, building relationships and then bring and build back in the Nordics.

Henrik Göthberg:

And you said already before, if we simply push the Swedish companies on a highway to Silicon Valley and they leave, that's a failure, because the real thing is to build our hub here at the same time. I think so, I think so too, if we can't.

Minna Sandberg:

and there is a lot of really good programs like BASB, for example. They have people who are two years at Stanford and then they get the opportunity to move back For people that know, basb is this big research program in Sweden.

Anders Arpteg:

That is, I think, about six billion Swedish crowns, which is one of the biggest, if not the biggest, research program in Sweden. That is, I think, about six billion Swedish crowns, which one of the biggest, if not the biggest, research program we have, and they have a big focus on AI, of course.

Minna Sandberg:

Yeah, they also have a postdoc program where you can study two years at Stanford and then you can travel back to Sweden and get paid two years to find a job back in Sweden. So it's really, really an opportunity. So, yeah, wallenberg are doing a lot for Sweden, which is great, but also for the big companies, so a lot of the really big and famous Swedish brands that we have. They can get so much out of using AI in a greater extent. So traveling to Silicon Valley but also getting a closer connection to the ecosystem here back in Europe is an opportunity for Sweden that I think would make a big difference when it comes to physical AI and the usage of AI within heavy industry and manufacturing.

Henrik Göthberg:

And I think this is a good segue because if we started with Silicon Valhalla and we started very much focused on the startups you have done a lot which we would categorize as industrial AI, or even physical and industrial AI, and which is really also about really what I think VASP is all about strengthening the really long-term standing industries and companies. In Sweden we don't want to be disrupted by startups all the time. We want to make sure that our great brands stay healthy and in that sense I think, have a less disruptive world for most in Swedish society. But could you elaborate on the difference and what it means when we say we want to focus on physical industrial AI and maybe how that differs in approaches?

Minna Sandberg:

Yeah, and I think examples is a good way of talking about it. I see here I mean, for Sweden, the biggest money is really what in what you talk about here? The big companies should stay healthy. I mean it's really really important with scale up and being the building that ecosystem, because that is what brings innovation. I think it's rare that big companies create something good internally, like I don't.

Minna Sandberg:

I don't believe in it because we have a history of building uh, hardware, not software, but what we have is that we have a history of building hardware, not software, but what we have is that we have a lot of domain knowledge, and what is not very common in Silicon Valley is actually production or anything like that. So if you are a B2B AI startup in Silicon Valley, the only thing you need is Sweden.

Anders Arpteg:

You need to apply your.

Henrik Göthberg:

This is our USP we have a USP here to export to them.

Minna Sandberg:

Yeah, and it's really unique. So if I come with the strong brands that we have in Sweden and propose a collaboration, that's the only thing they need to get. Their next investment round is to prove their technology with a strong brand, which is what we have. So there's a lot of opportunities.

Minna Sandberg:

But if you look at from the Swedish side, for example, shipping Stena have done a really, really cool project together with Google DeepMind, where they could analyze satellite data and decrease the amount of traffic all over the world, and they save millions by doing so. So and it saves. It saves on the environment as well, because they don't have to travel or they don't have to wait outside of harbors, because it's so much better optimized. When it comes to logistics, we have one of the largest logistics companies or tracking companies here outside of Stockholm. They have tracked the logistic inventory so they could decrease up to one billion spare parts just by knowing when something is coming and where it is to decrease the amount in all the different steps in the logistic change. So that is like a few examples of how easy it is to actually optimize using AI.

Henrik Göthberg:

And here we have an AI divide the companies that were great, but they don't really are great at software or machine learning. They haven't really embedded those capabilities in their companies yet, and this is, of course, how they stay healthy.

Minna Sandberg:

They have to, yeah, exactly.

Henrik Göthberg:

And this is the challenge, this is what I work with every day To reform a company that is a physical company to have software in their DNA. It's not the same as AI and software sugarcoating right.

Henrik Göthberg:

So now you need to start with collaborations and you need to simply learn how to operate like a software company. I guess and I can sort of see how we can foster that or how we can support that. In my opinion, this is how I interpret industrial AI that we need to not only work with the innovation, the invention, but also how to reform them by learning how to do it.

Minna Sandberg:

I mean, you can do a lot of really cool POCs or pilots, but that will not be anything.

Henrik Göthberg:

Exactly.

Minna Sandberg:

We have that and we have done that, but that's why I brought so many delegations to Silicon Valley. So I would be sick of even counting them. But what I realized after a while is that I'm not doing that anymore. I will bring them into a one-year program and you have to be senior and you have to be C-level at the company, Otherwise you're not welcome. The reason is because it takes time to change the organization. You have to think long term. You have to travel back and forth uh, for many times. You have to interact with daily, uh. So you have to have experts that you build your own network and interact daily with this.

Henrik Göthberg:

You know experts from silicon valley to really, really understand what's all about I shouldn't talk with my voice, but it really excites me what you're saying now, because we can't be invention focused. We need to be adoption focused. We need to basically think about, okay, the people that we're now, that experiences this need to be people who then can go home and think about what they need to reform as well.

Minna Sandberg:

Yeah.

Henrik Göthberg:

Which is like if you only meet the tech guys and you know it's super cool, but it cannot be, it ends up on the prototype graveyard Exactly All this. My summarizing it.

Minna Sandberg:

And I'm thinking I mean, I have been thinking maybe too much on this, but I think that you have to have someone accountable at the company. That's really the big thing. So, instead of just like okay, you have to have someone that feels like it's my job to change this entire organization to make it work to scale AI in a greater extent. That's my responsibility, I'm accountable, I will deliver that, because if you don't have that, no one will take that responsibility and you will have, like you know, ice cream and balloons.

Henrik Göthberg:

But this is the big problem, right, you have someone responsible for the tech and it ends up as a prototype and they got you know the innovation theater galore. We did it, we built it. It is great, but it wasn't their responsibility or accountability to adopt things at scale, because that's a different story altogether. So what you're saying now, I think it's some big learning from making these networks work, but it's a big, big learning to take home to anyone who wants to innovate with data and AI in an enterprise setting. You need to think like that.

Minna Sandberg:

Yeah, exactly, and you have to be serious about it. Yes, and you have to put a lot of funding into it. But I mean, after the vacation I went to Stockholm for one week to meet with all the really good founders that I have been following for a long time and kind of picked their brains on what can we do better in Sweden? And the B2B startups I met with they said that for us we cannot lose the time it takes to work with Swedish industry and this is very sad, but I mean to do a proof of concept for like 500K or 1 million Swedish crowns. It doesn't make sense. It takes so much time and then after that they have no plan or even intention to scale it. You have to have a scale first mentality and the day you sign the pilot you should already think of the next step. You should already have a scaling plan. This should only be the opener.

Henrik Göthberg:

This is the basic think big, start small scale fast.

Minna Sandberg:

If you don't have the think big in front of it, maxine, it will not work and that's why the few B2B startups that are really good in Stockholm, they have the US and China and other places as their customer base. They don't even look at Swedish companies. But we can be better. So that's why I'm kind of bringing these groups closer together because they should know of each other and it. If we, if we make that happen, we will be.

Anders Arpteg:

We will be silica valhalla, but we have to take advantage of, you know, everything we really have in stockholm it's strong, and in Sweden so we need to learn a bit about the grandiose plans that perhaps the Valhalla people is no, sorry, the valley people are doing and bring that to Valhalla. I guess in this case yes.

Henrik Göthberg:

Before we leave Valhalla. What's your trajectory, what's your idea, what's your dream and what? Your idea? What's your dream and what is more concrete? Where are we going? What are you going with Swinode, but also the vision? What's the short-term, mid-term, where? Do you want to take this. What do you want to be part of building?

Minna Sandberg:

So I want to for Swinode and for Sweden and the Nordics. I want to position the Nordics as the leading AI region of the world and I want it to be in a global context. I want the entire world to see what we are seeing, and the step towards doing that is starting to think Nordic. I don't think we are big enough in Sweden so connecting all the really fast fast growing networks in the Nordics. There is actually a lot of drive from the VC networks in Norway, in Copenhagen, we have started to see a little bit what's happening in Stockholm. We're starting to see it there. We have in Finland a few initiatives that are really groundbreaking. So I'm putting a lot of effort into doing it in Nordic and also to use the big playgrounds and the connections that we have to show off what we have in the Nordic summer global setting.

Minna Sandberg:

So one thing is, for example, at the big nvidia gtc conference in santa clara, we are bringing uh. Last year we had around 100 companies. Now we're hoping for 800 companies and people from the nordics and we are showcasing success cases as a building up, a momentum prior to this. So that is the moment that I see uh. I can't do it myself, I will have to connect with all the different, uh, people in the nordics that are actually building, actually doing stuff, but we, we, one of the problems, uh, I think, here is that we have so many small groups that are not connected. Yes, and just connecting them will will make it so much stronger, is my belief.

Henrik Göthberg:

Maybe that's a good, isn't that one of the pro? You know what? Okay, this is all fantastic. What's the blind spot? What's the problems? I think you said it the disconnect between the different things yeah it's one of our problems, because it makes us way weaker than we could be exactly.

Minna Sandberg:

So we have to and we have started to see that kind of mentality by here in Sweden that is not really Swedish which is kind of introducing people to each other and make this kind of take an AI coffee a big thing, or you know this initiative.

Anders Arpteg:

I mean, so many people say that we are so good in collaborating in Sweden and the Nordics and I don't buy it. I don't believe it at all.

Minna Sandberg:

Sweden and the.

Henrik Göthberg:

Nordics, and I don't buy it. I don't believe it at all.

Anders Arpteg:

We're super fragmented we're fragmented as hell. So we need to fix that right.

Henrik Göthberg:

This fragmentation piece is so obvious. I mean, like you, even see it down in how we invest in AI factory and AI verkstam and it's even funny, right. And of course, on grassroots level, I'm and others are working to push them together. But to build a community, to put it together, to connect the dots hard, focused work and it doesn't come naturally right. And in this sense, I don't think we are clicky Now. We are clicky Even going to a bar in Sweden. We are clicky compared to Australia, yeah, yeah, of course.

Henrik Göthberg:

You can see it in the bar mentality yeah, and.

Minna Sandberg:

I agree, we don't. I mean, but we are very, we really like to be good at. You know, we really like to be proud of Sweden, so that is kind of what will bring us together and also the opportunities. I mean, there's a lot of buzz around you know, these fast-growing startups that we see right now. People want to be part of it. The fomo is definitely there. I have never seen as many startups and founders traveling back and forth to silicon valley. Uh, that as it is now it's like like every week we have something happening. So it's definitely like it's not about me at all. It's about all the different people that actually they want to build this, they want to do it.

Anders Arpteg:

But you do amazing work, I must say, Mina, and I think we have all the preconditions really to make Valhalla a really great place and we've done and seen so many companies actually do succeed here. But I think we can do more and if we combine the positive aspects of the valley and bring it here, who knows what can happen right.

Henrik Göthberg:

But I must also applaud, like Silicon Valley. Halla is interesting, right, ida puts out the logo. Someone was trying to trademark the logo, or you know, and it was by the domain. Oh, by the domain, sorry, not trademark by the domain. And it was an outrage on linkedin. No, this is not the fucking point exactly, and maybe that's the whole thing. Right, sometimes I feel like we are starting something, but but at the same time, it's very important for each of whoever is starting to be the top dog and all of a sudden, we are not promoting each other if we are not the top dog. In this context, and what I feel with Silicon Valhalla is this is the first time in Sweden where it's not so obvious that someone is trying to be the top dog in a certain ecosystem.

Henrik Göthberg:

Exactly Because the true ecosystem doesn't have a top dog. Do you see what I mean?

Minna Sandberg:

Exactly, but I think it's I think back to Lavables and others like if a lot of people invest in it and not only I mean they could invest in time, they could invest in being a part of the ecosystem or actually be an angel investor, Like if that's I mean I mean in silicon valley people talk about this a paid forward mentality. Yes, it is, but everyone gets something out of it. Right, it's not like people help each other, but they do it because they know that that's expected and that's what everyone else is doing and that's why the ecosystem is so fruitful and why it works in a good way. But they always want to get something out of it, and that's okay.

Minna Sandberg:

It's okay because it should be, that. That's how it should work. So in Sweden now, when a lot of people say, oh, it's like 100 people that are investing in this company and super successful and I can be a part of it and I want to do it, then I mean people will engage and be a part of it. You can now. Anyone can build right. Yeah, will engage and be a part of it.

Anders Arpteg:

You can now anyone can build, right, yeah, so so why not cool? And with that it's time for ai news. It's time for you by ai podcast. So we usually have this kind of break in the middle of the podcast to just share some thoughts about recent news that we've heard and read about. So let's try to do a single or some few stories each one of us. Mina, do you have something you'd like to talk about?

Minna Sandberg:

um, yeah, so from big news, uh, I think, for example, the um, there had been a lot of discussions in Sweden on how much money we put into a few employees in Silicon Valley and the hype of that. I think that is really funny and it's a big hype. And, yeah, what's your thought about that?

Anders Arpteg:

No, but I guess you're referring to Meta then.

Minna Sandberg:

Exactly.

Anders Arpteg:

The hundreds of millions in signing bonuses etc. That they are pouring into a few or a single person, and it might sound super strange, but I heard you had more details about this, so I really love to hear it.

Henrik Göthberg:

Yeah, you spilled the. Silicon Valley gossip that sort of clarifies what this is really all about.

Minna Sandberg:

Exactly. So, friends, at Meta they said that oh wow, yeah, this is really good news. I mean, it's a hype, it's not true. It's basically the IP, and I mean what they built is more like okay, you're thinking of, like, buying a startup, buying the technology, but also what the person know about and that shouldn't come out or shouldn't leave the company. So it's really, it's really juicy, but yeah, that's typical.

Henrik Göthberg:

It's really juicy right.

Minna Sandberg:

And the typical Silicon Valley that you want to build a news around something that is actually smoke. So sometimes you shouldn't believe everything you read.

Henrik Göthberg:

Yeah, so. So. So in plain English, someone had done something really fucking smart, right?

Minna Sandberg:

Mm-hmm.

Henrik Göthberg:

At another company, and how do you get a hold of that Exactly? You buy the guy. You buy the guy Exactly.

Anders Arpteg:

Or the girl.

Minna Sandberg:

Thank you. Yeah, exactly, you can buy, in this case, a really young person. People think, wow, he must be amazing.

Henrik Göthberg:

Yeah, he's amazing, but that is what he did Exactly. He has done something and they want to do exactly that algorithm.

Minna Sandberg:

Very true.

Anders Arpteg:

And just to give some more background on this. You know Meta went on a poaching spree, if you call it that. They tried to get people from OpenAI, google, amazon and so many more places and just give them insane salaries and signing bonuses, but some of them actually went back.

Henrik Göthberg:

Some went back, some refused. But obviously if you're in the Valley, people in one company know which engineer cracked it. And okay, what tinkering did he do? We don't really know, but he knows. Okay, we need this tinkering. It's worth billions if we could do that tinkering with our system.

Minna Sandberg:

Yeah, is that?

Henrik Göthberg:

simple right.

Minna Sandberg:

Yeah, exactly. I mean it's always been a thing with 10X engineers, like of course that is a thing, but now we're talking about like 100X.

Henrik Göthberg:

It's 100,000X. Maybe on somebody's top that's right.

Anders Arpteg:

That's an alternative to traditional. We call it that industrial espionage.

Henrik Göthberg:

You were saying that let's not go there Plus you know, it happens.

Henrik Göthberg:

But because we're in the world of AI and software, someone has figured out the fundamental way to tinker with something. I mean like we have all this hype, so many scaling problems, so many fine tuning problems, and if you can get them right, right, and in the end, some of this stuff, even with the transformer if I take the transformer, it's not that, not that super diff. I mean like when to do it, first, right, but now, when we have learned about, it's not that we, we know what it is.

Anders Arpteg:

So all these things is happening all the time yeah, you know, the difference is in the details, you know. So what's really published? That's one thing, but really to make it work you need to have details, and that's usually embedded in the people and the one that actually worked with it and used for real.

Henrik Göthberg:

And in the Valley people know exactly who did it.

Minna Sandberg:

Exactly, I think I mean that's probably the change that I'm seeing in Silicon Valley right now that people, they're really good developers and they're the, they will, I mean, have a crazy, uh um, amount of you know. Their salaries will be crazy, and and I mean the, the tech guys is still a thing I mean they do, they do hang out, they do see competition as a good thing, they spend time that's why it's so expensive in this area, but uh, um, I mean, then it's a lot of smoke as well yeah, but in the end this is a simple roi.

Henrik Göthberg:

If we can get that guy and he can fix that, how much can we say? What can we do with that?

Anders Arpteg:

yeah, I see that I see goran put up some stats here on the salary in silicon valley versus sweden. I don't know. I want to read this even.

Minna Sandberg:

I mean it's kind of sad, I mean one funny thing is that I have a few friends that moved to Silicon Valley because they sold their startups and they got some in Swedish amounts crazy money. But the thing is that now you know they live in a quite small home, the standard is lower I mean much lower than we would have in sweden and they pay like six thousand dollars for each kid at kindergarten and then they are, you know, working 13 hours a day. So it's like the the standard of life is. Since everything's so expensive, they will be super rich if they move back to sweden, but when they are still a part of the, this ecosystem, it's just like numbers rolling in and out on your account and anyone that's just playing the, the money game.

Anders Arpteg:

I think they they're going to lose in one way or form anyway. I mean if you just want to make money, you can go on the street and sell drugs, I mean and that that's not really I mean anyone that's just going for that.

Henrik Göthberg:

It's not a good game to play yeah, but let's see on the news, what's the next news? Yeah, any other news, and do you have something, or should I? We need to mention sauna, and, uh, you will. And then, and even here we have a scoop, you, at least you.

Minna Sandberg:

You met with matias quite recently right, yeah, like today one hour, one hour ago right so we have the latest, freshest on the sauna, you know.

Henrik Göthberg:

So, without saying anything, what's your take on this? You should explain it. Oh, you set it up, or you set it up and you do it better. No, please go ahead.

Henrik Göthberg:

Okay, I mean, like so sauna, just um, sauna just uh, got acquired by workday. Workday is, uh, you, in my opinion, erp system. Are they more HR? It doesn't matter. It's one of the big software vendors who sells B2B software for big enterprise and you use it in order to manage your flows within your organization transactional flows, mostly. Erp is a good example of this. So Workday is a competitor to SAP in some ways, right. So they went in and they bought SANA For $1.1 billion, right, for $1.1 billion, right, yeah.

Minna Sandberg:

I had that coming because I saw a few things that I think. So, before the summer, sana was announced as a big collaboration with NVIDIA at NVIDIA GDC Paris, and we saw the same with Silo. Like a few weeks before they was acquired by AMD, they closed a really big deal with NVIDIA, and NVIDIA is not only and the only case, of course. They closed a lot of other really important deals and did a lot of publications about it, and you see that you know the patents when people have a focus on doing some really cool publications and then somehow they got bought. So I just discussed that before the summer. That was my guess, but I think it's like it's really important for sweden, because now the what happens is that we bring that you know mentality back to all the other builders, uh, all that you know people thinking of doing something like this. It's really important for to have, uh, someone to look up to. So I it's really important to have someone to look up to, so I think it's really good for Sweden.

Henrik Göthberg:

But it's an interesting one. Now, spotify went one way headquartered and become Spotify Klarna. They put themselves on Nasdaq. We can talk about that. They went to Nasdaq not long ago, exactly, and they went that way. And here now we have a proper exit. It's a little bit different. An IPO is an exit as well, isn't it? Yeah, of course it's an exit. Okay, sorry, they're all exits, but the IPO way versus now being acquired by what I would call a more traditional software vendor. Have we seen many of them? Spotify did the same as Klarna, right? Yeah, the IPO way On this magnitude, I mean.

Anders Arpteg:

But you can think, you know, is this good or bad? If you take Silo AI or Sona Labs, you know getting acquired by a big American company and in some way it's good because you know it gets a lot of capital into the Nordics in some way we need to say we have Sofie Nass.

Henrik Göthberg:

She was here on the pod. She's one of the early. Do you know, Sofie? She's one of the three co-founders, so she was in very early as the first marketing person in. Sona the first marketing person. She was here, so we've had some. We have Anton here, you know he's worth 4 billion he's apparently worth the same.

Anders Arpteg:

I think it's a trend If they come here then the company will go unicorn.

Henrik Göthberg:

Any cool people who hasn't? Made it yet come here, we fix it for you.

Minna Sandberg:

I look forward.

Henrik Göthberg:

Now you know what's coming. Now you know what's coming.

Anders Arpteg:

That's cool, no, but I guess it's a very positive thing that we have these kind of amazing exits for Swedish startups that scale I think SANA went like 10x in one year or something in valuation.

Minna Sandberg:

It's one way of getting into the Nordics. It's not easy to scale here. Coming back to, for example, with Silo, the network they have around in the Nordics of companies, I think for AMD it's actually the way of getting into that in a natural way. It's a fast track.

Henrik Göthberg:

What's the Workday logic? Sorry, why do Workday acquire Sona? What are they after?

Anders Arpteg:

Yeah, good question. Is it for internal usage? Is it just for improving their own way of working? I mean, if we break down Sona Lab, I mean they have, I think, two major things. I mean in the beginning they were more of an e-learning kind of platform, if I'm not mistaken.

Henrik Göthberg:

Now they're assistance and agent.

Anders Arpteg:

Yeah, now it's more an AI platform can serve so my take.

Henrik Göthberg:

Sorry for jumping. I think they want to get that. They realize software is changing. Software is going more co-worker style UI, so they want to be good at that. This is one angle. Why SANA? The second angle SANA is that what SANA has been pivoting towards is enterprise learning on the fly. So instead of you know how do we keep up with the pace? So if workday is having more and more and more advanced features, they also need to have in-built learning of how to adopt those features. I think it's a double whammy of getting the ai agents or being able to take the work they're making into making making, simply flipping that to agent style plus the learning in order to drive the adoption yeah, but is it for internal use or for their products?

Anders Arpteg:

I think it's for their products in this case.

Minna Sandberg:

I choose, I think so too, and I think it's like I mean, if you always think like if they wouldn't do it, I mean a lot of companies will not be around anymore when we see opportunities for coming back to lovable and wipe coding and kind of also building, anyone could develop anything. Right now, I mean, my father is 70.

Anders Arpteg:

Don't hype it too much, no.

Minna Sandberg:

I'm not only referring to vibe coding. I'm thinking of, like building and learning from uh getting your you know uh chat gtp to kind of give you support on like uh yeah, my 70 year old father is kind of learning how to code right now it's amazing yeah, exactly because he can get the feedback back and forth all the time. That wouldn't be possible otherwise.

Henrik Göthberg:

But it is logical. If you flip it Workday, imagine they haven't really gone fast enough on this route. And they want AG agentic approaches and they want embedded learning. How the fuck can they do it better?

Minna Sandberg:

Exactly so. It's a fast track and it's also survival mode. They need to do it, otherwise they fast track and it's also survival mode. They need to do it, otherwise they wouldn't be around I guess.

Anders Arpteg:

So it's a good or bad thing. I mean, it's good for work day, I believe it was a very good for them. But then what about valhalla? Is it good or bad for valhalla that they got acquired, you think?

Minna Sandberg:

I mean it will be a lot of. I think that they are very the. The founders of Asana has been a very important part of building the Silicon Valley and they are very active in a lot of the networks around it. So I think they will continue and hopefully they're ready for the next big thing. Yeah, they will hopefully reinvest most of it.

Anders Arpteg:

So they'll probably get the golden lock for some years in the birthday, but then probably spin up some kind of new thing with the capital they have Exactly.

Henrik Göthberg:

So, it's good.

Minna Sandberg:

It could potentially be good.

Henrik Göthberg:

It all depends on how you and Mattias and everybody's dealing with it. But I think it could be positive.

Minna Sandberg:

Yeah, of course, Of course. I mean, that is one way of seeing it. It could be a problem that we have American companies and Chinese companies owning so much of Sweden.

Anders Arpteg:

Because you know before we've said on this podcast, you know, a big problem is that we have a lot of talents, we have good universities, we bring the people up, they have actually some capital and they have good startups, but then they get the brain drain by you know a big American company that move them out of Sweden and Europe, and this is partly that as well.

Minna Sandberg:

So in some sense.

Anders Arpteg:

I think so. But the question is then if we can keep some of the at least capital and if they can move on to something that actually creates work and prosperity in Sweden. And I hope we can, but in some sense it is a brain drain move this one.

Minna Sandberg:

Yeah, and I'm comparing it. I mean, if we're thinking, if they wouldn't be bought and they would continue to build and do an IPO later, would that be better? Like what would that give back to the ecosystem? Exactly, and I'm thinking then, because I think it would be better.

Henrik Göthberg:

What would that give back to the ecosystem Exactly?

Anders Arpteg:

And I'm thinking, then, because I think it would be better. By the way, yeah, I think so too.

Henrik Göthberg:

Yes, but the big IPOs is Nasdaq. It's not in Sweden.

Anders Arpteg:

But still. So they become American companies anyway. No, no, but later. You don't think so?

Henrik Göthberg:

No, I think so.

Anders Arpteg:

They still can operate in Sweden and do the original business plan they had, not so for Sana or for Silo. Okay, that was a long topic but a very good one. I think Perhaps one more Should I do the boring, I think a small, but I think it's an interesting topic is the extreme focus on AI for coding, and OpenAI just released their GPT-5 version of Codex, their kind of AI specific version of using AI for coding. And why is that interesting? It's just another model. Yes, it is, but for one, they said when they released GPT-5, it was specifically one of the most important part was for coding, but it didn't really perform that well and it wasn't really better than Anthropic and the cloud models that they did have. But then they focus a lot now and they potentially are betting I haven't tried it yet, I must say, but it seems to really show how important coding is for the big AI labs apps and in some way I consider AI for coding to be one of the best metaphors or examples of what AI will be in the future.

Anders Arpteg:

So for coding, you actually do have AI that can work with large amounts of data, which it can do for many cases, but it can also do reasoning surprisingly well for coding purposes and also agentic tasks already today for coding. Not so for other processes and business processes we do have, but for coding yes. So in some way you can see into the future, but see what you can do today when it comes to AI and coding.

Henrik Göthberg:

So you can study what is happening here and understand the trajectory, which is then you can prepare for it better.

Anders Arpteg:

And I play around with myself, I can see that I become this kind of agent manager. In some sense, I became like I have my own team of not people but AIs and I can tell them what to do and I can just speak English. To do so, I need to understand what to do, otherwise I have no clue of telling if they do something wrong or not. But again, I get personally like my own team.

Minna Sandberg:

Yeah.

Anders Arpteg:

And I think that's a great metaphor to what we will see in three, four years. For more or less any kind of process that you have in a business, each human will have a team of people-ish or agents that they can use and that will tenfold their productivity.

Henrik Göthberg:

And hint hint. Looking at the profile of what makes it work in software, we can kind of see the trajectory. Which areas is it more suitable to do next? And next, and which one is a little bit slower and this is quite obvious if you look at the profile of the problem I think I mean one.

Minna Sandberg:

What I think is so interesting about this is that this was kind of the vision that OpenAI had very early and they have kind of.

Minna Sandberg:

This is at least very close to the vision that they had. I actually had the opportunity to meet OpenAI one year before they released ShutGTP the first version, and I saw the news. You know a lot of people were talking about it in Silicon Valley because it's, of course, yellow mask, but also they had they got their hands on a lot of really big data sets from Meta and somehow Meta strategically I think now afterwards really released the data sets and I went there and then I was so convinced there was actually one of the big reasons why Swinode was founded Because I went there and I then connected people from Vinnova with OpenAI and I said you have to go there. There Something groundbreaking will come out from Silicon Valley, from OpenAI, I'm so sure. And they went there and they also met with the team and met with a lot of others, and when then ShutGTP was released a year after, I think that was triggered why I was asked to be leading this initiative, because I could see that come.

Anders Arpteg:

So this was in 2022 or something.

Minna Sandberg:

Yes, already. Back then, when we talked to the team, this was the story. They had a very strong vision very early and they have been working hard to get there. I think the competition is, you know, yeah, that's why they are succeeding, but I don't know if they will be leading forever.

Anders Arpteg:

I think OpenAI has trailed behind a bit and they think they will continue to trail behind. But we'll see.

Minna Sandberg:

And they actually I mean Vinnova actually connected them with AI Sweden who have done this collaboration, and they got to test the GTP5 due to that. So that was actually my contribution to Sweden, to that story as well. So a few in Sweden were first um of testing that new model awesome.

Anders Arpteg:

Did we have something else going on that you would like to bring up?

Goran Cvetanovski:

I will do just a five minutes rent and then we'll be done with it. Um, I will not talk about work today because you can see this map and you will know why they need to stay ahead of the competition. But um, let's follow the money, you know, because that is my part. So nvidia bought like five billion stake in intel.

Anders Arpteg:

That was a big one, and this week um, and also the government invested in intel, right, yes, that's kind of interesting.

Goran Cvetanovski:

Well, it was falling behind a little bit, uh, behind the amd um. So it's a very strategic move from NVIDIA. I think that they are getting like 3% stake or something like that within the company after the entire what is called deal has gone through, but it's very strategic because Intel has the packaging, it has the manufacturing and everything else in the United States.

Goran Cvetanovski:

So it's a good investment. But bear this in mind because in the whole concept of what today I'm presenting is actually a little bit interesting. Then we had Microsoft now sign the 66 billion deal to rent AI compute from N scale. That is in no way right, yeah.

Goran Cvetanovski:

Keep in mind they already invested in Sweden. Sweden now has a number of data centers actually that are in AI centers that are happening in Sweden. Sweden now has a number of data centers actually that are in AI centers that are happening in Sweden. So the Nordics is actually becoming a little bit more as a you know let's call it like an ecosystem for data centers and AI factories. Yeah, we have a good climate. We're talking about Silicon Valhalla, but that is the innovation startup. But we need to also separate that. We also have an industry that we need to take care of.

Goran Cvetanovski:

And the industry is actually what makes the Nordic market to be so innovative and, as you said, that is the advantage. So this type of deals are actually helping the Nordic market to be very close to where the compute is and where the data is, and etc. And that will give us a boost, more capabilities and innovation in enterprise setting not only in the Valhalla part.

Goran Cvetanovski:

Then we had in Sweden. We had Ecodata. Yeah, that was a new deal as well. Ecodata Center secures 600 million. Yeah, it was something like that. Give me a second, let's find out where it go. There it is Swedish digital infrastructure provider. Equal Data Center has secured 600 million euros in debt financing from Deutsche Bank to expand its AI data centers. It says on Tuesday, etc. So this is actually two things that are happening in similar things that are happening in one week.

Goran Cvetanovski:

And then I was very excited about this and I was actually expecting this is going to happen. I was speaking with you know events. I was speaking with a lot of companies from telecom and et cetera. And telecom, it makes sense that they should be the new data center or the cloud providers and we actually discussed this at large a couple of years ago back because they're already very regulated, they have the infrastructure, they have relationship with public and private sector and etc. And and they have the 5g, they have the networks and everything else.

Goran Cvetanovski:

So it makes so this week, uh, telenor uh announced their first ai factory, which is basically a new service from telenor that you can companies can actually uh tap into. Uh and this it's, this is all into this year team, which is Sovereignty in Europe. Make mega great again. Make Europe great again thanks to Trump. So I think this is super good and just shows that there is much more happening in the Nordics than just Silicon Valley. It's actually it's really really a right time to be in the Nordics. Just, uh, silicon valhalla, it's actually it's really really a right time to be in the nordics when it comes to ai and data and innovation and etc.

Anders Arpteg:

So, from all angles, public sector just got like a lot of money, so things are happening and the big um investment from the swedish government as well, like 500 million sec per year in five years in a row. So it's uh, a lot of stuff is happening. It's the best time ever. I think too.

Goran Cvetanovski:

But keep in mind, all of these things also points out to one thing that we want Microsoft to invest in Norway, because they understand the centralized what is called data centers. The business model doesn't work, so a decentralized compute, closer to the people that are innovating, is the next thing to go, and we will see more from Google, we will see more from Snowflake and many others. So I think data centers and AI centers will happen. So the innovation will happen to the nearest clusters, to where data and AI factories will be in the future, and we are blessed to be exactly where we need to be to innovate. So that's it. That's all Good rant, perfect, exactly where we need to be to innovate.

Henrik Göthberg:

So that's, it.

Anders Arpteg:

That's all Good rant, exactly Perfect. And if we could, mina, I'd like to move a bit more into more philosophical topics and perhaps also in general. If you start like this, you have a lot of insights into startups, I think, specifically, but also larger enterprises, but still a startup that want to succeed, and you can see both from the Valley point of view but also Valhalla point of view. Can you give some simple advice? You know what are the biggest mistakes. How could they potentially get in contact with you and get some good traction and become the next sana labs or or silo ai or lovable or, etc yeah, I mean, I think it's uh.

Minna Sandberg:

I usually think of it as very different processes, depending on if it's a b2b or b2c, or I mean uh kind of a ai native scale up uh startup like sauna or uh, lovable I see that is, or at least lovable is, a product. That is different uh. But if you look at the from the perspective of b2c, uh startups, I think you should have a global approach from start, meaning that you should have a global mindset, you should think big. That you should have a global mindset, you should think big otherwise.

Anders Arpteg:

Think big or go home.

Minna Sandberg:

Exactly. Don't do the mistake of spending all your time on one slow customer in Sweden. At the same time, we have the opportunity of having a lot of really good companies here, so taking advantage of that is a dream that I think that I can support with, because we have the contact on a higher level at a lot of companies.

Anders Arpteg:

So contact me, can I just slightly challenge you on that, because I do agree, of course, that thinking big is super good and you can have a huge customer base from that. But if I were to quote Elon Musk, we haven't even mentioned Elon here for some time.

Anders Arpteg:

I think, he gets angry with us soon. But he said something which is if you are a startup and perhaps have a B2C kind of agenda, then either you focus on having a huge impact for a small number of customers or a small impact on a huge number of customers or a small impact on a huge number of customers. Wouldn't that still make sense? We could potentially, as a B2C company, still say that, but just for plumbers. We will make such an amazing experience. That could be a fine business model as well.

Henrik Göthberg:

But this is still global potentially, because if you go narrow but B2C plumbers, it could still be global human plumbers, you know what I mean what I'm trying to say. It's bad example actually you're commenting, but and you're highlighting different aspects, but they're not against each other at all, but there are example.

Anders Arpteg:

I think it was a bad example from my side that plumbers, but still you could think of something that only applies to sweden, or something that that still could be worth for a startup to do potentially yeah, I mean, um, what I would do if I had like a b2b startup myself?

Minna Sandberg:

then I would actually try to first, um, I mean, of course you have to have a few, like a few customers, improve your products and stuff like that, and that you can do in Sweden, but as soon as possible, try to get as big customers as I can on a global basis, because if you have one really big, important customer, then you can easily get the second and third, because that's how it works.

Anders Arpteg:

It's like this A big reference customer that gets you started.

Minna Sandberg:

Exactly, and you can take the opportunity here, because we have really strong brands in Sweden. You should also have quick and fast sales cycles. You should see some really good results only in a few seconds, like really good demos, and I think one startup that is doing this really well is Legora. They are working with enterprise customers, helping them with their legal work and NDAs and stuff like that, and they are growing really fast. And they have the Silicon Valley mindset of being a part of the velocity program, traveling with the equity and others to Silicon Valley. That's how they started and then they have been a part of Y Combinator, so really like building a global first mindset, having a Swedish customer to start with, one of the biggest law firms that could really prove that they are something, and then they have just like continued and they are growing really really rapidly. So think big, don't be shy, go bold or go home kind of, and then try to secure as many customers as possible quickly.

Anders Arpteg:

Yeah, that makes sense. B2b then.

Henrik Göthberg:

Because we're looking at ARR, we're looking at subscription rates and then all of a sudden we get into someone who is working much longer sales cycles. I mean like selling enterprise sales for something that is more sort of harder to get into the core of the enterprise If you have some great ideas but it's more B2B or physical and industrial AI.

Minna Sandberg:

Yeah.

Minna Sandberg:

I mean we have some really good examples. In Gothenburg, for example, we have mostly B2B customers close to the automotive industry. Some of them have done some really good work together with Volvo, which is a global brand, and that took some time, but then they could go to Silicon Valley and, through our network, get opportunity to get customers like Toyota's AI lab. They got Nvidia as a really big customer, one of the startups, because they are building the Omniverse platform. So I think, going to Silicon Valley when you proved yourself with, in this case, the biggest brand in the world, which is a Swedish brand.

Henrik Göthberg:

So it's the same. It's actually a good thing either way. B2b, b2c then.

Minna Sandberg:

Yeah, B2C, I would say you can build in Sweden then you have to put so much effort into the marketing part so you can go and take the real big money in Silicon Valley from investors. If you're a B2C startup and that is exciting that's what we have seen with Lovable and others but you just have to put so much time and effort into being smart when it comes to market yourself and grow as fast as possible, because time is of essence.

Henrik Göthberg:

So timing is a really big thing here.

Minna Sandberg:

Yeah, and also building really good products, because otherwise you will never win.

Anders Arpteg:

Speaking of products, and you also mentioned enterprises and Volvo and Jock Gothenburg and companies like that, and I think as a startup, you can get into the enterprise world because a startup can move faster and potentially help them in some way.

Anders Arpteg:

But I'd like to just flip that a bit and say, if you are an enterprise now, you're a Volvo or something and you're all feeling the FOMO, so to speak, that they are trailing behind and some competitor will move faster with AI, et cetera. But I think sometimes they are so focused on products and miss out on actually the people that they have, the employees and the business processes they do have. So I think sometimes an enterprise, they may have some nice products actually, of course, they should be AI optimized as well but they can do something with AI that the startup never could, which is actually to optimize their own. You know people make them ai empowered and also processes and actually a very good use case as well. So what I'm trying to say is that ai shouldn't just be to improve and empower products that's important but for existing companies yep, knowledge workers and uh yeah, that's a great workflows.

Henrik Göthberg:

To remove internal friction is a massive thing of course.

Anders Arpteg:

And then you know I've studied so much about, you know how Tesla works and how they operate internally and how they use AI for a lot of their internal processes and for the people, and you know I'm amazed how much they do, and I think any company that don't start to think that way not for the product, but just how they operate, you know internally they will get left behind.

Minna Sandberg:

I would say yeah, I mean, coming back to volvo is a good example. They they had a really big performance at the cs conference last year or this year, actually in january, and they were they were the main keynote and they talked about this transformation, where they went from having free strategic collaborations with software companies and now they have 50. So they realized that it's not possible for them to build everything themselves, but they should use AI in a greater extent for everything, and collaborations is the way forward here, which is really important, and I agree with you. I mean, in big companies they have a lot of domain knowledge and in Sweden you can work for a company for 20 years and that's not even weird, it's like how you do it. It's not even weird, it's like how you do it. So to focus on your knowledge workers and to make them better at what they're doing is really an investment worth doing.

Minna Sandberg:

But in big companies, I believe we are working a lot on this change management. How should you apply AI, mass trainings, having internal champions in different groups I've seen different strategies that worked at companies and I don't have the answer, but I think that if the leadership really truly believe that this is the transformation, they would like to go on. I think then it's when it's really going to happen and that you can see in the investments like how much have they actually invested in this? And we will see. We can already see now that a few companies are not even thinking of it as a serious thing, and then, of course, it will have an impact in a few years.

Anders Arpteg:

And just continuing on the question about people and employees and I know you're a strong advocate for women as well in tech- yes. Can you just elaborate a bit more from your own side? What kind of challenges do you see? I mean, you're obviously successful in the field, but still, if you were to just think more a bit about the challenges that potentially women have to get further in AI and tech, what would you say?

Minna Sandberg:

I mean looking at founders. It's a lot of buzz around like cracked founders right now and people are, you know, working day and night to build stuff, but it's very rare that we see any female founders uh, being able to do that. And, um, when you are, especially in silicon valley, the best, um, the best startup founders are usually around your 30 to 40s, and that's the way also the same time that you are thinking of building a family. So to be able to, to do that, you have to have the entire support system behind you. So I think but I think we are in an era where that is changing. At least in Sweden, we are really good at supporting each other. We have the opportunity to get kids, work from home, you know, you can work from your computer, you can travel, you can do stuff. So I meet a lot of really strong female engineers and founders.

Anders Arpteg:

Isn't that a perfect example of where Valhalla actually is better than the Valley? Because at least for women. We have better opportunities than to actually do take or make a career here.

Minna Sandberg:

I have one of my best friends. She's an AI lead at Meta. She was previously a founder of a startup together with Andrew Yang, and she has done an amazing career. She's one of the best AI engineers in, I would say, silicon Valley. But when she worked prior to Meta a year ago, she worked at Accenture as the global AI lead and she was responsible for eight different continents, meaning she had to stay up for meetings all over the clock and they paid her to froze her eggs so she could actually work more and longer. And now she had been struggling for two years with the evf process and it unfortunately didn't work. So she had this false belief that I mean I they paid for this frozen egg program that they have for all the women. They have that at many of the big companies.

Henrik Göthberg:

And now I haven't heard about that story.

Minna Sandberg:

No, it's uh it's interesting, so now they, uh, and it's, that's not like you can't. Heard about that story. No, it's interesting, so now they, and that's not like you can't. I mean, it's sad.

Henrik Göthberg:

But you have led us into another dimension of entrepreneurship.

Minna Sandberg:

Yeah.

Henrik Göthberg:

And this is life balance. And if we want to talk life balance, it's very, very interesting to talk about that angle with you. Yeah, even the way you have managed, as a family, to live three months in the valley and then back and forth. You have three year olds now, but could you paint a picture for the listeners what is what I'm talking about? And then lead, lead it into your thinking and or rant on this topic? I think you, you, you obviously thought about this a lot and you're making it work yeah, I mean I, um, I love my work.

Minna Sandberg:

That is like you know. I have, uh, that top of mind all the time and I think that's my, my first baby. But then, uh, I have a three, now almost four year, old. The first time we traveled to Silicon Valley together she was three months old and we struggled to get all the vaccines done. Before we came over to Palo Alto, I walked with my stroller back and forth, having like online meetings in my headphones in the morning with and late afternoon, especially early mornings because of all the people in Sweden. We walked back and forth on Hamilton Avenue, which is the same street as Mark Zuckerberg. So after a while his guards actually started to follow me because they thought I was a spy, typical stalker, you know, having a stroller. So that was a really fun story. But yeah, now she's almost four, she has been visiting I mean, I have a lot of friends at Meta, google and other places, so I bring her and she loves going there eating ice cream.

Henrik Göthberg:

Yeah, you said that Is it the Google ice cream or the Meta? That's best.

Minna Sandberg:

The Meta is better. They have a much cuter pink ice cream shop, but they have, I mean, one story that I have a friend that works with HR at Meta and she told me that the reason why they built like an entire city in the Meta office is actually because of the 10x engineers, because they a lot of the people that are really smart and good developers. They are also scared of doing new things, so they if when you come into Meta you have your own the doctors, the kindergarten, they have a lot of restaurants, they have a lot of coffee shops, you never need to leave the office, you can meet the same people every day and that they know that they can't compete on salaries. It you know you can always offer someone a higher salary but you can't offer them a safe environment. So that is all created to sit for the the few, very few people that actually don't want to leave that environment.

Minna Sandberg:

So that's the reason why you that's a really amazing headquarter. But yeah, and we with my kid, she, she said last night when I read a moving story she said, um, that moran is, she's so cold so she doesn't have any friends, but with the humanoid we met that figure they could actually become friends because they are not cold. So she's already thinking of you know, the Silicon Valley mindset within her play and, yeah, it's really fun.

Henrik Göthberg:

So it's one way of living, but your lifestyle I mean like it works, of course, because your husband can share the same rhythm with what he's doing. Did that just evolve? Or was that something that you too, even when you had kids, you realized? This is how we want to design our lives. How did you come there?

Minna Sandberg:

We took, yeah, so I mean we have been together for a very long time 12 years now so I mean even before my first trip to Silicon Valley. He has always had the same, similar lifestyle, but we had, I mean, some kind of decisions that we made. He got a really, really big offer to be responsible for one of the biggest AI initiatives at a big company in Sweden, but I was really into you know what I'm doing. So we took the decision to not do that right now. So, of course, sometimes you have to, because that wouldn't be able.

Henrik Göthberg:

Then he would have been stuck in that initiative rather than being able to follow the rhythm.

Minna Sandberg:

Yeah, and he will probably be need to travel all the time and I will be home with kids not being able to work. So it was a decision that we took together, but it's not. I understand that this is not a lifestyle for everyone, but it's basically just being able to work behind a computer, having digital meetings all the time. People doesn't even know where you're based, which make it possible. So I think the past COVID era makes it possible for a lot of women to actually work and have kids.

Anders Arpteg:

That's interesting, Great and perhaps if we were to move a bit more into philosophical topics. I mean AI of course, can help with so many things, and it can also help with technology and meetings and all the things that enable women to actually do work in different ways. But I guess there are a lot of challenges as well and perhaps it would be interesting to just hear in general sense, you know, what do you consider the more ethical challenges and considerations?

Minna Sandberg:

maybe as we get more and more ai empowered, if you call it that I mean, what I'm scared of is actually the, the in this change, the um, I mean the knowledge, or and the old information that now you can actually know everything about the person and the big companies know more about you than yourself. I mean, yeah, so that is what scares me the most.

Anders Arpteg:

Did you hear the example of I think it was a Google ad thing where Google ads can track you so well you know what you're doing and your behavior and it was a female that suddenly started to see a lot of ads for diapers or something right, and then it turned out she was pregnant.

Minna Sandberg:

Yeah, and she didn't know it. I remember that one. I thought it was like how I remember it was.

Henrik Göthberg:

I had some maybe it was a similar story where someone else in the same uh house yeah started to see uh yeah the dad started to see direct marketing email or sorry, mail for a diaper diaper discounts to the who she was leaving at home.

Minna Sandberg:

Exactly.

Henrik Göthberg:

And this is how sort of everybody found out.

Anders Arpteg:

Yeah, exactly, and I mean you can figure out so much from data and people. I don't think understand that really.

Minna Sandberg:

Exactly, and now you see people traveling to the US and they're not allowed to enter the border because they liked something about someone you know. So it's really some in some sense. I think it's scary and but also, um, on the opposite side, I think there is. I mean, it will help us live longer and it will be, you will be able to have a much better life. I mean, we are working a lot on, for example, apple Heart Study, where you could, in Sweden right now, get the Apple Watch and you can live your life as normal and you can follow the heart from home, which is one way of using this physical AI movement as an enabler for a lot of people to live a normal life. And this is what is a huge opportunity in Sweden, because we trust the system, we trust our government, which means that we can use new technology much easier than we in the US.

Henrik Göthberg:

This is the big game. How much investment should we do in preventive healthcare versus the other way around? You know, and this is such an obvious business case and an obvious better for society. But and it's a big but you need to then do the work, do the trust system in order, for that. Preventive healthcare means you need to show them the data.

Anders Arpteg:

But I think she has a good point. We actually do have a big trust in both tech and the government in Valhalla, which actually provides us with an opportunity to actually potentially gain more benefits.

Henrik Göthberg:

I mean it would be easier to sort out preventive from a cultural perspective in Sweden than in many other places, simply because of this.

Minna Sandberg:

We have, as one of my strongest focus areas is AI in precision health, and that's because it's one of the strongholds and we have the biggest opportunities here in Sweden due to this, so we can. One of the products I'm deeply involved in is where you use your Apple Watch to detect sudden death Sudden death.

Minna Sandberg:

Yeah, sudden death is when you die of a heart disease that you didn't know of. You are healthy in all other areas. You have no history in your family, but you can use the smartwatch this could be Apple Watch, or it could be a Fitbit or whatever with an EKG To indicate before.

Minna Sandberg:

Yeah. So what they have done and this is actually back to Marcus Lingman and Regine Haaland, together with CR Ogermeyer from UC Berkeley they have found a way to not only detect and see deviations using large language models, and they have actually built an AI that could tell. So that means that a person or a doctor couldn't see the differences between a sick person and not a sick person, but an AI can, and that means that in region Holland, they did something really cool. Instead of just collecting, they went out to everyone that lived in the region and said do you want to be part of sharing your data or not? And it was only eight people that said no. Sweden.

Minna Sandberg:

Yeah in Sweden, and that is one really good example.

Anders Arpteg:

We trust the tech right, we trust the tech and we trust Sweden. Yeah, in Sweden and that is one really good example we trust the tech right, we trust the tech and we trust Fikir Halla.

Minna Sandberg:

Yeah, and they don't have. So now they can do a lot of preventive health which will give back to the entire region and to Sweden in general.

Anders Arpteg:

So at Stanford, for example, it's privately owned, which means that the hospital, the Stanford School of Medicine, if they should do anything, they could do it, but it's a very small study but speaking about health and health is, of course, a very regulated area and if we speak about the valley versus valhalla, one big difference is regulation then, of course, I'm not sure if you want to go into this topic, but you know, the trump actually recently in the summer, made a big AI action plan where one of the biggest points was really that you want to minimize regulation and really maximize the innovation possibilities for AI. And then some people claim that Europe is taking the opposite route here and adding more and more regulations for companies and adding more and more regulations for companies, and it's also like a rather fragmented situation where each country in Europe have different ways to implement these regulations, which cause potentially some problems, or what do you think?

Minna Sandberg:

I mean, yes, I think GDPR, for example, was really like it banned so much innovation to happen. People were so scared so they just decided to not do anything. And I mean, we are using all the data that is created from social media and is US, so we don't own anything ourselves. The only thing we have is health data and from industry data. So I think that's a big lost opportunity for so many years, and that's also why I'm quite surprised that we are actually having Silicon Valley movement at all, because I wouldn't see that coming. So I think it's a big mistake that we are going that way On the other side. Personally, I think it's good, because I don't want, as I said, the integrity is really important.

Anders Arpteg:

And it could be the strength. I mean, I think no one argues about the intention of the AI Act and GDPR etc. But then the question is how it should be implemented, and you know we've spoken about this so many times on this podcast. But to at least know, before the law becomes enacted, how to become compliant could be a good idea, I think. And it's not. It wasn't for GDPR People, just the uncertainty of how to become compliant caused a lot of fear, and now the same thing is happening again, exactly.

Minna Sandberg:

It causes fear. That's the same thing is happening again. Exactly, it causes fear. That's the big thing that you don't even know how to. You're so scared so you don't do anything at all.

Henrik Göthberg:

So this legal uncertainty we're talking about and we talked especially about this a lot in the last episode with Louise from Assa Bloy, who is really strong in this area, ex-eriksson from Assa Bloy, who is really strong in this area, ex-eriksson and I mean like so we talk about the act coming In reality. What we really are all waiting for is to harmonize standards around this. What I can even live with to some degree that there is a due course here, as we talked about is you cannot get to due standards until you're done, so you need to lock it as something to work against, fine. Get to do standards until you've done certain. You need to look at something to work against, fine. What I'm really, what annoys me over, and over is that we have defined the definition of done too short.

Henrik Göthberg:

We have not looked the life cycle of the law, and so when you look at investment, it's the same. It's the same thing as we are invention focused, not adoption focused. So it's the same problem. When I have a cool tech idea and I want to bring that into Scania and it ends up on the prototype graveyard, it's the same fucking problem.

Anders Arpteg:

It's a good point. I mean we have forgotten the change management required. We just build a prototype, right More or less, but we haven't really built a product and how to adopt it.

Henrik Göthberg:

So once again, think big, exactly.

Minna Sandberg:

Yeah, I think I mean, yeah, that is sad.

Anders Arpteg:

I agree. Yeah, but still, perhaps we should move to the last topic. But okay, in some way you can argue that if you compare the valley to Valhalla, it could potentially be a benefit as well if we just got away from this uncertainty that we have. So I mean if we know that Swedish or European companies do have to comply with good regulation, which I think is no one that don't want to have any regulation, of course.

Anders Arpteg:

We want to have good regulation, which I think is no one that don't want to have any regulation. Of course we want to have good regulation, but that could potentially cause that companies in Europe and Sweden become more popular because you can trust them in some sense.

Minna Sandberg:

Exactly, I agree that is one good thing, and we have a good brand. I mean, in Europe and Sweden especially, we have a really good reputation. People trust the companies. What we're doing, so I mean I don't think we should try to, at least not from the bigger companies perspective. We shouldn't try to build, we should adapt and use what is out there and don't be scared of doing so, because I think AI Act is so much better than, for example, gdpr.

Henrik Göthberg:

They have actually thought about a lot of things, but people just have to read it and understand how it applies to their specific area because if you think about, if you flip it, if we crack harmonized standards, if we crack that, we pour a lot of investment into having, I mean like thefts and everything like the sandboxes. So it's in Swedish det ska vara lätt att göra rätt it should be easy to do it the right way.

Henrik Göthberg:

Then all of a sudden that opens up that, together with our culture of trust in the system and that therefore actually sharing data in different ways, it could flip the whole arena. But it becomes more and more obvious what you need to pour a lot of money in to crack it to engineering standards, not just legal standards.

Minna Sandberg:

No, definitely yes.

Anders Arpteg:

I mean I think no one disagrees with the goal from EU Commission, et cetera, to have trustworthy AI. I mean that think no one disagrees with the goal from EU, Commission, etc. To have trustworthy AI. I mean that would be perfect and we shouldn't lose that. I think we should keep that. That should be the edge that we do have here. It's just that we need to crack it properly with harmonious standards or something else. Yeah, Mina, time is flying away here and I'd like to ask you a very simple question. I'm being very ironic, it's a very hard question, but still, if we believe, or do you believe, that we will have AGI at some point in the future, you can start that. Yes.

Minna Sandberg:

Yeah, I mean I have been interested in this field for quite some time. I think there is some really really good researchers in Silicon Valley, especially at UC Berkeley, actually within this field, Talking about robotics, for example, I had a roundtable that I hosted for a few years within the field how to use AI and AGI to accelerate that usage. So, yeah, I think it will take some time before it's actually common use, but there is Three, five, 15, 50?. I would say five years maybe.

Henrik Göthberg:

Are you distinguishing AGI in a lab versus out broadly? Exactly, that's the question, because I think in the lab, out broadly.

Minna Sandberg:

Exactly that's the question, because I think in the lab, yes, already existing Almost, yeah, or existing even. I think it is existing, but I think it's similar to the discussion about quantum computing right?

Anders Arpteg:

No, it's not similar. Don't go there. Don't go there.

Minna Sandberg:

But I mean on the time you can argue, I mean in.

Minna Sandberg:

Silicon Valley. I was waiting for the AI boom for many years before it happened, so it's hard to tell. But I think AGI it's interesting. I think I'm I mean, I brought two kids to the world, so I'm a believer that we will have a better world, that we will work uh humans and ai together, and I think we will soon live so much longer with uh all the technology out there so we will be able to focus so much more on our health. I think that the big change will be that people would focus mainly on living a healthy life and the life they want, and then secondly on the work, instead of working and then prioritizing a few activities afterwards.

Anders Arpteg:

So it sounds like you're pushing to the positive side of this, but there could be a negative side. It could be AI used to create World War III, if it's not already started.

Minna Sandberg:

That's the dark side, I see. I mean we have a lot of investments in the defense it's the fastest growing industry right now.

Minna Sandberg:

That is probably the first application for aeg already and the. I mean a few years ago, in europe at least, we focused all the funding on green transition. I never even hear people talk about that anymore. It's all about defense innovation. So, yeah, I see that. I mean Mark Zuckerberg put all his own money into building an island, an escape island. So if you are behind the big companies and you are such a strong believer in that that will end the world, then and building an escape route, of course that's something uh.

Anders Arpteg:

But so what do you think? If we take the two extremes here, one could be the very dystopian, like the matrix, you know, or the terminator, with the skynet coming and machine killing us all, etc. Or the more utopian version where I'm not sure if you see Nick Bostrom's the Deep Utopia book, but they speak about the future and a world of abundance where we have AI that solved a lot of our health issues, cured cancer and so much more, and we have no energy dependency anymore, we have more or less free energy and basically the cost of goods and services go towards zero in some sense. Where do you think we will end up?

Minna Sandberg:

in some way, I think we will end up in both, meaning that a few people in the world, very fortunate people like us and people in Silicon Valley and other areas, will have a really good life. They will continue to have a better life, a longer life, a more healthy life. You will work less and have a better. You will live longer. But then we will see even more wars and other parts of the world where people will die in millions, and it will not affect my life and your life, but it will. We will have less people on earth, but it the the rich and fortunate will be having an even better life and I think the poor ones will have even worse so you think the divide is growing?

Minna Sandberg:

yes, the inequalities is growing yes, I think so and I think that is uh one of big like. If I look into the future, I think a few companies will grow faster and bigger, and I think also the differences in Silicon Valley it's like differences between being really poor living on the street, which they have a lot of, and the rich, normal people is huge and it's growing even in, especially in san francisco, because in palo alto. They, they. The police is taking them to east pal alto, which is one of the most dangerous places.

Anders Arpteg:

But yes, I mean it's a good metaphor in some sense in the valley, because it's such a big difference and you have a lot of homeless people there and a lot of people doing crack and whatnot, and then you have the super rich.

Minna Sandberg:

So it's a huge difference of class there which could potentially spread to the rest of the world in some sense yeah, I think that's the future, that more, yeah, the differences between people will be so much bigger and we we unfortunately that will have. I mean, that is the quite sad future.

Henrik Göthberg:

But isn't this, then, one of the key goals missions we need to have how we create inclusivity, diversity and battle the AI divide, so to speak?

Henrik Göthberg:

So we as an anecdote that was part of the mission statement while we started this pod, because we think how do we get more people to be on the, how do we close the AI divide, so to speak, or how do we make sure Sweden is not on the wrong side of the AI divide? But what are we fundamentally doing, or what becomes important, in order to minimize the divide or not just happen in certain way, but steering it for good, so to speak? What do you think about that?

Minna Sandberg:

That's a good question. I think what we put both our effort but also funding into. So, as I said, we see a lot of movement into defense and other areas right now, which is, you know, maybe not the only use case that we should focus on. A few companies have really good intentions to make the world a better place and focus on how to actually solve cancer, how to solve the energy and the transformation towards a more sustainable world.

Minna Sandberg:

I think that is the really good opportunities we can have. We are already forefront in Sweden on thinking in these terms, but that is what we could contribute with and inspire the rest of the world so sharing success stories and really be forefront in the Nordic values.

Anders Arpteg:

I'm trying to keep that kind of culture, perhaps in the Valhalla and not, in that at least aspect, be inspired by the valley, because then we have something better, here in Sweden at least, than the Nordics, right? That's what we think, and I think so, yeah, well, let's hope that we bring the best of the valley to the Valhalla, but not the worst then, and if we do that, it could be a really beautiful future here in Valhalla Right.

Minna Sandberg:

I agree Everyone should tap into the Silicon Valley ecosystem, but let's build and let's create the strong. Nordic values. Based on the Nordic values.

Anders Arpteg:

Sounds amazing. Perfect ending, I think, saying thank you so much, mina Sundberg, for coming here and I hope you can stay on for some future off camera discussions. Perfect ending, I think. Thank you so much, mina Sandberg, for coming here and I hope you can stay on for some future off-camera discussions, but it's been a pleasure to have you here. Thank you so much.

Minna Sandberg:

Thank you for the opportunity. It was a great pleasure. Thank you.

People on this episode

Head Shot

Anders Arpteg

Host
Head Shot

Henrik Göthberg

Host
Head Shot

Goran Cvetanovski

Co-host