The Global Framework AI Needs Before It Controls Our Future

Eleni Floros: AI is transforming fintech at breakneck speed, but should we be thinking about how to slow it down? Today on fintechbytes we're asking the bold question, can we actually control AI before it controls the future of finance? When we talk about pausing AI, what does that actually mean in practical terms? Slowing development, strict regulation or something else?

Well, the bottom line is that AI is only controllable in a limited sense. We can slow it down, definitely. We can pause it, definitely. But we need to have some kind of global framework that we can collaborate and work together.

example, CERN Switzerland is a... a collider that has scientists from all over the world working together to make it an international lab for the advancement of science, put it in simple terms. So we need to do the same with AI. And that's the thing that's not happening because AI now is in the hands of corporates.

And as long as AI is in the hands of corporates, there is the component. of profit and the element of market share, market dominance and having your own customer base. when you work as a public institution, you're working for your country citizens. You're working for the of the global population.

So that's the biggest shift that we have to do. We have to move from the corporate to the government public sphere. and it should be even open source, some of it, and then we can do something about it. When it's corporate, becomes difficult to control.

To not give you a living example right as we speak. Anthropy came out with this very powerful ⁓ called Mythos. It's not ⁓ a week and they're scared of it. I'm in first.

Cyber attacks. for cyber attacks is so efficient. It finds bugs, you know, that nobody else have found. Even bugs that have been in data centers and systems for the past one bug was 27 years in the system and nobody noticed it.

And it's very accurate. So Anthropic realized that. And what did they do? They went to the government and to the big tech companies and said, listen, you know, we'll open it under this thing called the the glass ceiling, they open it up to certain companies and to select few experts in universities and the government to see what they can do, how they can use it to the benefit.

So basically to counter cyber attacks, not to let it get in other people's hands so they can use it for cyber attacks. Well, unfortunately, a day or two ago, some enthusiast AI cyber attack specialists, they hacked a simple code to just figure out where it might be stored on the servers. And now the model is out and about. And when I mean with out and about, it's not taking a walk around the block, it is on the internet.

And that means it's somewhere that anybody can access it. you know, it can end up in the dark web and then you know the criminals they they they hired the experts to find it and use it and then before you know it they they will use it for nefarious and malevolent intentions they will they will use it for for to gain money to to hack and and to use it for ransom attacks and so on so there is already a living example why why AI in private hands is never guaranteed. And for one thing, even if you would catch the person or the group or, you know, just hackers, they can say, well, you know, we just ⁓ it on the internet.

You cannot even do something about it. But it's in the hand of the Pentagon ⁓ or US government or the British government or the EU, you know, breaking federal laws in the US and you face jail time. And in Europe they can also take you to court and then you can go to jail. Whereas in a corporate setting, worst case scenario is you get a slap on your wrist or they say don't do it again or they might say you have to pay us a fine.

If you're like a hacking company or a bunch of hackers, you just don't pay up or you go bankrupt. So this is the key element here. It needs to move from private and corporate profit seeking entities to government and public institutions that have the interests of the general public at not the interests of the investors ⁓ the big tech companies. I also want to add, think pausing AI doesn't mean pulling the plug on innovation because it gives us a lot of possibility.

It means putting guardrails in place. ⁓ Think of it more like controlled deployment, restricting high risk use cases, enforcing audits and slowing down where the consequences are unclear. And that's something where the EU is ahead with the EU AI Act, but a lot of countries are still lagging behind. Definitely.

Is it even realistic to pause AI innovation in a globally competitive landscape? Well, by the looks of it, it doesn't seem like it's going in that direction because everybody is just building data centers because the more data, the bigger the data centers, the bigger the LLM models, the bigger the that is ⁓ used training. ⁓ So ⁓ they much scraped all of the internet. So now they're looking at synthetic data to create data that they can use for what specific AI models want to build.

Now the problem It kind of hit everybody in the face because AI came out of nowhere and now we're looking at models that are like beyond belief in efficiency. And the problem is, yeah, and it's beyond human comprehension. The AI models black boxes that not even the ones designing it and using it can explain what's going on and why do they have hallucinations. They're trying to figure it out.

⁓ that's the reason why should pause it in the first place. Let's get a grip on what we have. and human comprehension. and use it to do the narrow intelligence what we have.

So what is narrow intelligence? Narrow intelligence is like, okay, you teach AI to build ⁓ ⁓ that can beat a chess player, which we've done ⁓ years ago. ⁓ AlphaGo, we beat the the most Lee ⁓ Sedol, who is from Korea, ⁓ and he's number one Go player in the world. He got beat.

And then, you know, we had Alpha Fold where we were ⁓ talking about proteins, ⁓ would help in medical research. Now that's something to look at, you know, so we could deploy whatever powerful systems that we have at hand and focus on that. Let's make new materials. Let's make new medicines.

⁓ do something for the health, climate change. Let's nuclear fusion where we would solve the electrical problem that we have with those data centers. which are going to consume massive amounts and they are now emitting green gases. They are destroying the planet.

And everything that we were talking about zero carbon, ⁓ know, carbon footprint and 1.5 degrees has just gone out of the window. So that's the thing that has to happen now. But instead of that, we have everybody going to IPO looking at trillion and two trillion valuations, SpaceX.

Open AI is up there also looking to float and you have Anthropic, they have private capital, private equity, which is money coming out of nowhere. It's not in any kind of system. It's just money from private and trusts and all that. Just trillions is made available to spur the race.

So the focus from the industry is wrong. We want AGI super intelligence. And the rest of the people, us, people say, you know, we don't hate AI, we don't say stop it and pull the plug on it. Let's slow it down and use it for what is good for us.

know, you could devise a new economic model that works for everybody. And then people would be more inclined than to rebel and protest against AI. They might be more inclined to adopt it and use it, which we are doing on a daily basis. if you look at the stats, It's gone up the AI usage.

So people using generative AI like Chai GPT and Claude and so on. It has gone up, but we're still below the 50 % across Europe and the UK. Now I don't have the stats for the rest of the world, but still 50. So we still have room to grow there.

So, you know, we can have a winning situation, a win-win situation for all of us, for the suppliers, the creators, the experts, the tech companies, the government, everything. Let's work together. and make it work for everybody. Because AI is a powerful tool, the genius at the bottom will never put it back in and we can use it to our benefit.

So let's do that exactly. AI is too economically valuable and globally competitive. think that's the key point. And if one country pauses, another country will accelerate.

The real question is, we pause? It's can we coordinate? ⁓ that's much harder. And I think that's what we already previously discussed, is to be on the UN level.

yeah. consider excuse that need to be ⁓ first in the ⁓ AGN or superintelligence, this argument is absolutely useless. China will give in to the US and the US will never give in to China. So we're going to have two ⁓ systems running parallel.

opposing systems. opposing systems and there is no finish line and there is no win. The only one that's going to win is AI itself. AI one day will be able to replicate, reproduce itself without any humans.

And you don't know how it's going to react towards human beings. It's not to say that AI is an evil spirit that just came out of a computer and is going to just civilization. It's not that. AI might just see us as an impediment, an obstacle.

And basically we are, because we are the ones that we are building and destroying planet Earth and outer space all on our own initiative. We are the ones that do the polluting. We are the ones that do everything wrong. And we keep on going.

And now the excuse is we need to have AI, super-intelligence, AGI, and building these data centers and all that. And that is the... or is it threat? wrong mental setting that we have in the corporate world and the only thing that makes it they justify that we are going to make profit down the line and we're going to create abundance abundance who ⁓ which are we talking about and here comes the reality the math the economics the numbers just don't add up none of the companies that are creating all these fantastic super duper models with some exception is maybe Anthropic, which has ⁓ targeted a nice, out a little nice ⁓ niche in the corporate world ⁓ and is making All the other companies are not making money.

They're losing money. Facebook and all of them, they are losing money. ⁓ They don't even the money to invest. And we're not talking about millions of dollars anymore.

We're talking about billions and trillions. So they don't have the money to do that. So they go to private equity. And so there's a whole dynamic coming into play that was never there before.

And at the end, they're going to control the markets, the private equity, because they're going to say, well, you know, say we pull the plug on whatever loans you have, whatever money you owe us, you know, we just pull the plug on you. So we are shifting from government control supervision to private interests, which is totally insane. ⁓ that's the reason why it's not going in the right direction and it needs to stop because even if we build the data centers that are planned today the income is not going to come tomorrow it's going to come in a couple years time in the meanwhile people are going to lose their jobs there's going to be massive pollutions there's going to be energy crises we're going to have water shortages and now with the middle east conflicts we're even going to have famine and food and hunger across the world so we are Our priorities are completely skewed.

And then they are saying to the rest of the world, don't worry. You know, the five people that are at the top, Simon Altman's and, you know, Elon Musk and all that, and Jeff Bezos, like, don't worry, we'll regulate ourselves. We're to make sure that everybody's going to do good. We are going to fire whatever we can and replace it, all humanity, whatever we can with robots.

But don't you worry because we got it under control. No, that's not the case. And it's sickening to ⁓ them talk on TV ⁓ it's a great decision when in the end they're saying, we're just going to destroy the world we know and the world you know and live in. That's the underlying message.

We're going to destroy everything and you're just going to be quiet and be happy. So now it's going in the wrong way. ⁓ that's the reason why we have the worldwide protests ⁓ the pause actions and so on, because we are moving ⁓ AI adoption That's the underlying message. to AI fear, ⁓ every negative word.

We are looking at it now as ⁓ a sort digital COVID. That's how we look at it. And it's only a matter of time that we're gonna have some kind of lockdown because we're gonna have some zero day action where complete infrastructure is paralyzed because of some cyber attack. And it's only a matter of time.

It's gonna happen. It's not a question of. if it's a question of when it is going to happen and the people wake up is like, oh, what are we going to do? And again, I'll give you another example.

We were without energy in France, Spain and Portugal. Well, part of France. Now, okay, there was no foul play, it wasn't a terrorist attack, it wasn't computers. But it goes to show that if one infrastructure is debilitated or rendered non-functional, when It affects the entire country.

People couldn't even get money to get to pay a bus fare to go to work. So now just imagine if on the other side of this problem there is a robot that just kind of went tilt and just disabled the complete transport system of London or all the airports in the world. Can you imagine what kind of impact it would have on society? And nobody's ready for that because the problem is even the react time You know if something happens, there's a cyber attack, it takes time Even if you have the best experts encoders and hackers it takes time to set everything back on track and to reboot the whole system and get everything back up and running Now if you have an AI that is doing all of this You could send an attack out and while they're repairing the first attack send out a second attack and a third and fourth and in an endless loop It's really You know, you don't have to be a cyber attack expert or a cyber security expert.

⁓ know that it's a system that once it gets out of hand, you cannot control it. It just keeps on going. So ⁓ have now the life example of Mephists and I'm really worried about it and I can't, I hope that they're gonna ⁓ that out because it is, yeah, stepping in like it is, but it's national, Step in. is like critical government security that has to address it.

So ⁓ see how that pans out Who should have the authority to hit pause? is it governments, regulators, or the companies building the tech? The ones that should have their hands on the button, Trump, Altman and Bezos, they should have the pause No, I'm being silly, ⁓ but it of looks that that's what they want. It shouldn't be their hands.

And actually, Sam Altman got kicked out of OpenAI by the board on the wording, the wording is the trustees, the trustees, do you want to entrust Sam with hand on the button? That's the exact words that they use. So ⁓ while I'm around about it, he actually got kicked out. And because of influence and power and all these shadow powers in the back, they got him back into the company and he kicked out everybody else or at least He kicked out half and the other half left.

For example, Dario Amadei, who then opened up Anthropic and is now beating him at his own game. So yeah, I think OpenAI days are counted because he just don't have the money. The numbers don't add up. that's again, the other side of the coin.

Everybody talks, we're going to do this and we're going to do that. Financial markets, they look at the facts. They look at numbers. They look at like, okay, you're going to spend 1.

4 trillion. Anthropocene. How are gonna pay for it? You know, what you're say?

We're gonna get some more money and raise some more do another IPO in between or a bigger IPO or another issue of shares or whatever It doesn't make sense. So We're living in very turbulent and concerning times and ⁓ me it has to be handled by government There's got to be ⁓ regulatory framework binding globally around the world that we all play by the same rules And it's not an impossibility. We did that for climate change We did that for nuclear arms. We do that for the financial markets.

When you do something wrong in the US or you do something wrong in China or in Africa, you go to jail. That's as simple as it is. Those are the rules. You can't break the rules.

And the same we need to have with artificial intelligence. It cannot be in the hands of a few companies and five people calling the shots on eight billion people around the world that are opposing it. It's insane. I agree with you.

think it can't just sit with one group. need legal authority. Companies need operational control and independent bodies need oversight. So if any of those are missing, the system breaks and I think it can have disastrous consequences.

What the biggest risks of not having control mechanisms in place for AI? ⁓ one, loss of control. I think we saw the example also with Anthropic where he decided to not co-work the Pentagon, ⁓ but OpenAI did. Well, that's the other thing.

We have a loose cannon regime in the US that everybody hates, like from Trump onwards down. mean, downwards, just about his whole cabinet, whole government, even the Republicans are turning against him. mean, you know, ⁓ Iran war is an example already of total of ⁓ credibility on world military scene, because you're not winning the war, they're losing the world. And destroying the economy and the people at the bottom are going to pay the price without ⁓ food.

That's the price that the people are going to pay. But it's not going come tomorrow. It's going to come in a week or two ⁓ a couple of months. But it's going to hit us everywhere.

And we're going to see higher prices. So, you know, that's another thing. ⁓ Because is the whole point. We are now living in a totally interconnected world.

Through the internet, have come even closer. Things are moving at lightning speed. ⁓ Before... what used to be done took days, weeks, months.

You had to send a letter, you to call, you had to send the fax. Who remembers the fax these days? Nobody. ⁓ it's a click of the button and it's going through fiber optic cables ⁓ lightning speed.

So things change immediately. ⁓ then the computational power is so rapid that there's no following for a human mind. I mean, he can do a million calculations within a second. We're not talking about quantum computing.

When that comes into play, we're talking about a billion calculations per second, even ⁓ per nanosecond. So these are things that are coming our way. And they have legislation which is like a turtle. So you have on one hand, you ⁓ a ⁓ running and have a turtle trying to slow down that leopard.

⁓ trying to catch up. trying to catch it, you know, it's ridiculous. it's not going to change because ⁓ takes time because you need to have people around the table and taking a vote and deciding. You need to ⁓ objectively and not be influenced by the powerful oil lobbies ⁓ military ⁓ lobbies the ⁓ tech lobbies.

need to be able to decide independently as much as possible if that's... still possible these days. But the problem is that you need to establish a framework. So you have this framework in banks.

We the framework for GDPR, ⁓ was for Europe and has been adopted by the rest of the world unofficially. There is some kind of, ⁓ know, we play by the same rules. These are the rules here and we don't move the goalposts. And with AI we move in the goalposts.

There is no plan. There is no... strategy, is no referees, there is nothing. There's just five people controlling all of it and just calling the shots.

And we're going to do this, we're going to do that. And when things go wrong, they're like, oops, ⁓ you know, I didn't, you know, how did that happen? Yeah, the whole thing is, the whole thing is about not letting things happen in the first place. So it is really, really, really challenged.

We have on one hand an unregulated industry as far as AI is concerned. And then we have the government, which is too slow to legislate and to be able to bring about change. Even if they pass an act, the EU act is quite an impressive feat because it got 28, 27 countries around the table to decide on one thing and it applies to everybody that does business in the European zone. that affects Americans, the companies, it affects the Chinese, Koreans, everybody.

So it's amazing that they did that. And that needs to happen again. But if you do now in the US something, Congress is deadlocked between two big parties fighting each other, the Republicans and the Democrats. ⁓ is this polarization, this division, this ⁓ and right and all this ⁓ and silliness that's going on.

And you have then this whole circus the circus king, Trump calling the shots and nobody, it doesn't make sense what he's saying or doing. And that has to change and that has to stop because as long as we have a loose canon deciding on the future of 8 billion people we are in trouble I think again, most frameworks weren't built with AI in mind. They assume human decision-making AI breaks that assumption. So regulators are constantly playing catch-up.

Exactly. Exactly. Exactly. ⁓ global AI governance work?

very tricky, but it could work because we have done it before. The Paris Accord, had, if I'm not mistaken, 158 signed the Accord. We have the Nuclear Arms Proliferation Treaty, which has been signed by the major economic powers and military powers, including China, Russia, and the United States. We have GDPR, is being used by everybody, but it's not officially put into legislation.

have the International Maritime Organization where they have ⁓ a assembly authority in relation to maritime law and everything. It's semi-governmental but it's not really governed by ⁓ in own sense but whatever they ⁓ bring all the resolutions they get adopted by each country that is signed up to the IAMOM. there is possibilities to do that. It just takes effort on everybody's part.

And Guterres from the United Nations, he has already established an AI commission and we need to have at highest level at the United Nations, just an AI assembly, not even the general assembly, just like you have a security council, we need to have an AI council or an AI assembly because the council, ⁓ the council only has like the seven big major... ⁓ military powers like Russia, China and the US. But we need an AI assembly where every country has one representative in that assembly and they take a decision and then it's done.

And everybody signs that agreement ⁓ whatever comes out, whatever proposal is coming out of that assembly gets ratified. That means brought into law, turned into law. by each respective country that signed on to that assembly. And basically when you do something like that, like for example, NATO is another example that we have.

And again, the Americans are pulling out or threatening to pull out. NATO brought so many countries together that say, you know, we're going to stop this killing each other, World War II, World War I, we're not going to have World War III. And it pretty much prevented it from happening. So far, it's a good eight years down the line.

And we need to have that again, an AI assembly. that prevents people from doing things. You sign up to the assembly and you are like a strong front. somebody does, yeah, and if somebody doesn't sign up, you can exclude them completely.

And that's the power of such an assembly. If let's say North Korea, which probably wouldn't join the AI club, you know, so to speak, you cut them off completely. Or maybe Israel. I don't know if Israel would join.

You're not... or certain countries might not want to join, maybe Switzerland says, well, you know, we don't care about AI, we have sovereign AI. You know, those are the countries that might not ⁓ join assembly. But if you have an assembly that brings about ⁓ resolutions get taken on to ⁓ and made law, it's a very move.

And it just sends a clear signal to everyone. Guys, girls, we're gonna play by the book here. We're gonna follow the rules. Because that's the only way this is gonna work.

If everybody is on the same page playing by the same rules and the goalposts don't move. And what are some real world examples where AI systems have gone wrong and did we learn something from it? Well, there's a couple examples. seen the end of it.

There's been some ⁓ situations that have caused ⁓ havoc. And then we have this concept called Zero Day, which basically ⁓ is when ⁓ a computer system a complete infrastructure and, you It affects not just one person, but entire countries. the zero-day concept is something that is real and could happen any day. And that is something that concerns everybody.

⁓ then you have what they call a single point of which can be ⁓ anything, any infrastructure, any organization, any company, meaning that if one thing goes wrong, the whole system collapses. For example, if you have a nuclear power plant, it needs constant water cooling. So it doesn't heat up and explode like it did in Fukushima. Once you turn off the cooling, you know, it starts leaking and then, you know, you have a nuclear disaster, which happened in Fukushima.

So that's the single point of failure. So what it did is, which is for me incredible. ⁓ not a nuclear scientist, but I understand that you need cooling water. to cool down the nuclear plant.

So what did the Japanese do? They built the nuclear plant next to the water so they can just pump out the water from the sea, desalinate it and use it for the cooling. But they built it next to it. They could have just built it up the hill because it's hilly area.

They could just put it up, let's say, above water level 10, 20 meters. The plant just move it inland, up. then they could have had longer pipes, pump up the water to the high place. ⁓ The tsunami, even if it would have hit that part of the land, it would have not shut off the power plants and the cooling systems.

So this is a single point of failure that you need to look at from every aspect. And that works when people put their brains together and their ideas and work together to avoid stuff like that from happening. But single point of failure is everywhere. It's in banking systems, it's in infrastructure.

electricity grids, governments, you name it. And that needs to be taken care of. And it's very crucial that there are safeguards built in to avoid things like that from happening. ⁓ that's why we do have the EU AI Act, which says everything has to be under government supervision.

⁓ the US military that is now boycotting, the ⁓ is boycotting, and thropping because they didn't want to play by their ⁓ rules. you They have systems in place that specify that any robotic or automated system must be supervised by a human being. Meaning drones or robots or whatever that goes without any human person inside of the machine, so to speak, is supervised by a human being somewhere far, far away. you which gives a good advantage is that when you send out robots and they get damaged, nobody is hurt because, you know, it's different when you send people to go fight a war.

So they have the systems like it is in the US government. They have that in America. They have that across the world. So why don't we do the same with AI?

Looking ahead five to 10 years, do you think we'll have meaningful control over AI or will it outpace our ability to regulate it? That is a big question that I don't think anybody can answer and I think more people are more scared than feeling secure and if you ask the big experts they say there is a 5 to 10 percent chance that we lose total control you know and in my extinguish civilization some are saying it's 40 50 other people saying it's a 50 50 some people say it's 100 what is guaranteed and between the experts is that if you do not do something now the window of opportunity to keep it under control is closing very fast and we're talking one to two three years so that's the reason why everything needs to be done now ⁓ control and...

Nobody's saying let's not use control. Nobody's saying we don't need AI and we're not going to use it. Just let's slow it all down. You know, let's forget what the corporates want to do and make more money.

Let them do other things, but let us do what is available right now. We have enough computing power as we speak, and we have enough artificial intelligence that we can use to do things that would make a massive difference. for all the people on this planet, including the governments, including the corporates, including everybody. So let's focus on that.

What's gonna happen in 10 years is unknown at this stage. I don't think even if you have a crystal ball, there's so many variables and so many things that go wrong because you're not only looking at AI now, we have geopolitical tensions. We're on the verge of World War III, to put it simple. Then we have climate change, Trump says it's a hoax.

unpredictable. I don't know where he gets that from. And other people it's real. And if it's not real, the pollution is there, the poverty is there, the hunger is there.

So climate change is affecting people all over the world. Either you have too much water or too little water. Either we have drought or we have floodings, ⁓ have landslides. ⁓ is not just...

This is nature at work. This is not man-made. We caused it. We are the causation of all of that because of irresponsible reckless...

pollution of the world. But this is natural disasters that we have no control over. And, you know, again, we have been hit by a huge tsunami. Tsunamis can happen anytime.

you know, the polar ice is melting and chunks are breaking off. You know, you'll see water rising in places and entire countries just disappearing. And then it's only when that happens that everybody says, oh, hold on, what just happened? We need to...

Let's wake up. Let's wake up, need to do something here. They always react when it's too late. That's the craziest part of government anywhere in the Hmm.

Yeah. I think with AI we'll have better tools, but AI will still move faster than regulation. Control won't be absolute. So it will be about managing risk, not eliminating it.

And AI isn't just a tool, it's becoming infrastructure. People are getting so used to it and the real challenge isn't building it faster, but building it responsibly. So the question we'll leave you with. Exactly.

better. If we could pause AI today, would we even agree on when to press play again? that's a good question. I mean if you cannot agree on where we're pause everything let alone trying to agree on when we gonna Start again.

I think what is meant by pause is More like let's slow it down. So we're not gonna stop all the big tech companies say okay stop now finish What we would probably what everybody wants is okay. Let's not start building all this and you 33,000 data centers across the world and one in the space by Elon Musk. Let's pursue recklessly AGI and super intelligence.

Let's get the AI that is working and make it work better. That's, ⁓ the bottom ⁓ And it will do everybody a lot of good that if we could use AI. And one day we found a cure for cancer or Alzheimer's or Parkinson's or we develop a new material that we can use, new medicine. We let's say, nuclear fusion where we have ⁓ unlimited clean energy abundance, would be something that people say, well, know, AI, you know, let's go for it.

⁓ then people would be like, ⁓ probably start doing the opposite. ⁓ want AI, ⁓ want AI, you know, that's probably going to be the start sign when people going to say, yeah, ⁓ use it for other things, you know. But the vast majority of the world, in my opinion, is looking at what can we do positively with AI? How can we use it responsibly?

And not recklessly. That's the thing. And the crazy part is, it's only a handful, and I mean really a handful of people that are deciding the fate of the world and influencing everything and affecting everything without our permission. These people didn't get voted by us.

They didn't get, they ask our permission and they shouldn't have that permission. And then you have the tech companies. Okay, you're a tech company, have billions and trillions. Who...

gives you the right and the authority to decide over the planet, over the population of the planet. You have no such right. And that's the thing that needs to change. The mindset has to shift from not what you want to do and make more profits, like what we can do with AI to benefit humanity.

That's why I'm a firm believer in human AI, keeping humans in the loop and let us decide for ourselves. That's what I say. That's a wrap for today's episode. Thank you so much for tuning in to Fintech Bytes.

And if you enjoyed this episode, please share with someone. And thank you, Jerry, for your insights. Thank you, have a nice day.