저자: DanIves · 외국인
요약 보기
AI 수익화 단계 진입을 강조하며 하이퍼스케일러 실적과 주문이 기술 섹터 성장의 확인이라고 주장. 칩 수요/공급비 13~14:1, 자본지출 1달러당 5~6달러 승수 효과 제시. 반도체 공급 부족과 메모리·칩 업체 '골든 에이지'를 언급하며 Palantir·Apple(중국 AI 모델)에 긍정적 시각. 나스닥 13,000 목표는 AI 혁명 진행률(10~15%)을 근거로 유지. 차입 비용 상승은 자본지출 규모 대비 미미하다고 평가.
다듬은 글
It's been such an exciting earnings season, and the tech sector's reaction has been very mixed. Very strong results, very weak ones, slight fluctuations, big jumps—we've seen it all. What do you think are the real drivers of sentiment at this point? Listen, I think earnings have really caught up with estimates.
I mean, if you look at the results of hyperscalers and then neo-cloud services like Nvidia's CoreWeave and others , the monetization phase is beginning. And I think that was the kind of confirmation that investors wanted to see —not just capital spending, but the beginning of monetization in businesses moving into second, third, fourth derivatives.
And I think that's why we continue to see the growth of the technology sector. This really became something of a tipping point in what we're seeing with the AI revolution. We heard yesterday from Chuck Robbins of Cisco on our channel that Cisco stock was under pressure, despite very large orders. That's $4 billion per quarter from some hyperscalers.
And he emphasized that these were very large clients with very strong balance sheets that placed these orders. How much should we pay attention to this, because it seems like if something goes wrong in this very cyclical ecosystem, it could disrupt the demand story. So how do we evaluate the orders we see today compared to the risks that still remain in the system?
Yes, supply and demand, and we estimate the ratio for chips to be 13-14 to 1. The reality is that everyone is lining up, when it comes to enterprises, you need to reserve your spot for data centers and GPUs, because the reality now is that the use cases are starting to multiply across industries. And I think Microsoft sees this in the example of its customers.
Robbin, Cisco, they see it. I think IBM sees this. And if you look at everyone, this is the confirmation side. I think that's why this reporting season was so important. I mean, I think especially when you look at Microsoft, it really stands out because it shows that businesses are doubling down. And because of the capital expenditure, for every dollar of capital investment, there is a five to six dollar multiplier for the rest of the technology sector.
So betting against that, you know, it's actually a very, very difficult bet, given that we believe we're still in the early stages of this growth cycle. Don, you mentioned the word "confirmation ." Let me turn that question back to you, because it seemed to me that Nvidia had confirmation from every top Wall Street financier this week.
Everyone rushed to sign up for this half-trillion- dollar computing power asset class. It is hard to believe that these extremely experienced investment managers could have missed anything while conducting due diligence on financial arrangements. So, how do you evaluate this $500 billion in funding?
I think the concerns about supply chain financing —we've been talking about this for months. I...there are certain aspects that have yet to manifest themselves in the end. But this was a huge confirmation that the market needed to see, because these are the best banks, in fact the best banks in the world, who have now made a bet on this .
And when you look at what AI godfather Jensen and Nvidia are doing, you see a supply and demand problem. I think everyone recognizes that—both in the industry and from a funding perspective , and actually in the fact that computing power is becoming a separate asset class. And it's not stopping, because we're still at a point where the US and China—it's, of course, a race between two leaders, but we're starting to see it in the Middle East, in Australia.
And, you know, there are also sprouts appearing in Europe in terms of AI , but where does the computing power come from? That's why you're looking at Korea— you can count the memory and chip manufacturers on one hand, they're currently experiencing a real "golden age" of demand. Dan, what do you think about the talk of Silver Lake acquiring Workday?
What is your opinion on this ? I don't see a "SaaS apocalypse," you know? When that happened, it only proved that the smart money sees real value. I'm not saying there aren't nuances for Workday, Adobe, or Salesforce, and that they don't face challenges due to AI, but it just highlights the value of their customer base.
And I think this is very important, because against the backdrop of the “SaaS apocalypse” and the fears associated with it, software will become a key part of this AI revolution. Was that today? No. But if you look at the use cases in the 6, 12, 18 month perspective, Palantir is, after all, a software development company.
And they saw it in advance, because they know how to, you know, look a few steps ahead. Others are just trying to follow them. When you look at the P/E ratio and evaluate whether it was worth it, it is a very positive signal for the market in its development. Don, Cameron and I talked earlier. The VanEck Semiconductor ETF has risen just over 2% over the past 3 months.
Coffee futures rose 13% over the same period. Are we focused on the wrong thing? I think the reality is that when you think about where the growth is right now , it's the market that's going to be the driving force. We could talk about crypto, the Fed, or commodities, but the market will move up or down precisely because of the technology sector , as we see huge capital investments and derivatives in industrial and energy sectors around the world.
So, of course, there are other asset classes that are worth focusing heavily on , but technology is the focus right now . They are under the spotlight on stage. I think any investors betting against technology are watching this from a waiting position because they don't see AI in their spreadsheets.
Dan, let's move on to Apple. They are training a large language model for the Chinese market using Alibaba's technology. Is this more than just Apple trying to fix its business in China, where AI is developing very quickly and potentially disrupting their business model? Is there more to this than just a story about China?
Yes, I mean there is, because this is a company that has actually been at a loss for two years on AI issues. And now, finally, with Gemini and a real AI development strategy, they can start going on the offensive. 20% of iPhones come from China. And when you look at Alibaba, it was a kind of two-way collaboration, and that's the bet they're going to make.
I think what they're doing is half defensive and half offensive. But I think the market is finally starting to realize what Apple is doing, because on the consumer AI highway, they will essentially become tax collectors. Donna, I want to ask you about some of your predictions in the past. You predicted 13 thousand points for the Nasdaq.
This year we are still quite far from that figure. One of our guests' thoughts earlier was that he liked Nasdaq as a place to invest. What is your opinion? Is it possible to reach 13 thousand points this year? I think the reality is that we are only 10-15 % of the way through the AI revolution. When you look at the technology sector , especially the results of the reporting season, I think we will see this in October and later this year.
This is the kind of confirmation investors are looking for, because when earnings catch up with expectations, stocks that seemed expensive start to pay off, and derivatives come into play as well. You can stock up on popcorn while watching what's happening in the tech sector. I think it was very, very risky to play against him, given the real flurry of capital spending that we're seeing right now .
Done. I am also interested in changes in the cost of sovereign debt, because even though expectations for interest rate hikes have eased due to lower inflation this week, the cost of borrowing for the US government has still increased. And this, of course, has consequences for all other borrowers. And hyperscalers, large technology companies, are now partially financing the development of AI infrastructure through borrowing.
They issue new debt obligations. How sensitive are these investments and huge amounts of capital expenditure to rising borrowing costs, the cost of money? Do you think this could hinder things in any way? Yes, that's a great question. Look, I think these are still insignificant amounts compared to their balance sheets.
Because we are talking about a trillion dollars of capital that they have and a free cash flow of 300-400 billion per year. But you hit the spot: investors want to see monetization happening. But look at Oracle—it's a good example of what investors couldn't digest, and the stock started to react. So I think we'll continue to see this tug-of-war as companies raise capital through equity or debt, but this is an arms race, and you can't just stand by.
Because if you stand aside, you will miss everything. You'll be, you know, using a VCR, essentially, right? Or a typewriter —however you prefer to think of it. They understand that there is a window of opportunity, and they need to maintain a balance in raising capital so that investors see prospects for the future.
Comments 0