저자: ChamathPalihapitiya · 외국인
요약 보기
AI 토큰 가격이 '배럴 단위'로 급락해 수익이 downstream으로 전이되기 어렵다는 우려를 제기. IBM 관련 논란을 AI 가격 경쟁의 신호로 봄. 나스닥·S&P500이 기술주 매도로 밀리는 가운데 S&P 50일선 유지 여부가 중요하다고 강조. 엔비디아는 밸류에이션 부담이 줄며 자금이 메모리 등 공급망으로 이동 중이고, 마이크론은 메모리 수급 호조로 현금흐름 기대가 크다고 언급.
다듬은 글
Glad to see him here again. Chamath Palihapitiya is with us. He is the founder and CEO of Social Capital, CEO of 8090, and of course, host of the All-In podcast. And sir. You are dressed, you are wearing a suit, no less. For our sake. I...this... My job is to sell enterprise software. This is what I do.
This is what you do. So before we move on to anything else, and I want to discuss everything that's going on in the Valley, AI, everything in the world. We're going to go back in time, talk about SPACs, because people have been asking, I know these are your favorite topics, but just... your thoughts on the IBM situation, how much do you think this has broader implications, or is it maybe just IBM-specific?
I think we know that there are several companies whose revenue is growing by a billion dollars a day. And at some point, the ecosystem downstream also has to start making money. And then the end buyer of these tokens should also make a profit. So at some point you have to ask the question: if two or three companies are generating $200, $300, $400 billion a year, and that amount is doubling and tripling every year, where is the rest of the money going to be made and who is implementing it in a way that those profits can be made?
And I think you're starting to see things go a little awry, at a very basic level. If you think of AI as oil, let's use this analogy for simplicity: a barrel of crude oil is worth, what, $86 right now? Now, unfortunately, the price has increased. Yes. No, no. WTI is 8. WTI is 81. 81. Okay. Let's say a barrel of intelligence equals a million tokens.
Good. You can buy this barrel for $26. M-hm. In a really good Anthropic model. You can buy it from OpenAI for $26. Anthropic's latest model will cost you $56. M-hm. Elon sells you a barrel of intelligence for a dollar. Zuckerberg is going to sell it to you for a dollar fifty. Demis and Sundar are trying to sell it to you for a dollar.
The Chinese will sell it to you for 50 cents. So this rationalization is inevitable. We have the same resource, which has such an insane value. If you bet on one of those extremely expensive barrels of intelligence very early on and try to pass on those costs, you may run into difficulties later on, and this process should come to a natural conclusion.
And do you think that's reflected in the IBM story, or do you think it's just a broader comment? I think Arvin did a great job. Listen, there's no denying the trajectory of the business, and he was able to change course and orient it towards cloud technologies, right? And I think he deserves a lot of respect for that.
But the reality is that everyone in this ecosystem, including me, is trying to figure out how to price these products so that the end buyer earns more and grows faster. And that still remains in question. I think you're getting to the story really, I mean a lot of the strength is coming from your Micron, AMD, Intel, SanDisk, and we've seen a really noticeable pullback in those stocks this week, which is only amplified today with the NASDAQ down almost 2 % this morning.
Good morning, Jen. Yes, there are many topics worth discussing, and they are very controversial, aren't they? Let's start with inflation. We had a great week on inflation. CPI and PPI data are all positive from an inflation perspective. We had geopolitical risks, the sixth day of bombings in Iran. We had a chip sell-off that, frankly , at one point late yesterday morning, 10 of the 11 S&P 500 sectors were in the red, but the one that was in the red, information technology, continued to pressure the entire market until the entire market collapsed.
Now, excuse me. Today, I think, will be a very critical day. Why? I think about 73 points below where we are now in the S&P 500, which is , frankly, right where the market is going to open. This is the 50-day moving average for the S&P 500, which we have been holding above for quite some time. I think from a trading perspective, with everything we have to understand, take in and digest today, watch this 50- day moving average for the S&P and see where we close today.
I think this will be vital. The NASDAQ, which is so heavily dependent on technology, has already passed its 50-day moving average, but the S&P , if it's really a rotation from some of these names to others. Keep an eye on where the S&P will close today. Yes, I will say that it will be interesting, because now we are not just seeing the weakness of the NASDAQ, which I do not ignore, but I find it more telling that even the S&P 500 futures have gone down.
I mean 1% today. We have been hovering around these historical highs for some time. Therefore, we become greedy. As of now, we closed about 1% off the all-time highs yesterday. I'm not saying that this pressure is cause for concern, but there are certain themes brewing beneath the surface. Another one, of course, is related to energy.
We saw a rise today, and oil is over $80 again . Kevin, what are your thoughts on the overall energy situation, given that events abroad continue to cause volatility in global energy trade? Yes, it was a good week for inflation, but not a great one for energy, as oil prices rose from $70 and below to $80 and above.
Why? Well, six days in a row of bombings of Iran. Meanwhile, Iran was striking Qatar, Kuwait, and Jordan. They are attacking US military ports in Bahrain and Syria. So what we need, what the market needs—is some form or hint of something very unusual happening with Nvidia right now . For years, one of the main arguments against owning stocks was their valuation.
Investors looked at Nvidia's explosive growth, huge market cap, and dominance in artificial intelligence and concluded that excessive optimism was already built into the price. But now the conversation is starting to change, and the company remains at the center of building a global AI infrastructure.
The demand for computing power continues to grow, hyperscalers are preparing to spend huge sums, but Nvidia's valuation has fallen significantly compared to the early stages of the AI boom. This creates an interesting situation: the company has not lost its technological leadership, the development of AI has not stopped, data centers are not disappearing, and the largest technology companies in the world are not abandoning their infrastructure plans; instead, investors' attention shifted.
The money that was previously in Nvidia has moved into businesses, and other parts of the AI supply chain . To some extent, Nvidia has become a victim of its own success. The market has become so accustomed to Nvidia's history that investors have begun looking for the next opportunity. And this search led me straight to memory, because the AI revolution doesn't just run on GPUs.
You can build the most advanced accelerator in the world, but without enough memory, network capacity, and manufacturing capabilities. This computing system will not be able to operate at full capacity. That's why Micron has become such an important part of the AI investment debate. The demand for advanced memory has created extraordinary conditions.
The offer remains constant. Customers need more capacity, prices have increased, and memory manufacturers have gained significant negotiating leverage. This combination has created extraordinary expectations for Micron's future cash flows. If the company generates even a fraction of those huge free cash flow projections that are being talked about, Micron will face a problem.
Most companies would love to have this problem: what to do with all that cash? And the answer starts with investment. Micron needs to increase capacity. The industry needs more memory. And the development of AI infrastructure is putting pressure on the entire semiconductor supply chain.
Comments 0