I'm Scarlet Fu.
The S&P 500 up 1.1%, the first back-to-back gains since mid-August. The bond selloff taking a bit of a breather. The 10-year yield coming down one basis point. Oil prices modestly higher. Brent is unchanged but we have Iran and the U.S. resuming military strikes. Israel indicating it is prepared to intensify its role in the ongoing war. The dollar is weaker today but the big mover is the Japanese yen, gaining as much as 2% now after the BOJ senior official struck a hawkish tone and traders speculate whether there would be intervention coming.
Let's take a look at some of today's equity movers starting with Tesla, shares up more than 6.5%. That is before its event in Boston where they will reintroduce driverless cars as part of the robotaxi service. Looking at Coinbase, gaining 9.6% after Bitcoin topped $80,000 once again. Looking at the decliners today, Tyson Foods leading those declines. 7.5% decline after the meatpacker cut its full-year earnings forecast again. Moderna shares down 3.6% after the stock lost the steam as analysts at Rothschild called it overvalued and cut its rating to sell.
The mix of assorted individual company names. Let's talk about volatility and equities because it has faded. The blue line shows the VIX now sitting at 14.5, well below the 12-month average of 18. The white line is the S&P 500's 20-day realized volatility which tracks how much the S&P has been moving day to day. The line goes up when daily swings are large and falls when the market settles into one direction or another. If you go to the far right side of the chart, it has settled down from its peak in early April. All of this may be the quiet before the storm because September is historically the worst-performing month for the U.S. stock market.
This volatility really screams late-summer. The weekend before Labor Day, almost, and look what volatility is doing. A lot of risk events coming up. Later this week, we have the jobless report tomorrow. Next week, we have inflation data as well that will be crucial for what the Federal Reserve does next.
Let's kick things off with Trevor Slaven. Given all the big catalysts that are coming up for investors over the next week, how much do you read into the fact that the stock market has managed to stay pretty much stable where it's at even with all these different dynamics and headwinds potentially?
I think there's an incredible dynamic in economic transformation that is taking place today. It is really this battle between economic resilience and maybe some resistance going forward in terms of the traditional setup. Maybe as a quick backdrop on the economics side, over the last three quarters 94% of all the real GDP growth has come from essentially AI capex. Investment in information processing equipment and IP. The rest of GDP that the other 85% contributors, household sector, those have been running close to zero for about nine months now. There is this building tension between maybe lackluster growth in some of the traditional parts of the economy where consumption is slowing down, wages are slowing down while the AI capex boom has put the economy on its back.
Putting aside the AI capex boom which we know is in no danger of slowing down, what will you be paying attention to when it comes to the rest of the economy?
Everything we look at is on the fundamentals and valuations and technicals. One, valuations across the market are pretty tight. Credit spreads at the index level are very tight. PE multiples relatively elevated, sentiment is pretty bullish. I would say the market has a pretty constructive view in terms of valuations. Against from the fundamental perspective, yes, the capex boom on the AI side is really impressive but there are some looming questions. We may be facing for the first time ever a structural decline in the labor force. We are seeing wages start to fall as well. I think there are some unanswered questions around how does AI capex compete with the labor market, what does that mean for budget deficits? To that point, the market has gotten to this moment where it is looking at maybe a slowdown in the labor market to say even if I get that slowdown, all that means is bigger fiscal deficits on the other end and therefore more supply. That takes me to what is the biggest story in markets today which is this massive supply we are seeing from the Treasury market. $30 trillion gross, $50 billion a week in net Treasury issuance. The new dynamic is on the investment grade side. Over $2 trillion of debt being issued this year. So much of that increase with the hyperscalers and other AI-tangent borrowers.
Where do you see that playing out? We know that hyperscaler issuance has been a big story and part of the reason why is we have seen yields hit the highs we saw around 2008 and 2007. What is that story this year?
There is the growing tension that this year dollars of borrowing is so immense. There is so much hedging that needs to take place across the issuance. There is this story in the background of concern around sticky inflation. Concern around the path forward for the Fed at the same time they are undergoing their own transformation. The one thing I would say as we get into September and October is we have seen so much of this investment grade issuance, $400 billion in issuance from the hyperscaler since the beginning of last year. Almost all of it is underwater today because of credit spreads widening and higher Treasury yields. You have a pretty substantial, almost $20 billion of market-to-market losses before the AI-centric bonds over the last year and a half. I think that at some point starts to cause a little more caution from lenders. I either need more clarity around how these business models will evolve or I need more margin of safety in terms of my evaluation. I think that could be a bit of a speedbump going into September and October because September will be another massive month of corporate supply.
We only have about 30 seconds left. You want to be cautious heading into September. At what point do you want to play offense again?
Because valuations are tight and because we have these unanswered questions, the supply is pushing investors down the road faster than they probably want to go of trying to navigate this transformation. You would love to see a little stability in the consumer or so much about it is a valuation story of can we get a little bit of a bigger margin of safety on credit spreads and PE multiples and then deploy more capital?
Such a pleasure for you to come in and for us to see you in person on the Thursday before a three-day weekend. Head of Multi-Asset Solutions at Baring's. OpenAI CEO Sam Altman joins the program live. It is a can't-miss conversation with the latest news out of the ChatGPT operator. All of that and more coming up.
Ciena shares on pace for the worst performance since -- beat expectations for third quarter results and issued a better-than-expected revenue outlook for the third quarter despite supply constraints. Joining us is the CEO and President Gary Smith. Thank you for joining us. Impressive results. Shares are down. What do you think investors are missing from these?
Generally, you think about AI infrastructure stocks and I think it is really the discernment between the winners and losers. Really a broad brush being played across all of that AI investment stocks as opposed to who are going to be the long-term winners and drive growth and operating leverage in the business.
Very interesting. Some of the analysts we spoke to in the back half of your earnings really pointing to high expectations heading into the event. In some ways, that indicates the company is a little bit of a victim of its own success. As you mentioned, there is a bit of a rift at the moment in the trade, the AI trade. Investors looking for winners and losers. What do you think Ciena needs to do moving forward?
I think we are part of that winning group. It is a critical enabler of AI infrastructure. You think about what's happened from the AI story so far, everyone has been very focused on compute, power, data centers, etc., and quite rightly so. Now that data is beginning to come out of the data center. Without optical networking, there is no operational -- operation of AI or monetization of AI. The network is underinvested writ large around the globe to be able to support the whole AI infrastructure as it scales up through training, inference and agentic use of AI and eventually into robotics. It is all about the network. The network is now the critical path.
You said in a press release that you talked about AI driving compounding waves of network investment. That suggests even within this once-in-a-generation AI infrastructure buildout, there is a waxing and a waning cycle. What does that look like? How long do those waxing moments last, and how long the waning moments will last?
I don't think you will get many waning moments and that is what I mean by the compounding effect. What you see is the establishment of the cloud connectivity amongst all of the data centers. On top of this, you've now got training on top of that, you will have inference traffic which is only just starting to happen. Then, you've got the whole adoption of agentic AI, when you think about that from a network point of view, that will increase traffic dramatically. On top of that, you've got the robotics piece, the physical AI, if you will. All of that is to come. I don't think you will see any waning moments over the next few years. I think you will see a multi-year massive buildup of optical capacity across the globe and within data centers.
Every tech hardware company faces supply constraints that is kinda preventing them from being able to fulfill all customer demands. Talk about what you are seeing and how you are addressing that.
The whole optical ecosystem is expanding and investing to meet this multi-year demand that I just characterized for you. We've been able to secure supply for the next couple of years to scale up our business and also we've got commitments at the front end of our business we talked about today will be leaving the fiscal -- backlog of over $10 billion. And frankly, if we can get all the supply from that, we would have larger revenues. This is not about demand, it is about ramping supply. In the industry are driving the investing in that supply.
Let's take out our crystal ball and look forward to 2027, what will be the key driver of growth next year? The acceleration in AI infrastructure spending?
It is going to be supply. We have visibility to next year. We already talked about early indications, we gave early indications that baseline growth for next year is 30%, we grew about 30% this year. We can grow even greater than 30% if we can secure more supply, which we are working through. It is not about demand in the optical network in space, it is about supply.
Very much a supply story. That is Gary Smith, CEO and President of Ciena. Coming up, the latest from Kris King, Chairman of Skyworks and the first female semiconductor company. This is The Close on Bloomberg.
Our next guest has spent more than 50 years in the semiconductor industry, roughly half at IBM before making her mark as the world's first female CEO of a chip company, AMI Semiconductor. Kris King is the Chairman of wireless chip company Skyworks Solutions and author of Breaking Through the Silicon Ceiling about her success story. She joins us live. Thank you for speaking with us. You've clearly seen your share of boom and bust cycles in this industry. Where do you think we are right now in the current memory chip boom? Close to the peak or nowhere near it?
I don't think we're anywhere near it. I have seen so many cycles, from PCs, the internet, mobile devices, now the AI boom. It is exciting. I don't think we've seen a time where we've expressed demand in things like gigawatt capacity. It is very exciting. I think we are early innings, as your last guest said, we are even earlier in the build.
How do we know when we are at or near the peak? What will be the markers?
It is always tough to see those markers because it kinda falls off of the cliff. There are so many aspects that demand drivers, right now we talk about processors, GPUs and memory as the main thing. But I think as we see this continue on, we will see other areas like power supply management and connectivity that are all ancillary. So I think it is going to take quite a while to see the rollout of this cycle.
One of the trends that are emerging now in this space is the whole idea of custom silicon where hyperscalers are designing more of their chips. How disruptive do you think this trend could become in the industry for traditional chipmakers?
I think it is a really interesting trend, and it could have a big impact. In the semiconductor industry, we've had these trade-offs that keep repeating themselves. Custom versus general. Things like centralization versus distributed. They keep recurring. I kind of think in the AI processor scene, we have seen the general processors, the XPUs, a company like Broadcom partners with hyperscaler as well as the AI companies. And then we see these companies getting into their own semiconductor capability. Not just design, but also fabrication. I think a great example of that is the Terafab, when we talk about SpaceX and Tesla requirements for tons of semiconductors. It is a very interesting trend, and it will be disruptive.
One of the trends we are seeing is in terms of the geographic divide, we have a lot of U.S. listed and based hyperscalers taking the lead here in a country like China. Maybe a few years behind on those trends. How durable is the U.S. lead in this space and when we might see China start to catch up?
I think for sure China has been catching up incrementally all the time. But in every cycle, I think we have seen the U.S. lead, particularly in semiconductors. Not only in design, but manufacturing expertise. We will see that continue. I see the U.S. being in the lead for quite some time to come.
When there are supply constraints like right now, investors can invest in new production facilities to meet demand. How quickly can a manufacturer build up that capability? Has that timeline accelerated in the last five years versus when you first started in this industry?
I think it is always a trade-off because we have more complexity and more connectivity. I think the cycles are still pretty long. Hopefully that continues to compress. When we think about the supply line to actually build a fab, there is a lot to it with land and energy. One of the interesting things about this cycle is will we see energy become a constraint? That is one of the things I worry about. I think our energy infrastructure can use some of the techniques we have used in semiconductors and networking all along. That would be interesting to watch as well.
I'm glad you diagnosed the problem. How would you ask policymakers to address that before it becomes a big issue that you are worried about?
I think that obviously, our whole energy and power infrastructure has been very regulated. I myself have been a power company board. And I think as we always do, a reduction in regulation would be really helpful because I think we are going to have to use some new thinking and innovation in the overall grid and the new power sources that will become available to us.
One of the themes as well that we have seen is the idea of the money being used for the build. A lot of hyperscalers cash rich, they have taken down the cash flow reserves to finance this. Now we've seen this trend of turning to debt to get the funding. When does it become a worry, or is it going to become a worry for investors?
I think we see this being viewed as an investment. I think the hyperscalers, as well as the semiconductor companies helping finance some of the new large language model companies, finance this buildout, is interesting. It is heartening they want to invest in this. We don't want anyone to become overleveraged, but clearly it is a good sign.
What do people, and this can be folks in adjacent industries or investors or policymakers, continuously misunderstand about the semiconductor industry? Today and including 5, 10, 15 years ago?
I think people understand semiconductors better than ever. When I got into the industry, nobody knew what that was. They thought a conductor on the train. We've come a long way. I think the understanding has gotten better. The importance of semiconductors as a foundation for our entire technology infrastructure. As I always like to say, you can have hardware without software, but you can't have software without hardware. I think I understood much better. And in these acquisitions with semiconductor companies buying software assets or other assets that are adjacent to their specific product line. I think that is a great trend as well.
Final question, 30 seconds, we are still very much at the peak of the semiconductor super cycle. What are you looking forward to to mark the end of that?
You asked me that before, I think it is a hard one. I think it is when we start seeing no constraints on capacity that everyone has what they need, then you know it will be over soon.
Very bullish on the road ahead here for the industry that she has been such a pioneer in, Kris King is author of Breaking Through the Silicon Ceiling. Quite some perspective up from both her as well as our previous guest, the CEO of Ciena, about how the real constraint is the ability to deliver on everything you want.
It is very much a supply story. Looks like that will be the narrative into 2027, which will be very interesting in terms of sifting out the winners and losers.
Our first guest was talking about if you back out of the capex into AI and everything else, the economy has not really done a lot because it is all being driven by this capital expenditure. We are going to put this into context when it comes to the markets. Coming up, a live interview with OpenAI CEO Sam Altman on the latest with his company and the AI landscape in the U.S. and around the globe. This is The Close on Bloomberg.
This is the countdown to the close. The data out of S&P Global highlights the AI build. A show in no signs of a slowdown. The Hyper Five which includes Alphabet, Microsoft, Meta and Oracle have spent more than a trillion on the past five years with more than 5 trillion expected by 2030. Joining us now is Daniel Sandberg at S&P Global. Thank you so much for joining us. My first question is the fact that five companies are responsible for more than half the increase in the S&P 500 capex since 2021. Do you see that as a warning or something that is maybe a strength?
Great to be on with you. As you said, we are in unprecedented waters right here. We are seeing a capex spend we really have never seen before and it's not what is being spent today but what the expectation looks like. We are already at roughly $1.7 trillion of annualized capex across the S&P 500 and the Nasdaq 100. That number represents a 42% year-on-year growth. Really, it is about what's coming next. Visible Alpha consensus sees that rising almost a trillion from here. More than half of that is coming from a concentrated set of companies. Does it represent a concentration risk? It is hard to say. Right now, the big spenders are also the ones with the earnings to back it up. I like some of the prior periods where we have seen capex intensity like this, it does not seem to be much of a disconnect between the earnings coming in and the capex that is being spent, at least not yet.
You say that periods of high intensity have coincided with higher earnings growth, elevated expectations and higher federal funds rates. What is the actual relationship? Is there a correlation or actual causation?
The framework we lay out, we go through prior capex cycles and we classify the periods as build, stretch, overshoot and digest. In the build phase, the capital funding environment is accommodative. Earnings expectations look very attractive. Companies have a lot of incentives to spend the money. In the stretch phase, the financing environment starts to tighten up but the return on investment looks promising. Companies spend despite that more constrained fiscal environment. That is the stretch. Overshoot comes when the earnings expectations start to turn, you get the revisions lower or misses. At that point, we start to see the pullback in capex, and that is when you see digestion of the spending.
Really fantastic to get you to break that down for us, Daniel Sandberg at S&P Global. We will stick with AI and some news from OpenAI which just released its latest model. This new model is OpenAI's latest push towards artificial general intelligence or AGI. OpenAI says paid users will get access to a version of Astra with renewed guardrails after a potential risk. Joining us now is Bloomberg Tech host Ed Ludlow who is standing by with OpenAI CEO.
We are live with OpenAI CEO Sam Altman. The way that OpenAI is framing Astra is basically an early step towards AGI but I thought we can start our conversation with basically what is fundamentally different with Astra relative to prior generations of model.
First of all, thank you for having me. It feels like a new step in this process towards models that can really help us create value, do work, discover new science, help start new companies or create new products. Using it subjectively feels very different than any model before.
What is the principal use case that is different this time around? We can get into the generations of models that were very coding focused. A lot of the development of a cybersecurity focused, but taking it away from the software engineer, what does the everyday person unlock in AI that they were not able to do previously?
One thing that people will immediately notice is if you have an idea and want to work interactively with an AI to get a complex piece -- a whole piece of software built, this is the first model to meet where I can tell someone just give it a try, there's a good chance it will work. And I have watched people make computer games. I have watched people do home sort of DIY electrical engineering projects. I have watched people do very complex simulations for some piece of science they are working on. A lot of the work where you would normally sit down and have to build a financial model here and a PowerPoint presentation around it and figure out how to make a little interactive piece of code to try different simulations. That stuff is so doable now. What I hope will happen is people, I think people will be surprised at the beginning but then as they build up more trust in the model, they are like that can really do this. Just start throwing harder and harder tasks and more creative ideas at it and they will find out what the model can do for them.
The specifics of how Astra is being released is important. A version of it with specific guardrails, what was the thinking behind that and what are those specific guardrails in this early release?
Yeah. The challenge of our industry is we have these models that are getting incredibly capable and incredibly useful, and that people want to use for everything to make their lives a little easier to starting new companies. On the other hand, the risks we have to mitigate also become more serious. Models can do damage if they don't do a good job at that. This model took us a little longer to release them. I think it will be worth the wait but we wanted to spend the time on the safety and security alignment of this model. As people understand the power and impact and capabilities, they will be happy we did. We will have different tiers of access for this model. Today, we are rolling it out to trusted access partners and then in the coming days, if everything goes well, we will roll it out more broadly. You could say this model is -- has a cyber problem, why let anyone use it for cyber? I think it is important to note that the world is very close to a complete change of landscape of cyberattacks and the only way we see for society to collectively defend itself against this coming wave of models from around the world and other companies is to use tools like Astra to rapidly defend against these new kind of cyber threats. So, we will have multiple programs for people at different levels of verification and trust. But we think it is quite important the world use these models to collectively defend ourselves.
What happens prior to release is important. You determine that Astra can find zero-day exploits without human intervention. So, you paused some of the work to strengthen the safeguards. What specifically was it that you saw that made you hit pause? The behavior of the model?
It is worth pointing out that the model we are launching today has been done training for a while. The model we talked about pausing is a future model. With Astra, we did hit cyber critical. That required under our preparedness framework a new set of safeguards because of those capabilities.
I've got a lot of questions for you about Astra and chain of thought. I don't know if that comes as a surprise to you but as Astra and future models become more autonomous, can OpenAI say with confidence that you see and understand what the model is planning before it acts? In other words, you have visibility into the reasoning, that basically users of the model cannot see. Can you go into that a little bit?
We have talked for a long time about the importance of monitorability of models and we also believe in defense. This is only one part of it. Sandboxing most importantly, there is alignment. A number of things come together to make the safety guarantees that we want to be able to make with a model. Monitorability is an important thing. We have been talking for well over a year about the importance of chain of thought monitoring and now we have made various decisions that help us preserve chain of thought, even if it means we don't maximize the capabilities we could get. We think that is important. It is an important component. We want to be clear, we don't think that is the only component that matters.
We are live on Bloomberg Television and radio. We are speaking to Sam Altman, the CEO of OpenAI. There is a lot of bad PR in the world of AI at the moment. Everyday citizens worry about the impact. One way I wanted to put that to you is if you could say unequivocally if Astra or any future generation model behaved in a way that was dangerous, that you would be able to detect it and shut it down?
The obvious hard part of that question is that people have different opinions about what models should and should not be able to do. We think there are things that model should clearly not be able to do. There are some clear redlines. We talk a lot about being able to detect those, but people have strong opinions about things within the broad bounds of what is possible about how to use AI and how they want to be able to use AI. This will be a difficult question for society because there will not be agreement on what is and is not acceptable use of AI. I think it is very important in the same way that people do things with electricity that I may not always like but I think everyone has a right to electricity. We say there will be some rules for society. You cannot electrocute someone else on the street, but this is an important platform that belongs to all of us and we need to be able to use. There will be an ongoing conversation, negotiation in society. There will be some things we agree on that people should not do with AI but there are edges we have to feel our way through. Giving people autonomy and freedom to use AI is important to our mission.
At the G20 this week, Scott Bessent said the AI industry had done a terrible job of explaining how AI benefits ordinary Americans. What is your view of that and whether you on reflection consider OpenAI to be in that bucket?
I think the industry has done a terrible job on the whole. I think we have done a bad job ourselves, maybe better than some others, worse than some others. It is difficult to explain that this is -- the world should want us to worry a lot about this technology. The world should want us to think about all the ways that can go wrong, to talk about what we see coming, but not to sort of stir up doom or concentrate power for ourselves. In the history of technology, the way you solve the hard problems is to look at them, talk clearly about them, have scientists debate about them, mitigate them. That has been our intention, but clearly, sometimes it ends up scaring people. There's another thing that I think that has gone very bad besides talking about the ways this could go wrong, which is I don't think we've done a great job as an industry of really talking about how to empower people with this technology. You hear some people talk about, well, we will give you a cure for cancer and that is great so we will take these risks and you will have to have less power and autonomy. I don't think a cure for cancer is enough -- a wonderful thing to have but we should set our sights higher. We want these tools to enable a boom in creativity, entrepreneurship, people figuring out new ways. I think that has been lost in the messaging and I think there are some companies that have a different opinion there. But it is very important to us that this technology is something that people give more power and autonomy, not less. I think people rightly get afraid when they feel like AI leaders may not want that for them.
That is in part what I wanted to ask you about, pricing on Astra but also generally. The direction of travel has been you cut prices, there is competition with the other frontier labs but how much is that the balance of access to the technology, you need to move quickly in releasing the technology, and also still trying to find fair market value of how you price the technology?
We want to create incredibly capable, incredibly low cost, abundant intelligence. I think Astra is a great step forward. The top and efficiency is so impressive to me. The amount of work you can get done with a relatively small amount of tokens -- I think is just remarkable. Only a few weeks ago we cut our price of Llama by 50%, and our goal is to have the best price performance at every level of intelligence on the way along the curve. We will continue to keep dropping prices dramatically. I'm confident in the shape of the technology and the work we are doing to enable that to happen. Of course, we will make some money along the way, but the opportunity in front of us is so huge, we will be able to continue to bring prices down and still have huge revenue.
Let's end our conversation with -- Astra is a big step toward AI actually doing work autonomously. I have come to see the emphasis on that. If you are staying ahead in the race it requires OpenAI to keep investing, more heavily than maybe expected to keep pushing out the boundaries of what the model is capable of doing, the cadence of release, where do you sit right now on considerations doing that as a private company versus doing that as a public company, where the financial incentive and fiduciary duty for a public company is not necessarily the same as being committed to developing technology, getting the technology released out into the world.
If we go public someday, which I assume we will, I will spend time reminding people who might want to buy or may not want to buy our stock that we are mission first, and extremely long-term oriented. You are going to make decisions for what we think is best for society as a whole for this technology, a multi-decade time frame, and we think we will do great as a company and we will put those things first.
OpenAI CEO Sam Altman on Bloomberg Television. Thank you very much, indeed. Ed Ludlow is going to stick around and break down some of the comments Sam Altman, the CEO of OpenAI, shared with him. I appreciated your asking about what the AI industry at large could do better and could improve upon, especially communicating the benefits of AI. We have seen a sea change in terms of public sentiment towards AI and data centers. What surprised you most about his answer?
The answer to that question was similar to why Astra is different. To this point, the model allows you to pose a question and get a response, rather than the model allowing for an AI to do real work on your behalf. That kind of is the technological change, agency, giving the AI agency to do something that is useful. There is a bit of a fear of that in society. To what degree you are comfortable in leaving an AI to represent you, do your work on your behalf? There's lots of benefits to it. The benefits of that he seems to be agreeing have not been conveyed very well. Like this can improve productivity, it can improve your life, free you up to do things outside of the workload on your desk. It is interesting the sheer acknowledgment, in the context of the Treasury Secretary making the point in the first place, he could not have been clearer he agreed with that position.
The OpenAI President calling this new model a real shift in what kind of work people can delegate to AI. It basically the comments from Sam Altman as well. Do you think that is actually true, or is it marketing speak?
In very simple terms, what Astra is good at is doing work on a computer on a human's behalf. If you can think of the things you do on a computer each day, that you sit there and type, if you have the capabilities, a lot could be done for you. At the other extreme, this is a very cybersecurity focused model where the specific guardrails or this version of it that have been early released relate to that capability. What Sam Altman was going through in that conversation is it is of great benefit in the context of cyber defense, the risk as is always the case in any model that is released and the capabilities on paper is what if that model were to fall into the hands of a malicious or bad actor? It is so capable that it poses a threat on the offense rather than the benefits of its defense in the cybersecurity context.
We know that there was an inadvertent hack carried out by AI models in July. That is top of mind for everyone in the industry right now. What did we learn from Sam Altman when it comes to any kind of timetable for going public? You were delicate in the way you asked it, and it is expected Anthropic will be the first mover here between the models to go public. Maybe September, maybe October. For OpenAI to follow after. Are we thinking later this year, or is it a 2027 kind of thing?
OpenAI has filed confidentially. They said going public is one of several available avenues to them, and they have a number of financial strategies. The reality of it is the public markets require financial discipline. Maybe the capital expenditure story of the last 12 months would be a big counter to that. The training models, it is complicated, expensive on the compute side, the talent side, and it is a difficult process releasing the most powerful models in the world. Those goals are not always in line with the goal of being a profitable public company. What I'm trying to get from Sam Altman is if you go public, what is the priority? Your duty to shareholders, or what it takes to release the model? He towed the line. My expectation was not by the way, here is the latest update, that is not realistic at this time point, but we will wait and see.
Fantastic interview, thank you for coming back and joining us. Ed Ludlow is host of Bloomberg Tech, catch him every weekday at 11:00 a.m. Eastern time. We are less than 10 minutes away from the closing bells on this Thursday. Let's look at where things stand as we get ready to close out the trading day. The major indexes for equities rose and got to their session highs, what we thought was around 11:00 a.m. to 12:00 p.m., the Dow down 0.4%. We're just kind of drifting along these levels right now with the S&P 500 gaining more than 1%. A little bit of a breather in the Treasury selloff. This is really a global selloff we have seen. The 10-year yield at 4.76%. The big jobs report comes out tomorrow at 8:30 a.m. Oil prices have been bouncing around. WTI at $91. And the dollar is weaker. You can look at the big move in dollar-yen. Not normal to see a 2% drop in the dollar versus the Japanese yen. There's a lot of talk about intervention, about whether the Bank of Japan will be raising interest rates any time soon. A BOJ board member did strike a hawkish tone. The Treasury Secretary is encouraging them to raise interest rates as well. From our market analysis, let's welcome Allie McCartney, Managing Director of Wealth Management and Private Wealth Advisor at UBS Alignment Partners. Good to speak with you. Just give us a sense of what the conversation you're having with your clients are like right now on the heels of what has been another blockbuster year for U.S. equities.
We are talking about the third year that has been a blockbuster year. There is both general enthusiasm about everything from earnings to secular growth and technological revolutionary tailwinds, to a lot of concern about the interest rate environment, the conflict in the Middle East, what that is doing to oil. So in a sense, as we have seen over the last couple of weeks, a seesaw between these dueling narratives, inflation and interest rates, will that take interest rates up? If so, will that destabilize the earnings that we are seeing that is really taking the market pretty close to its highs today?
What is the conviction of earnings growth we saw in the second quarter, which was blockbuster by most accounts, can continue to stay at that pace, if not accelerate?
It is a feverish pace. Just like we have seen with AI and technology companies, the top 10 titans, you cannot grow these rates forever, but we do see this continuing. Not only continuing in the U.S., but continuing in Europe but emerging markets in China. We are anticipating a 24% earnings growth this year for the S&P 500. What is the most notable and giving the breadth of those earnings. Not 10 stocks growing at 150% and everything else barely growing or shrinking, the average company is growing about 15% this year on earnings. Based on a lot of the productivity growth and the increased margin ability we are seeing as a result.
And we are seeing a number a little bit less than 14% to 15%. So to me, probably your next question, which side of that seesaw I'm on, certainly reason to be concerned about inflation or rates from every perspective, from our debt, the U.S. debt going up $40 trillion, Scott Bessent coming in on the Treasury. But it is hard to look at the earnings picture and understand we have not even seen most of that expansion in a year and not be bullish.
Very interesting. It seems we are in quite the Goldilocks environment in the U.S. I was interested when you mentioned you are looking globally to Europe, Asia, global markets. What markets are you seeing that U.S. assets don't necessarily offer at the moment?
It is not necessarily what U.S. assets don't offer, but a combination of one thing you just touched on, which is a weaker dollar, which certainly helps when U.S. investors invest in those companies in local currency and repatriate it. What is happening in the U.S. and driving our earnings growth is not just AI, but everything from deglobalization to reshoring, to a wave of capex, and all of those things are happening in those areas as this administration and political alignments and globalization of Congress has started to come apart.
Interestingly, I know you are optimistic about equity markets in particular, but you are also buying gold and short-duration bonds. To me, that sounds like a more defensive play. Is there more to that trade than meets the eye?
Let's start with gold. Gold is something I've had in my clients' portfolio a few years now. That has worked really well. I would say it is partly defensive, partly ballast for your equity portfolio. But it is partly a play on a lot of things we just talked about: a waning dollar, de-dollarization, countries less interested in buying U.S. debt and dollar-denominated assets, and more of a bid for gold. On the short-duration bonds, we continue to have a view that interest rates, in terms of the short rates, the Fed's next move is going to be lower, not higher. We don't think it will be until 2027, but we are not of the school of thought that the next move is higher. Based on that, we continue to expect interest rate volatility because of what we are seeing with the Fed and the inflation market, but we do think that the path is such ultimately buying those bonds now will allow clients to lock in not only that ballast, but a high yield in the next few years.
You mentioned a lot of the risks that percolate that exist, you have to keep in mind with other investors of those risks. Which are you positioning for, and which are you not bothering to position for?
I get a lot of questions from clients about the midterm elections and how you position around the midterm elections, especially geopolitically and politically, with everything going on in this country. That is the hardest one to get around. The way one positions or does not position tactically is diversification. Again, we expect, and historically going into midterm elections, you see volatility up until the election. And in a sense, irrespective of the result, the concept of the uncertainty falling away, you get a rally in equity markets even when there are not so many tailwinds. So I would say that is the one that we are waiting to see what happens and what that means economically and from a regulatory perspective to different stocks and industries. But it is not one that we are actively positioning or acting like a hedge fund in terms of trading around; rather, we are using the benefits of geographical and cross-asset diversification to protect ourselves there.
I appreciate the honesty behind that response. Allie McCartney is director of wealth management and private wealth advisor at UBS Alignment Partners. We are counting you down to the closing bell. Joining us for more market analysis is Justina Lee, a reporter for Bloomberg News—catch her writing on the Bloomberg Terminal—and the senior retail analyst at Bloomberg Intelligence to give us reaction when Lululemon reports earnings, and also to give us a sense of the consumer and how the consumer is doing as we get into the fourth quarter of 2026. We are counting you down to the closing bell, and it does look like we are on track for the first back-to-back gain of U.S. stock indices. And Christine, we have 1% gains for the S&P, the Dow, and the Nasdaq.
Finishing the day with their best performance in about a month. Very interesting, looking at some of the sectors that are leading today. It really tells you where things are quite positive.
Looking at breadth overall, we had close to about 70% of stocks advancing while a little over 30% were declining.
Let's dig more into those sectors. Consumer Discretionary topping the leaderboard as we close for cash equities, up by about 1.6%. A little surprising, because we got lower yields. But perhaps the overall relief from the market proving good for financial stocks.
I want to take a look at the gainers today. We do have HOOD, Robinhood. Tesla, this is the best day since April of 2025; that was during the height of the tariff tantrum. It all has to do with the prediction market growth that Morgan Stanley and some of these other upgrades we have seen at banks recently, especially with football season coming up. That is where you are seeing the strength. I have to point out what happened with Snowflake as well. SNOW is the ticker, having their best day since late May. This is a company that helps organize corporate data. It did have strong revenue growth and then an uptick in the AI assistant. And after the bell, Tesla is going to have its Cybercab, up about 5%. I want to point out, even year-to-date compared to some of those Mag 7, it is still down about 15%.
And there is still a lot of speculation that Tesla and SpaceX will combine as well. I know it was much easier to pick winners today.
I will start with Ciena, the worst performer in the S&P 500, down close to 11%, the worst since June 4. The communications equipment company reported quarterly results that gave a revenue outlook that was above estimates. Unfortunately, Vital Knowledge was talking about how those expectations were too high because the stock is still up about 40%. It had some strong gains relative to its peers. That is why potentially we are seeing some of that drawdown. Victoria's Secret, I know you were talking about the retail side, how you want to keep a close eye on the pulse of the consumer, down 13%. The stock is actually close to 60% up year-to-date. Unfortunately, those quarterly results were underwhelming to investors. Campbell Soup, when you think about their flagship products, as well as in sympathy, that stock is down about 7%, the worst since May 11: weak earnings, weak sales growth, and a dividend cut.
Especially for a name like Campbell's, which relies on that steady yield. Those are the gainers and decliners. Let's take a look at the bond market, because the global bond selloff took a respite today. Yields down across the curve, not by a whole lot. The 30-year yield staying pretty steady at 5.25%. This will likely change tomorrow when the August jobs report comes out and we get a better sense of what is to come. Fed Governor Chris Waller saying he is focused on what happens with that next inflation report due out on September 11. Let's get some earnings that just came out: Oura filing for an IPO. This is the smart ring company. People definitely thought this was happening. Christine has one of their rings on her finger.
Quite a big fan. A lot of analytics. You track your sleep and your activities. Not surprised that this is the next step for Oura, because it has been one of the leaders of that wearables market. A lot of the analytics you get are very popular with consumers such as me.
The symbol would be OURA for an eventual IPO from Oura Health. Lululemon results have come out. Let's start with second-quarter results because that is what ended: net revenue, $2.42 billion, missing the estimate. When you backed out the FX effect, down 12%, looking for a drop of more than 10%, a worse-than-expected read for the Americas. In terms of EPS, $2.92, analysts looking for $2.80. As for the outlook, Lululemon sees third-quarter net revenue of $2.32 billion, a long way off of the consensus estimate. That is the outlook that has been cut for Lululemon. The stock is down almost 8% right now. We will bring in our senior e-commerce analyst over at Bloomberg Intelligence. Poonam, these numbers do not look good, but given that the new CEO has not really started, how much are people going to read into this? Is this just more to fix?
More stuff for her to fix. The numbers are bad from every angle. Everything I see so far, things are worse than expected. It is really setting a floor for her to clean up when she comes in next week and cycles these comps. She will have easy comparisons, but at the same time, the issue with Lululemon is product. Their product and innovation continues to lag, and those are things that don't get fixed immediately. They take time. Even when she steps in, we are talking six to 12 months to see signs of a turnaround.
This is more of a turnaround story for the incoming CEO of Lululemon. Enterprise software company Zscaler has also come out with results. And for the fourth quarter, they reported revenue beating the consensus estimate, adjusted EPS also topping the estimate. Zscaler sees 2027 revenue of at least $3.9 billion, better than anticipated, and adjusted EPS looking for $4.90. The stock is up more than 10% right now. We have results coming out of DocuSign, which gained a lot of popularity during the pandemic: second-quarter revenue of $875 million, topping the consensus estimate; adjusted EPS, $0.16 versus the analyst estimate. Revenue for this quarter, $886 million; the midpoint of that seems to be the consensus estimate. Full-year revenue will be anywhere from $3.5 billion, a slight upgrade in terms of the range it had offered previously. That's good enough for investors pushing it up.
Absolutely. That takes me back to the early pandemic days. That is when the peak of popularity was for DocuSign. Very interesting that it looks like they are still going strong six years after that peak. It goes to show how digitized we have become.
It goes to the idea of how earnings have delivered this past quarter, even as expectations have—
Talks more about how this is a micro-driven market rather than macro-driven.
