Saturday Night Quarterback Part 2 says (on a Sunday), for the week ahead expect…
A week of Big Tech earnings reports will confirm or reverse what sure looks like a sector rotation away from tech shares and into metal and mining company stocks.
A week of Big Tech earnings reports will confirm or reverse what sure looks like a sector rotation away from tech shares and into metal and mining company stocks.
Will OpenAI go broke this year? 2027% Never?–with the company navigating its way to a company-saving initial public offering in 2026 or 2027?
The company’s current attempt to raise an additional $50 billion in venture capital from Middle Eastern investors will certainly help answer the question.
And the answer isn’t of interest only to those private investors who have put $57.9–$64 billion into the company across 11 funding rounds. OpenAI owes so much money to the companies supplying its chips and building its data centers–an estimated $1.4 trillion in contingent liabilities– that an OpenAI failure would dent–or worse–the entire technology sector.
Taiwan Semiconductor Manufacturing (TSM) has told Nvidia (NVDA) and Broadcom (AVGO) that its advanced capacity is effectively fully booked through 2026 and that it cannot meet all of their requested AI chip volumes.
I expect more volatility as forth quarter earnings season picks up speed. Next week, despite the short week created by Monday’s Martin Luther King holiday, 157 companies are scheduled to report earnings with highlights that include Netflix (NFLX) on Tuesday; and General Electric (GE),Procter & Gamble (PG), and Intel (INTC) on Thursday.
I’d argue that the bIg news out of the 2026 Consumer Electronics Show in Las Vegas so far isn’t about new chips, or flashy hardware across TVs, PCs, phones, and robots.
It’s all about speed: Nvidia (NVDA) pushed up the launch schedule for its new Vera Rubin AI computing platform by several months from late 2026 to the middle of 2026. Vera Rubin promises about 10x higher throughput and 10x lower token cost than the prior Grace Blackwell platform. The chip now looks be available to customers in the second half of 2026. Cloud partners like AWS, Google Cloud, Microsoft, and CoreWeave planning Rubin-based instances starting in that second half 2026 window.
I take everything from both extreme speculative bulls and extreme negative bears with a grain–or more–of salt. Members of each camp talk their own positions and certainly aren’t above cherry-picking facts to buttress their own views.
But I do take shorts like Jim Chanos very seriously. In his long career on the short side of the market Chanos has proven himself to be a well-informed analyst of company weaknesses and corner-cutting. So I think it’s very important to listen when he has joined his voice to those pointing to serious accounting problems at AI companies and the possibility that those accounting problems could lead some of the huge amount of debt that AI companies have issued this year to blow up. Chanos and other short sellers like Michael Burry have focused on the use of asset backed debt to pay for the huge build out in AI data centers. The problem, these shorts say, is that many of these new AI companies–particularly in the part of AI called “neoclouds–have used assets in the form of Nvidia (NVDA) chips to secure large loans to buy more AI chips to scale up their operations. But, and this is the crux of the shortsellers’ case, they are using unrealistically long schedules to depreciate the value of these assets–5 or 6 years when the life of these technology assets is more like 2 to 3 years.
I’d temper my exuberance a bit. Some of the big bounce today is a result of extremely negative sentiment on AI stocks in the last few weeks and some highly publicized selling by a few big name (Peter Thiel comes to mind). The big test will come in the next few days when wecsee if Nvidia’s earnings can ignite an end of the year rally across the broader market.
Nvidia’s (NVDA) revenue rose 62% to $57 billion in the fiscal third quarter, which ended October 26. Earnings were $1.30 a share. Analysts had predicted sales of $55.2 billion and earnings of $1.26 a share.
Nvidia (NVDA) will report earnings Wednesday November 19, for the quarter that ended on October 25. According to Zacks Investment Research, based on 13 analysts’ forecasts, the consensus EPS forecast for the quarter is $1.17. The reported EPS for the same quarter last year was $0.78. Visible Alpha reports slightly different estimates. Nvidia is expected to report adjusted earnings per share of $1.26 on revenue of $55.28 billion, each up more than 55% from the same time a year ago. Data center revenue, the chips Nvidia sells that other companies buy to train and run a variety of AI models, is expected to grow 61% and make up $49.53 billion of Nvidia’s revenue. I expect the earnings report to move the market. How could it not? It’s almost as if the financial market’s stage director had cleared the boards for Nvidia’s report.
Nvidia’s basic problem is that investor expectations are so high that the company struggles to meet them. In the quarter ended on July 27, results released today after the close of trading, Nvidia said it earned an adjusted $1.05 per share on $46.74 billion in revenue. Which disappointed the market. The stock fell 3.03% in after-hours trading.
what would a week be without an inflation report. And the week will end on Friday with the Personal Consumption Expenditures index report for July. The Fed’s favored inflation gauge, the core PCE was up 2.8% year-over-year in June and 2.7% in May. Another tick higher wouldn’t be good news for investors counting on the Federal Reserve to cut interest rates at its September17 meeting. But THE event of the week will be the earnings report from Nvidia (NVDA) after the close on Wednesday, August 27. Nvidia is expected to report earnings for the fiscal Quarter ending July 2025 of $0.94 a share. The reported earnings per share for the same quarter last year was $0.65. Nvidia faces a challenge to beat lofty expectations. Wall Street analysts expect Nvidia to deliver 53% year-over-year revenue growth in the quarter and expect a 48% increase in earnings.
In the long term, say two to three years, I think Huawei and its AI chips–plus the rest of China’s AI sector–pose a significant threat to Nvidia’s (NVDA) position as king of the AI mountain. In the short run, however, the negative aspects of that threat are far outweighed by the benefits to Nvidia of the Trump Administration’s decision to allow the company to resume selling its H20 chip in China.
The reason for that is pretty simple: in the short term whatever the technology capabilities of individual Huawei and other Chinese chips, in the aggregate China’s AI chip makers don’t have the capacity to manufacture AI chips in the volumes the global AI boom requires.