Pinned Tweet
Stock is up a shit ton...failed to participate...this stock sucks and everyone in it is dumb as fuck.
1
4
8,404
$DELL call from (Contra) Barron’s
barrons.com/articles/dell-te…
Our stock pick of DELL has performed exceptionally well since we initially picked it, though we feel we’ve overstayed our welcome.
So let’s face the issues. Huang’s ouster could threaten Dell’s server business. Its servers are on Nvidia’s chip platform, meaning more Nvidia AI chip deliveries translate into more deliveries for Dell servers. But if the Huang effect vanishes now that he is no longer CEO, Nvidia could lose market share. And if more hyperscalers make their own chips, they could also make their own servers, which would pressure sales estimates for Dell lower.
For broader data spending, the Democrats’ sweep of the congressional elections might bring about delayed data-center investments. That would reduce server demand in the near-term, and spread it across more years in the future, introducing an added layer of risk to Dell’s cash flows.
Either way, some hyperscalers could announce lower capital expenditures than they’ve currently guided for, to quell the market’s anxieties about returns on business investments. Most capex is going into data centers. That would hurt data-center server demand. Oracle might have dropped a breadcrumb indicating that day is coming: The software giant’s 2029 free cash flow guidance from its investor day when in October implied a year over year drop in capital expenditures. Yes, the market knows this, but Oracle might have been the first of a string of disappointing capex commentary.
$AMZN CEO Jassy to CNBC:
- long partners with NVDA, “I don’t see that changing”, re: the relationship
- on competitor side, “Trainium is doing very well, we have the co-opetition, room for all of us to be successful”
- “we don’t think of it that way”, if Amazon can be successful with wounded Nvidia…“never wish ill on a competitor, I think of us more as a partner”
- will there be a time you no longer need Nvidia chips? “I don’t foresee a day we are not a partner. Always will have customers that want Nvidia chips. But, the vast majority of inference on Bedrock is our own silicon”
- Where are we in the AI cycle? “Incredibly early innings in my opinion. Demand is barbelled to date, AI labs on one side, other end is enterprise with things like customer service. In the middle is vast majority of enterprise workloads. I understand why people are nervous, because the numbers are so big. If you have the demand signals like we do, the big chunks of compute are reserved in 2027, in 2028, and even a fair bit in 2029, if you have those demand signals, when you are growing this quickly and deploying this capital and then later the FCF and ROI is very compelling. We will be a much larger FCF in a few years”
- On the 2 yr guide: “I think in this case, we do have strong visibility into the demand. We are also aware of the unusual capex investment. The investment is working and I think it benefits shareholders to let them know”
- “we expect to spend more on capex in 2027, that’s what I will say”
- On $ANT: “The pace at which Anthropic has grown its business, have not seen anything like it. An impressive team, smart they focused early on in coding. Amazing partnership for us, they do lots of their inference on our chips. We have learned a lot, and we hope we helped them too, it’s early in our partnership.
- Converting the ANT converts? “We are believers in Anthropic. We invested not to make a quick buck. We anticipate being holders for a long period of time. We have a lot of ways to finance.”
- On funding for build-out, capital crunch: “It will have its own evolution, look at 6-9 mths. Will be a number of companies looking to finance. I think businesses are all different. The more AI you do, the more it drives non-AI, what we call core.”
WSJ
$NVDA
“Since my departure, I’ve seen an outpouring of calls from Nvidia’s employees, investors, business partners, and customers for me to return as CEO,” Huang said in a short statement. “If the Board of Directors decides that my return is in the best interest of our shareholders, our employees, the company, and America, I would be honored to answer these calls and return as CEO.”
The statement was Huang’s first since he resigned Thursday following news that he had extramarital affairs, including one with a suspected Chinese spy and another with an employee. The revelations prompted the board to announce early Tuesday that it had formed a new committee to oversee Nvidia’s national security policies and compliance with export controls.
$NVDA
FT:
Jensen Huang’s daughter was closely involved in orchestrating the public defence of Nvidia’s relationship with the Singapore-based company at the heart of the scandal that led to her father’s resignation as CEO.
Madison Huang, a senior director of product marketing at Nvidia, assumed a broader role outside her formal remit, helping to direct the company’s response to increasing press scrutiny of Megaspeed, the Singaporean cloud provider, according to three people familiar with the matter.
As the $3.8tn chip giant’s relationship with the little-known company first started to draw attention late last year, Nico Caprez, a fellow executive at Nvidia whom Madison Huang recently married, was promoted to a role that covered international deals with cloud computing clients in Asia, the people said.
The Nvidia CEO’s decision to install family members in senior roles at what was until recently the world’s most valuable public company had caused some concerns over governance even before Huang resigned last week, according to current and former employees.
3
2,052
$MSFT Nadella appearance on Pragmatic Engineer podcast:
“Own the intelligence”: avoid lock-in & treat agents as an insider risk. Satya cautions companies to avoid lock-in with any single model provider by ensuring that the harness, memory, and model are not coupled, so that they can be changed with ease. He also thinks more companies should consider using open models as a way to protect their core intellectual property (IP) from leaking out, and build private evals to decide what models to use and upgrade to. He also says that models are intelligent but unreliable, so it’s best to treat them as “insider risks” and build a separate system to verify that agentic systems work as expected.
Microsoft is betting on “hybrid intelligence.” Satya questions why we rely on the cloud for AI infrastructure when last year’s frontier-equivalent model can already run on a laptop, while the frontier-equivalent model from as recently as the previous quarter can run on desktops. He predicts the concept of “hybrid intelligence” will gain traction, of having local AI models running on laptops and desktops, and a model router deciding when to use cheap, “unmetered” local models, and when to route to a more capable, expensive one in the cloud. Microsoft is building infra to support such routing with its Microsoft AI (MAI) models and Surface Laptop Ultra lineups.
Code generation has become cheap, but verification is still hard. Satya reveals he’s been thinking a lot lately about how challenging it is to verify the correctness of code generated today. Models doing a “critique” on generated code is one solution, but it involves model alignment issues, such as a model achieving its goal in undesirable ways; for example, making tests pass by deleting failing tests. Satya also thinks we’ll start to build more verification loops based on production traces, in a kind of “recursive self-improvement” (RSI). But for this, models need to be able to verify fixes, meaning we return to the verification problem! We didn’t arrive at any definite solutions for this during the conversation, which tells me this is a problem space that will keep many of us engineers busy!
The cause of the junior hiring slowdown is that companies are figuring out the new way to work. I asked if Microsoft is hiring new grads and juniors, and Satya admits there's a slowdown there (and at many places). He identifies a process that's currently taking place as the reason why:
There's an internal reorg happening on team sizes and roles Teams are figuring out how to build "software factories" and how to balance human cost/effort and token budgets
When all this is figured out, then junior recruitment will pick back up
The next big engineering challenge: building “software factories” within teams and companies. Satya believes every company needs to control their AI infra – the harness, context, memory, model – and that verification of the generated code is a hard, currently unsolved, problem. Inside Microsoft, he sees lots of teams building their own “software factories” to improve team and org operations. As we talked, it became evident that building software factories looks set to be a widespread engineering problem across the industry because it’s a novel challenge without many existing “blueprints”. It will be slightly different for every team and company.
Satya loves how decision models like ‘Jev’ and Microsoft’s ‘Decision-1’ are a kind of new primitive. He says: “These decision models are kind of like Excel macros. What happened to spreadsheets with macros can happen with code with intelligence embedded as a primitive. This primitive has a distribution, it doesn’t need to be generated again and again, it’s token-efficient. I think that this is one of those places where software engineering remains very much alive. Decision models are like the stored procedure of the intelligence era.”
$NVDA
WSJ:
Just last month, the CEOs of emerging cloud companies that buy Nvidia chips and rent out computing power gathered in Berlin for Nvidia’s premier European artificial-intelligence conference. In a talk that surprised some, Huang’s message was clear: Nvidia’s days as an open checkbook were over.
Huang told his partners that if their companies wanted allocations of Nvidia’s prized GPUs, they would need to secure upfront payments from customers, people in attendance said. That would make their projects easier to finance and ensure that Nvidia wasn’t alone in being a source of financial support.
“You’re not charging what you should be charging,” he said, telling them that Elon Musk was renting Nvidia chips to customers at better rates than they were. “I don’t even need you guys. I could just give all this capacity to Elon! Elon wants it!” he said, according to two executives who heard the remarks and one who was briefed on them shortly after.
UBS analyst Tim Arcuri estimated in a recent note that an Nvidia financing pullback could take $50 billion or more out of the market, starving companies of support as interest rates rise.
Nvidia recently made an offer to Reflection AI, an open-source model developer valued at $25 billion, to license its technology and hire away some of its talent, according to people familiar with the matter. Huang had publicly praised the company. An agreement hadn’t been reached before he left, the people said.
Nvidia has also been finalizing a massive investment in Thinking Machines Lab, another provider of open-source models, valuing the company at $40 billion, according to people familiar with the matter. Thinking Machines is confident the deal will be completed, according to a person familiar with the matter.
Gerstner's message to the world:
Altimeter senior employees can sleep with subordinate(s) without risk of being fired...it's just sex.
Replying to @CGasparino @nvidia
Charlie - as you said “maybe all noise” or “maybe not.” Isn’t that the whole point? The Board should take the time to gather all the information required for an informed decision. This is such basic common sense & good Board practice that I find the objections odd. 🤷♂️
$NVDA Jim Cramer on Collette Kress
"One time, for example, Nvidia missed the quarter because of an issue involving bitcoin mining. Nvidia’s GPUs were being used by cryptocurrency miners in record numbers until a sudden plummet in price led to a huge drop in mining and therefore sales of its chips. Kress explained it to me and admitted that she had gotten the numbers wrong because she didn’t have insight into that mysterious crypto channel, so she didn’t know about an inventory build caused by a collapse in prices. She apologized. She didn’t need to do so. An explanation would have been enough, but in her patient way, she had to let me know that she would do better next time. And she did. She never missed again. That element of humility is very rare in these kinds of calls. Because of it, and her powerful bearing, I told people to stick with Nvidia when so many others were running from it. There were several instances where I questioned the valuation. She would patiently work me through why a certain valuation might be considered appropriate given the future. It always made sense. Again, it’s why I reiterated buy to you over and over and over.
I had been suggesting to Kress that Nvidia adopt a much more aggressive buyback, and I was quite strident. She took it to Huang and came back to me with some of the company’s parameters that differed with mine. I could tell Huang was far more lukewarm about it, even as I kept saying that the best CFO ever, the now-retired Luca Maestri of Apple, shrank the iPhone maker’s share count by roughly 40% during his decade-plus tenure. I dropped it as a cause because I did not want to alienate Kress. And I was also critical of Huang through Kress for not taking some profits on some of the big wins he had on his investments — in particular, Intel.
Kress is now going to run the company until a full-time CEO is picked. This thoughtful, razor-sharp person will do a terrific job because she listens, she learns, she explains, and she is not afraid to speak her mind. The coda? We know Huang reportedly derided Kress as a deer in the headlights. What we don’t know is what she was saying to him that he called her out with such vicious opprobrium. Was it the billions he wanted to give companies that perhaps didn’t deserve it? Was it the commitment to invest $100 billion in OpenAI in September 2025 that sewed up nothing because OpenAI announced a huge deal with Nvidia’s chief rival, Advanced Micro Devices , a couple of weeks later? Was it to question an initiative being led by Huang’s son-in-law ? Somehow, in retrospect, I think if we knew the context, you would know why I think the world of Colette Kress because I can assure you that she had the institution’s interests first. Sadly, I no longer can say that about the man himself. Kress has the backing of the board as the right person for this job."
$NVDA
JENSEN HUANG IS DISCUSSING A COMEBACK AS NVIDIA CEO WITH ALLIES, WITH A “WAR ROOM” OF SILICON VALLEY EXECUTIVES BEING PUT TOGETHER - FT
Jensen Huang is considering a comeback as Nvidia’s chief executive, as allies in Silicon Valley pressure the company’s board to reinstate him days after the tech executive resigned over reports he had affairs with a suspected Chinese spy and a subordinate.
Huang has discussed with confidantes the possibility of returning, according to multiple people familiar with the matter — even as an independent probe by Nvidia’s board investigates the conduct that led to a spectacular leadership crisis at the $4tn chip company this week.
One Huang ally told the FT they were putting together a “war room” of Silicon Valley executives to lobby for the former Nvidia chief’s return and would try to garner support from the White House.
“Everyone that matters in tech outside the frontier labs loves Jensen,” they said. The efforts had Huang’s support, they added.
The effort to reinstate the CEO is likely to face pushback from key policymakers in Washington, with Treasury secretary Scott Bessent on Sunday calling for a “dramatic shake-up” in leadership at the company after Huang’s departure.
"Even allies were unsure how to rehabilitate Huang’s reputation. “Having an affair with a subordinate is a fireable offence,” acknowledged one person involved in efforts to reinstate Huang."
$NVDA
BBG: Nvidia appealed to Taiwanese officials after prosecutors on the island accused an employee of helping to smuggle AI chips into China, raising fresh questions about how the firm views its own role in stopping the illicit trade.
The company expressed concerns to political officials about the July arrest of an Nvidia manager surnamed Chang, who was indicted in August on charges of forgery and breach of trust, according to people familiar with the matter. Taiwanese prosecutors described Chang as the “core figure” of a network that allegedly smuggled 74 servers containing Nvidia chips to China before attempting to divert another 56 that were seized by Taiwan’s Coast Guard in May.
Nvidia’s representatives told Taiwanese officials they disagreed with the charges against Chang, who is one of nine defendants in the case, and believed he should not be under investigation, said the people, who asked not to be identified discussing a sensitive matter.
In a statement to Bloomberg, Nvidia said it “investigated this matter thoroughly, as is our standard practice, and found no indication that our employee knew of the alleged diversion effort by employees at other companies.”
The political officials told Nvidia that they had no role in the legal proceedings, which are being handled by the Taiwan Keelung District Prosecutors Office, according to the people.
$NVDA
NYT:
Over the next few days, Mr. Huang received texts and calls of encouragement from friends, such as Elon Musk of Tesla and SpaceX, OpenAI’s Sam Altman and the financier Larry Fink. President Trump also called him on Tuesday with a word of advice: Fight.
Some allies began to make plans to turn the tide. Mr. Huang told many of them that the woman wasn’t a spy. She was just a business customer.
But four days later, The Times contacted the company and Mr. Huang about another affair. This one was with a woman who directly reported to him, Kimberly Powell, a 48-year-old who led Nvidia’s health division. The two had been caught on camera in an affectionate moment during Nvidia’s annual conference in San Jose.
The company declined to comment. Mr. Huang, who was reached by phone, said, “I don’t think you should be talking to me directly.” Then, he hung up.
After the news broke, the 10-member board called an emergency meeting. Members called in from remote locations and deliberated over a decision none of them ever thought they would have to make.
The revelation of another infidelity showed that Mr. Huang, who had extolled the company’s code of conduct rules over the years, had been violating them. First, there was the relationship with a customer. Then, there was the relationship with an employee.
But in the world of corporate governance, the worst thing a chief executive can do is lie, especially in a crisis. It is a reason they get fired.
$NVDA Damning article by WSJ
wsj.com/tech/jensen-huang-ba…
To build the most advanced AI models and chase the U.S., Chinese developers say they still need Nvidia’s top-of-the-line chips. When using the chips in overseas data centers, they sometimes carry physical hard drives with the training data back to China from Southeast Asia. The data centers can also serve customers outside of China who use Chinese-developed AI.
Alice Huang and her associates leveraged their connections with Jensen to get named as an Nvidia cloud partner, according to people familiar with the matter.
The Nvidia partnership program was in effect a list of Jensen favorites who got preferred access to scarce AI chips—often by proving they had a big customer lying in wait to use the computing power, say people in the industry. In the case of Megaspeed, that customer was Alibaba, they said.
Data center companies in Southeast Asia found that once they landed Nvidia chips, Chinese demand was insatiable.
Aolani Cloud, a Singapore cloud-computing provider, courted investors last year by saying the geopolitical situation provided a unique opportunity for companies that could connect Chinese customers with advanced chips. Aolani, a Nvidia partner, explained in an investor presentation viewed by the Journal that top Chinese internet companies had to rent the chips they were blocked from buying.
The pitch worked. Aolani’s investors helped it fund some $2.5 billion in hardware in Malaysia by the beginning of this year. Its anchor customer: ByteDance, the computing power-hungry TikTok parent that hopes its AI models will catch up to OpenAI and Anthropic within several years.
Aolani says it has systems in place to strictly adhere to U.S. regulations, and from the start has been focused on building a global customer base.
DayOne, a Singapore-headquartered company spun off from a Chinese developer and operator of data centers, is building data centers able to house Nvidia chips across Malaysia and Indonesia. A single technology company accounts for nearly 70% of DayOne’s business, according to a recent DayOne filing, and people familiar with the matter said that customer was China-based ByteDance.
The blanketing of Malaysia with data farms to feed China continues at Senawang, an hour’s drive from the capital of Kuala Lumpur. Amid palm-oil plantations, trucks this past week rumbled in and out of a data center under construction by Vena Group, which is backed by American asset manager BlackRock.
Around midday, workers crossed the highway to the rows of shops and restaurants, where they could buy a curry lunch for about $1.30. A grocery-store owner estimated that 80% of the workers hail from Bangladesh and the rest from Indonesia.
Like many data-center businesses, Vena Group will construct the facilities and lease space to tech companies that install their own hardware. One company planning to use Nvidia-powered servers inside, say people familiar with the matter, is Chinese tech giant Alibaba.
As smuggling cases proliferated and remote-computing arrangements drew closer scrutiny, Nvidia already began last year to limit China ties. It has required many of its cloud partners outside China to keep their Chinese customers below a certain level, which has shifted between 30% and 49%, though the rule wasn’t written into any contracts, people familiar with the matter said. Nvidia also urged some cloud partners to weed out Chinese nationals from top management and push out Chinese investors, the people said.
Cloud providers are setting their sights on regions other than Southeast Asia, such as Japan, Brazil, Central Asia and Europe, to build out new computing capacity that can serve Chinese customers.
The most amusing part of the $NVDA bears prancing and dancing...Many have been bearish since 2023...when the stock was a split adjusted $30-40, rising from $11. Back then it was "this stock has already priced in years of growth and this party is close to over"
Here we are, even after the stock has dumped mightily due to Jensen's departure...and it's still a 4-5x gainer since those days 😂
AkhenOsiris retweeted
BoA: NAND downside.
NAND appears to be a bigger concern. Three reasons: (1) spot-price corrections (down 20%+/- since mid- to late-Oct), (2) weaker orders among PC/smartphone OEMs and even China hyperscalers, and (3) potentially delaying 1Q27 contract price negotiation (no rush order). $SNDK
AkhenOsiris retweeted
Claude extended its lead with enterprise buyers in November, and the demand behind that lead is starting to surface as friction around cost and capacity.
I find the discourse around the US/China race, rivalry to be somewhat amusing. Every view of the two from an AI lens makes you wonder how the countries have a relationship at all (yes, yes, I know, the world would be in a depression if we suddenly stopped trade with China). The need to stifle China and its military is of paramount importance, but we gotta keep doing a HALF A TRILLY in business in the meantime 😂
I thought @dwarkesh_sp made a good point. That the availability of iPhones and Teslas in China didn’t halt the rise of superior indigenous competitors.
But I’m open minded to the possibility that chips are different than consumer products in this regard.
2
857
$ANT $MSFT $GOOGL
ETR:
Claude leads enterprise AI tools on satisfaction, spending plans, and coding, according to ETR’s November 2026 AI Tools Pulse. Microsoft 365 Copilot still has the widest broad use, at 68% of companies, but Claude is gaining fastest: broad use rose from 38% in June to 62% in November. Six months into the series, enterprise buyers keep choosing quality over price. The November 2026 edition surveyed 200 technology leaders on how they use, rate, and pay for four of the most widely deployed AI tools in the enterprise: Anthropic Claude, Google Gemini, Microsoft 365 Copilot, and OpenAI ChatGPT.
Key Takeaways
- Claude’s footprint keeps widening. Broad use has climbed steadily since June, and a sizable pilot base points to further conversions ahead.
- Satisfaction is translating into budget. Claude remains the highest-rated tool, and its buyers are the most likely to plan bigger spending over the next year.
- Coding is the stronghold. Claude Code is the most widely deployed coding tool, holds the largest share of coding budgets, and retains nearly all of its users.
- Demand is testing capacity. Claude users run into usage limits more often than users of any other tool, and buyers are responding in different ways.
research.etr.ai/etr-data-dro…










