To The Moon

The biggest signal in SoftBank’s $7 billion loss isn't the write-down. It's the rising cost of capital across the AI ecosystem. For years, investors assumed AI infrastructure spending was almost unlimited. Cheap financing funded data centers, venture capital backed compute providers, and semiconductor companies enjoyed expanding orders. That assumption is now being challenged. I wouldn't declare an AI capex peak just yet. Computing demand remains real. What may have peaked is the willingness to finance projects regardless of their returns. The next phase will separate companies selling essential infrastructure from those simply renting expensive capacity. Cash-rich hyperscalers could even strengthen their positions as weaker competitors struggle to refinance. My takeaway: Don't blindly sh

The Tell: SoftBank's ¥1.12 Trillion Loss Was No Accident On November 10, $Softbank Corp.(SOBKY)$ reported a quarter the market wasn't positioned for: a net loss of ¥1.12 trillion (~$7.0 billion) for the fiscal second quarter ended September, against analyst expectations of roughly ¥129 billion in profit — and versus a ¥2.5 trillion profit a year ago. The bleeding didn't come from OpenAI — that stake still sits on the books at a fair value of $99.6 billion, unchanged from March, with no impairment taken. The real damage came from a corner few people were watching: power equipment for US AI data centers. Silver Bands 7, a subsidiary held with Vision Fund 2, had pre-purchased generation equipment capacity for projects in Ohio and elsewhere; when pro

Whether the cloud rally can last depends heavily on whether $CoreWeave, Inc.(CRWV)$ CoreWeave and similar AI infrastructure providers can convert massive capital expenditures into rapid cash flow before heavy debt obligations force a reckoning. In this article we seek to perform examining whether hyper-leveraged special purpose GPU financing can translate massive capital expenditures into sustainable cash flow before debt maturities arrive. 1. CoreWeave’s Expansion as the Catalyst for the Cloud Rally The spectacular surge in AI infrastructure valuations over the past 18 months was not sparked solely by Big Tech’s internal capex announcements, but by the rapid operational scale-up of specialized GPU clouds. CoreWeave emerged as the vanguard of this

Institutions Sold S$900 Million — So Why Are They Buying These SGX Stocks? https://youtu.be/5JZfl0OgbD0 $Straits Times Index(STI.SI)$  

Disclaimer: Nothing I say or post should be considered financial advice. Please do your own due diligence before making any investment decisions.‌ So with crude oil projected to stay elevated into 2027, CPI and PPI data releasing this week on Wed and Thur, major banks kicking off the last quarter of earnings season and tech hyperscalers showing lethargy in their hyperzealous runs, should we remain ultra bullish on equities? Or should we start to take profit and risk off the table? Lets see how this develops as we navigate towards 2027. @PawsAndProfits - Specialist in combining FA and TA for Options selling and Swing trading.[666]

I think the bigger force at the long end is the term premium rather than just expectations for another Fed hike. Persistent fiscal deficits and heavy Treasury issuance mean investors need to absorb more duration, while inflation uncertainty makes them demand higher compensation for holding 10Y and 30Y bonds. That can push long yields higher even if the Fed eventually pauses. So I am watching Treasury supply, auction demand and the term premium closely. If those pressures persist, a Fed pause may bring limited relief to long yields, keeping valuation pressure on equities, especially long-duration growth stocks.

While the short end is anchored by the Fed funds target rate path, the long end is driven by debt supply, structural growth/inflation expectations & term premium demand. Long yields rising on hawkish Fed minutes reflects a market pricing in a higher neutral rate environment, persistent long-term borrowing demand, and the heavy fiscal burden of Treasury supply.

I think it’s more than the Fed. Heavy Treasury supply, rising term premium and renewed inflation concerns are pushing the long end higher. The key question is whether 5%+ yields become the new normal.

$NEBIUS(NBIS)$ Insiders Are Selling — But Is It Actually a Red Flag? Nebius dropped more than 5% Wednesday, putting insider selling back in the spotlight. At first glance, it’s easy to think: Insider sells at a huge valuation = warning sign. But the details matter. Nebius COO Ophir Nave sold 77,430 shares on October 5 at an average price around $232. The important detail? The sale was made under a 10b5-1 trading plan adopted back in May, rather than being an on-the-spot discretionary decision. The filing says the shares represented approximately 17% of his granted equity, and he retained more than 877,000 shares afterward.  That’s very different from an executive suddenly dumping most of their position after a disappointing business update. And t

All thanks to Citibank… that sudden pullback is an opportunity to accumulate additional shares! Surely funds managers is working with Citi analyst for this golden move. You sell they buy. So, some fools sell some fools buy!

The indices: all three fell back from record closes, and the Dow lost 0.66 per cent The three indices moved back from the previous day's record closes on Wednesday: the $Dow Jones(.DJI)$ Average fell 0.66 per cent to 51,179.90, the largest decline; the $S&P 500(.SPX)$ fell 0.22 per cent to 7,801.77; and the $NASDAQ(.IXIC)$ Composite fell 0.22 per cent to 27,538.69. The pressure came from the bond market: the 10-year Treasury yield reached 5.36 per cent at its intraday peak and closed at 5.28 per cent, while the 30-year touched 5.73 per cent and closed at 5.66 per cent. Minutes from the Federal Reserve's Septemb

The higher FY2028 target is encouraging, but I do not think US$20B alone is enough to carry MRVL significantly higher. The key question is whether earnings can catch up with the AI narrative. I would watch AI-related revenue growth, margins and whether hyperscaler demand remains strong enough to justify the current expectations. A raised target improves confidence, but once the market prices in strong growth, execution matters more than guidance. If Marvell keeps beating estimates and raising forecasts, the rally can continue. If growth merely meets the new target, valuation could become the bigger constraint. For me: bullish on the business, but increasingly selective on the share price.

Power and compute stocks rise together because artificial intelligence infrastructure links electricity generation directly to semiconductor demand. In this article, we will discuss and share a comprehensive financial and infrastructure analysis of the AI-driven convergence of Utilities and Semiconductors. 1. Introduction: The Convergence of Silicon and Electrons For decades, Wall Street analyzed semiconductors and public utilities as entirely distinct asset classes. Semiconductors represented cyclical, high-growth, innovation-driven plays tied to consumer electronics, enterprise IT, and global trade dynamics. Utilities, conversely, operated as defensive, bond-proxy instruments characterized by regulated monopolies, steady dividend yields, and slow, predictable capital expenditure cycles t

The AI trade is starting to look very different. $Vistra jumped more than 10% after the U.S. Department of Energy announced a conditional loan commitment of up to $4.2 billion for nuclear upgrades. The projects are expected to preserve nearly 4 GW of existing capacity and add another 433 MW.  At the same time, $Nebius rallied on new AI inference demand. These companies operate in completely different industries, but the connection is becoming harder to ignore: AI needs enormous amounts of electricity. The first wave of the AI trade was about GPUs. The next wave could be about everything required to keep those GPUs running. That means nuclear power, grid infrastructure, cooling, data centres, networking and storage could all become part of the same investment chain. What I find interesting

The AI trade may be entering its next phase. For years, the focus was on GPUs, chips and data centres. Now the market is increasingly looking at something much less glamorous — power. $Vistra Energy (VST) jumped 10.77% to $160.50 after the U.S. Energy Department announced a loan of up to $4.2 billion to support efficiency upgrades across six nuclear reactors at four existing plants. At the same time, $Nebius (NBIS) gained 7.44% following an inference-related order. Different businesses, but potentially the same bigger story: More AI compute = more electricity demand. That creates an interesting link between the compute boom and the power market. Data centres can be built faster than new generation capacity, while reliable electricity is becoming increasingly important for running increasin

I think the US$20 billion FY2028 revenue target is a positive signal, especially because Marvell $Marvell Technology(MRVL)$ has raised its target again, driven by AI data centres, custom silicon and interconnect. To me, this shows management is seeing stronger demand ahead rather than just short-term AI hype. However, I would not chase the stock purely because guidance was raised. At around US$287, expectations are already high, so I want to see actual customer orders, revenue growth and margins catching up. Guidance is encouraging, but execution matters more. Personally, I remain bullish on AI semiconductors, but I would rather accumulate on meaningful pullbacks than FOMO after a strong rally. If Marvell keeps raising guidance and delivers the

The indices: the S&P 500 and the Nasdaq both set record closes and the 10-year came back to 5.27 per cent The three indices pushed higher again on Tuesday: the $S&P 500(.SPX)$ rose 0.58 per cent to 7,818.93, closing above 7,800 for the first time and at a record close, having reached 7,844.52 at its intraday peak; the $NASDAQ(.IXIC)$ Composite rose 0.45 per cent to 27,599.89, also a record close; and the $Dow Jones(.DJI)$Average added 0.49 per cent to 51,521.28. The 10-year Treasury yield came back to 5.27 per cent from 5.31 per cent the day before. Last week's pattern, where weak data lifted equities, has

Power is definitely showing signs of embarking on its next leg up. As long as the AI story stay intact, we should definitely scale in whenever it rotates with Compute opportunistically.

A weaker dollar, renewed ETF inflows and improving U.S. regulatory expectations have pushed Bitcoin back toward $80,000. The next test is whether real spot demand can sustain the rally after the initial burst of enthusiasm. Bitcoin briefly traded above $80,000, reaching an intraday high of $81,237—its highest level since mid-May. BTC has gained roughly 28% in August, putting it on track for its strongest monthly performance since November 2024. It has since pulled back toward $78,000, showing that the $80,000–$81,200 zone remains a meaningful resistance area. What is driving the rally? 1. The “debasement trade” is back The U.S. Treasury’s expanded purchases of longer-dated government bonds have helped ease bond yields, but they have also added pressure to the dollar. When investors become

$Bloom Energy Corp(BE)$ Chips determine how fast AI can run, but electricity determines whether a data center can operate at all. A new Bloom Energy position disclosed by the Pelosi household and Aggreko’s proposed IPO are putting the AI power shortage back in focus. As semiconductor stocks pull back, companies that can solve the data-center electricity bottleneck are attracting fresh attention. The latest catalyst is a congressional filing showing that Nancy Pelosi’s spouse, Paul Pelosi, established and increased a position in Bloom Energy (BE). According to the disclosure, the Pelosi household purchased: A total of 15,000 BE shares A total of 200 BE call options A $100 strike price An expiration date of June 17, 2027 The transactio

The Tell: SoftBank's ¥1.12 Trillion Loss Was No Accident On November 10, $Softbank Corp.(SOBKY)$ reported a quarter the market wasn't positioned for: a net loss of ¥1.12 trillion (~$7.0 billion) for the fiscal second quarter ended September, against analyst expectations of roughly ¥129 billion in profit — and versus a ¥2.5 trillion profit a year ago. The bleeding didn't come from OpenAI — that stake still sits on the books at a fair value of $99.6 billion, unchanged from March, with no impairment taken. The real damage came from a corner few people were watching: power equipment for US AI data centers. Silver Bands 7, a subsidiary held with Vision Fund 2, had pre-purchased generation equipment capacity for projects in Ohio and elsewhere; when pro

The biggest signal in SoftBank’s $7 billion loss isn't the write-down. It's the rising cost of capital across the AI ecosystem. For years, investors assumed AI infrastructure spending was almost unlimited. Cheap financing funded data centers, venture capital backed compute providers, and semiconductor companies enjoyed expanding orders. That assumption is now being challenged. I wouldn't declare an AI capex peak just yet. Computing demand remains real. What may have peaked is the willingness to finance projects regardless of their returns. The next phase will separate companies selling essential infrastructure from those simply renting expensive capacity. Cash-rich hyperscalers could even strengthen their positions as weaker competitors struggle to refinance. My takeaway: Don't blindly sh

Whether the cloud rally can last depends heavily on whether $CoreWeave, Inc.(CRWV)$ CoreWeave and similar AI infrastructure providers can convert massive capital expenditures into rapid cash flow before heavy debt obligations force a reckoning. In this article we seek to perform examining whether hyper-leveraged special purpose GPU financing can translate massive capital expenditures into sustainable cash flow before debt maturities arrive. 1. CoreWeave’s Expansion as the Catalyst for the Cloud Rally The spectacular surge in AI infrastructure valuations over the past 18 months was not sparked solely by Big Tech’s internal capex announcements, but by the rapid operational scale-up of specialized GPU clouds. CoreWeave emerged as the vanguard of this

Power and compute stocks rise together because artificial intelligence infrastructure links electricity generation directly to semiconductor demand. In this article, we will discuss and share a comprehensive financial and infrastructure analysis of the AI-driven convergence of Utilities and Semiconductors. 1. Introduction: The Convergence of Silicon and Electrons For decades, Wall Street analyzed semiconductors and public utilities as entirely distinct asset classes. Semiconductors represented cyclical, high-growth, innovation-driven plays tied to consumer electronics, enterprise IT, and global trade dynamics. Utilities, conversely, operated as defensive, bond-proxy instruments characterized by regulated monopolies, steady dividend yields, and slow, predictable capital expenditure cycles t

The indices: all three fell back from record closes, and the Dow lost 0.66 per cent The three indices moved back from the previous day's record closes on Wednesday: the $Dow Jones(.DJI)$ Average fell 0.66 per cent to 51,179.90, the largest decline; the $S&P 500(.SPX)$ fell 0.22 per cent to 7,801.77; and the $NASDAQ(.IXIC)$ Composite fell 0.22 per cent to 27,538.69. The pressure came from the bond market: the 10-year Treasury yield reached 5.36 per cent at its intraday peak and closed at 5.28 per cent, while the 30-year touched 5.73 per cent and closed at 5.66 per cent. Minutes from the Federal Reserve's Septemb

The indices: the S&P 500 and the Nasdaq both set record closes and the 10-year came back to 5.27 per cent The three indices pushed higher again on Tuesday: the $S&P 500(.SPX)$ rose 0.58 per cent to 7,818.93, closing above 7,800 for the first time and at a record close, having reached 7,844.52 at its intraday peak; the $NASDAQ(.IXIC)$ Composite rose 0.45 per cent to 27,599.89, also a record close; and the $Dow Jones(.DJI)$Average added 0.49 per cent to 51,521.28. The 10-year Treasury yield came back to 5.27 per cent from 5.31 per cent the day before. Last week's pattern, where weak data lifted equities, has

$NEBIUS(NBIS)$ Insiders Are Selling — But Is It Actually a Red Flag? Nebius dropped more than 5% Wednesday, putting insider selling back in the spotlight. At first glance, it’s easy to think: Insider sells at a huge valuation = warning sign. But the details matter. Nebius COO Ophir Nave sold 77,430 shares on October 5 at an average price around $232. The important detail? The sale was made under a 10b5-1 trading plan adopted back in May, rather than being an on-the-spot discretionary decision. The filing says the shares represented approximately 17% of his granted equity, and he retained more than 877,000 shares afterward.  That’s very different from an executive suddenly dumping most of their position after a disappointing business update. And t

Disclaimer: Nothing I say or post should be considered financial advice. Please do your own due diligence before making any investment decisions.‌ So with crude oil projected to stay elevated into 2027, CPI and PPI data releasing this week on Wed and Thur, major banks kicking off the last quarter of earnings season and tech hyperscalers showing lethargy in their hyperzealous runs, should we remain ultra bullish on equities? Or should we start to take profit and risk off the table? Lets see how this develops as we navigate towards 2027. @PawsAndProfits - Specialist in combining FA and TA for Options selling and Swing trading.[666]

I think the US$20 billion FY2028 revenue target is a positive signal, especially because Marvell $Marvell Technology(MRVL)$ has raised its target again, driven by AI data centres, custom silicon and interconnect. To me, this shows management is seeing stronger demand ahead rather than just short-term AI hype. However, I would not chase the stock purely because guidance was raised. At around US$287, expectations are already high, so I want to see actual customer orders, revenue growth and margins catching up. Guidance is encouraging, but execution matters more. Personally, I remain bullish on AI semiconductors, but I would rather accumulate on meaningful pullbacks than FOMO after a strong rally. If Marvell keeps raising guidance and delivers the

AI Cloud -Investment Thesis. Is everyone clear about the business models of $CoreWeave, Inc.(CRWV)$ and $NEBIUS(NBIS)$ ? Both specializes in AI cloud-computing and infrastructure providers (often called "neoclouds") focused on renting out high-performance GPU computing power for artificial intelligence training, tuning, and inference workloads. My last post on both was dated 10 Apr 2026. Click here ! for details. It is timely to revisit these stocks since they have just released their latest quarterly earnings. For starters most investors frequently compare them to high growth plays that benefit from surging demand for AI compute res

The AI trade may be entering its next phase. For years, the focus was on GPUs, chips and data centres. Now the market is increasingly looking at something much less glamorous — power. $Vistra Energy (VST) jumped 10.77% to $160.50 after the U.S. Energy Department announced a loan of up to $4.2 billion to support efficiency upgrades across six nuclear reactors at four existing plants. At the same time, $Nebius (NBIS) gained 7.44% following an inference-related order. Different businesses, but potentially the same bigger story: More AI compute = more electricity demand. That creates an interesting link between the compute boom and the power market. Data centres can be built faster than new generation capacity, while reliable electricity is becoming increasingly important for running increasin

Institutions Sold S$900 Million — So Why Are They Buying These SGX Stocks? https://youtu.be/5JZfl0OgbD0 $Straits Times Index(STI.SI)$  

The AI trade is starting to look very different. $Vistra jumped more than 10% after the U.S. Department of Energy announced a conditional loan commitment of up to $4.2 billion for nuclear upgrades. The projects are expected to preserve nearly 4 GW of existing capacity and add another 433 MW.  At the same time, $Nebius rallied on new AI inference demand. These companies operate in completely different industries, but the connection is becoming harder to ignore: AI needs enormous amounts of electricity. The first wave of the AI trade was about GPUs. The next wave could be about everything required to keep those GPUs running. That means nuclear power, grid infrastructure, cooling, data centres, networking and storage could all become part of the same investment chain. What I find interesting

A weaker dollar, renewed ETF inflows and improving U.S. regulatory expectations have pushed Bitcoin back toward $80,000. The next test is whether real spot demand can sustain the rally after the initial burst of enthusiasm. Bitcoin briefly traded above $80,000, reaching an intraday high of $81,237—its highest level since mid-May. BTC has gained roughly 28% in August, putting it on track for its strongest monthly performance since November 2024. It has since pulled back toward $78,000, showing that the $80,000–$81,200 zone remains a meaningful resistance area. What is driving the rally? 1. The “debasement trade” is back The U.S. Treasury’s expanded purchases of longer-dated government bonds have helped ease bond yields, but they have also added pressure to the dollar. When investors become

$Bloom Energy Corp(BE)$ Chips determine how fast AI can run, but electricity determines whether a data center can operate at all. A new Bloom Energy position disclosed by the Pelosi household and Aggreko’s proposed IPO are putting the AI power shortage back in focus. As semiconductor stocks pull back, companies that can solve the data-center electricity bottleneck are attracting fresh attention. The latest catalyst is a congressional filing showing that Nancy Pelosi’s spouse, Paul Pelosi, established and increased a position in Bloom Energy (BE). According to the disclosure, the Pelosi household purchased: A total of 15,000 BE shares A total of 200 BE call options A $100 strike price An expiration date of June 17, 2027 The transactio

$Coherent(COHR)$ fiscal year 2026 fourth-quarter results were strong. Revenue, profit and next-quarter guidance all beat market forecasts. The debate has moved from whether AI optical demand is real to whether Coherent can add capacity on time and how much growth is already priced into the stock. Q4 revenue reached $2.046 billion, up 33.8% year over year and 13.3% quarter over quarter, passing $2 billion for the first time. Non-GAAP gross margin was 40.2%, up 2.15 percentage points from a year ago. Non-GAAP operating profit rose 62% to $446 million, while adjusted earnings per share grew 74% to $1.74. Profit grew much faster than revenue, showing gains in product mix, yields and pricing. Coherent fiscal year 2026 Q4 results Full-ye

NBIS Q2 Earnings Review: Orders Explode, 2027 Visibility Improves, More Room to Run? $NEBIUS(NBIS)$   surged more than 21% after reporting its second-quarter 2026 results. Revenue and adjusted EBITDA both beat expectations, while new contract wins accelerated sharply, customer commitments topped $40 billion, and the company raised its year-end contracted power target to 5 GW. Strong order momentum is extending Nebius' growth visibility further into 2027. With more large contracts set to come online starting in Q4, can a new Revenue Ramp drive another leg of valuation re-rating for NBIS? Key Financial Results – Revenue: Q2 revenue reached $582.3 million, up 454% YoY, above the $572.75 million consensus and about 46% higher sequentially.&n

The higher FY2028 target is encouraging, but I do not think US$20B alone is enough to carry MRVL significantly higher. The key question is whether earnings can catch up with the AI narrative. I would watch AI-related revenue growth, margins and whether hyperscaler demand remains strong enough to justify the current expectations. A raised target improves confidence, but once the market prices in strong growth, execution matters more than guidance. If Marvell keeps beating estimates and raising forecasts, the rally can continue. If growth merely meets the new target, valuation could become the bigger constraint. For me: bullish on the business, but increasingly selective on the share price.

$Cerebras Systems(CBRS)$ nearly quadrupled its AI-cloud business in the second quarter and raised its annual forecast. Its shares nevertheless fell sharply after hours, showing that investors are looking beyond revenue growth to the cost of supplying computing capacity and the speed at which a large backlog converts into recognised sales. Cerebras reported after the August 12 close for the quarter ended June 30. GAAP revenue increased 74% year over year to $180.1 million, while its core measure of revenue increased 103% to $209.9 million. Core cloud revenue rose 287% to $127.7 million, but hardware revenue declined to $54.1 million from $70.3 million. The company raised expected 2026 core revenue to $880–$890 million from $855–$865 million. Cerebr

I think the bigger force at the long end is the term premium rather than just expectations for another Fed hike. Persistent fiscal deficits and heavy Treasury issuance mean investors need to absorb more duration, while inflation uncertainty makes them demand higher compensation for holding 10Y and 30Y bonds. That can push long yields higher even if the Fed eventually pauses. So I am watching Treasury supply, auction demand and the term premium closely. If those pressures persist, a Fed pause may bring limited relief to long yields, keeping valuation pressure on equities, especially long-duration growth stocks.