$AMZN $GOOGL $MSFT all reported this week.
$AMZN: 2026 held at $220B, and Jassy says "meaningfully more" in 2027.
$GOOGL: 2026 held at $195-205B, with 2027 set to "increase significantly."
$MSFT: 2026 trimmed to $170B (from ~$175B, as some third-party capacity slipped into next year), with capex set to "accelerate" in the first half of 2027
The 2027 numbers get set on the Q4 calls in Jan/Feb.
Show more
$GOOGL AI mode and Agents will ultimately make Google Search more valuable to advertisers, not less.
AI Mode searches run about 3x longer than classic searches, and one in five uses voice, images or documents. A longer query tells Google far more about what the user wants, which is what ad targeting runs on. "Running shoes" is a guess. "Cushioned trail shoes for flat feet under $150 for a half marathon in March" is close to a purchase order. This moves ads closer to the sale, which is what advertisers actually want to pay for.
Today, advertisers often pay for clicks and hope those clicks convert. Agents could take users further down the funnel: comparing products, filling out forms, booking tables and eventually completing purchases. That creates an opportunity for Google to monetize not just discovery, but the transaction itself. Google has many of the pieces needed to make this work: Search, Chrome, Android, payments, identity and Workspace. The bigger opportunity is moving from monetizing attention to monetizing intent and outcomes.
Google is already positioning for this. Ads in AI Mode are launching internationally, multi-turn ads that follow a conversation are launching in the U.S., and AI Max matches ads to long conversational queries advertisers never put in a keyword list. Inside AI Mode, sponsored product listings already get about the same click-through rate as organic ones. Users aren't skipping the ads.
The skeptic points to paid clicks slowing from 13% to 8% while queries hit a five-year high. That gap is a choice. Google is shipping the experience first and monetization second.
Show more
RT @lipko: The 10-Q is out and we can see the breakdown between Volume and Price. It’s also interesting to see the massive decline in Netwo…
A lot of people are misunderstanding $GOOGL search slowing.
People are using Google Search more than ever, but Google is earning less on each search.
Query growth, the number of searches, is the fastest it's been in five years. Search revenue grew 12%, slower than in Q2. If searches are speeding up while revenue slows, revenue per search has to be falling.
Three of the reasons it fell are temporary:
•FX: a stronger dollar shrinks foreign revenue when it's converted back.
•EU de minimis change: the EU ended its duty-free exemption for cheap parcels, so Asian retailers advertising to Europe cut spend.
•World Cup timing: sports-related ad spend landed in June (Q2) instead of July (Q3).
The fourth reason is structural. A growing share of searches now happen in AI Mode, Google's chatbot-style search. AI Mode shows fewer ads than the classic results page, so those searches earn less. Google is adding ads there now, but it hasn't said how big the gap is.
AI is bringing more usage, but a growing share of that usage pays less per search, at least for now. If Google closes the AI Mode monetization gap, the higher query growth turns into revenue growth. If it can't, Search grows more slowly even while usage rises.
Show more
My key takeaways from the $GOOGL (Alphabet) Q3 2026 earnings call.
1. Search slowed while query growth hit a five-year high. Search and other ads grew 12% to $63.3B, down from Q2. The deceleration has four parts. FX was the biggest drag. The EU ended its de minimis exemption (duty-free treatment for cheap parcels), which hit APAC retail advertisers. World Cup media budgets landed in June instead of July. And AI Mode queries grew fast on a surface with "lower monetization" than classic Search. The first three wash out over time. The fourth is structural: more usage on a surface that earns less per query.
2. The EPS miss is mostly a SpaceX mark. Revenue grew 25% to $127.6B. Operating income grew 30% to $40.6B, a 31.8% margin. EPS came in at $2.12 against $3.02 expected because OI&E (other income and expense, the non-operating line where investment gains and losses land) swung to negative $8.2B. Most of that was an unrealized loss on the SpaceX stake after its share price fell. My rough math: after tax, that mark costs around $0.55 a share, so it explains more than half the $0.90 gap but not all of it.
3. Q4 Search is already guided softer. Lapping last year's Search acceleration cost about 3 points of growth in Q3 and will cost a couple more in Q4. FX gets slightly worse at current spot rates, mostly in Search and YouTube. The rest of Services won't make up the difference: YouTube ads grew only 9% to $11.2B, and Network grew 1% to $7.5B.
4. The AI Mode ad fix is rolling out but not proven. Ads in AI Mode go international for the first time, and multi-turn ads (ads that stay in a back-and-forth conversation) launch in the U.S. The evidence offered: inside AI Mode, sponsored and organic product listings get similar click-through rates. AI Mode queries run about 3x longer than classic searches, and most are incremental, meaning they would not have happened otherwise. What's missing is any number on how far AI Mode revenue per query trails classic Search. Agentic commerce (AI agents that shop and buy for users) is called additive, and consumer agents arrive in the Gemini app later this year. No revenue is attached to it yet.
5. Cloud doubled. Revenue grew 103% to $30.7B. Operating income was $11.1B, more than triple last year. Margin was 36.2% versus 23.7% a year ago. Backlog (signed contracts not yet booked as revenue) reached $561B, and just over 55% of it, about $310B, converts within 24 months. Two caveats. TPU system sales contributed to the growth but were not sized, and Cloud growth excluding TPUs was described as very strong with no number attached. Also, the rented third-party compute capacity came online only at the very end of Q3, so its margin drag lands in Q4.
6. TPU economics stay private. A TPU is Google's in-house AI chip, sold here as full racks with networking, storage and software. One analyst framed a TPU system at roughly $35B per gigawatt and asked how much of that becomes revenue and at what margin. Neither got an answer, because product-level financials are not disclosed. The cash cycle is clear. A signed contract shows up in backlog, Google builds inventory, and cash arrives on delivery. Inventory is only built against committed contracts. There are nearly 100 TPU customers, including Anthropic (directly and through Broadcom) and Apple, which is scaling pretraining on TPUs. New distribution comes through Oracle and Crux AI, a joint venture with Blackstone. The meaningful revenue ramp is 2027.
7. Free cash flow went negative. Q3 FCF was negative $9.1B on $51.4B of capex. Operating cash flow of $42.3B also absorbed the TPU inventory build and a $5.2B payment on a 2018 EC fine. Trailing twelve-month FCF is $19.7B. The 2026 capex guide is unchanged at $195B to $205B, and 2027 rises significantly, with numbers on the Q4 call. On top of that sit $924B in future purchase commitments and $152B in leases that haven't started and other obligations. Those are contractual maximums, and some have volume flexibility.
8. The return math is stated, not shown. Return on invested capital is said to stay meaningfully above the cost of capital, with no figure. Compute allocation is described as a long-range plan run on ROIC discipline, with no ranking of who gets capacity first. Cloud says it could serve a lot more demand with more capacity. About 65% of Q3 infrastructure spend went to servers and networking depreciated over six years, while TPU 8t claims up to 2.7x the training performance of the prior generation. My view: six years is a long life for hardware that loses that much ground each generation. Shorter lives would mean higher depreciation on top of the capex.
9. First-party model APIs process nearly 32B tokens (units of text an AI model reads or writes) per minute, up 40% QoQ and more than 4x YoY. In its first week on developer platforms, Gemini 4 Argon ramped 3.5x faster than Gemini 3, even with capacity limits. Customers who used Gemini Enterprise in Q3 2025 have more than quadrupled their spend. The Gemini app is at 1B users. This is the clearest evidence that the capex has real buyers.
Bottom line: The operating business beat. Revenue grew 25%, operating income grew 30%, Cloud doubled at a 36% margin, and the headline EPS miss is mostly a paper loss on SpaceX. The stock still fell about 10% after hours, from $322.22 to $288.35, because the market is now pricing cash, not growth. Quarterly FCF is negative, capex runs near $200B and rises again in 2027, commitments total about $1T, and the biggest profit pool is slowing as usage shifts to a surface that earns less per query. Management described every link between the spend and the payoff and did not put any number on them.
Show more
My key takeaways from the $GOOGL (Alphabet) Q3 2026 earnings call.
1. Search slowed while query growth hit a five-year high. Search and other ads grew 12% to $63.3B, down from Q2. The deceleration has four parts. FX was the biggest drag. The EU ended its de minimis exemption (duty-free treatment for cheap parcels), which hit APAC retail advertisers. World Cup media budgets landed in June instead of July. And AI Mode queries grew fast on a surface with "lower monetization" than classic Search. The first three wash out over time. The fourth is structural: more usage on a surface that earns less per query.
2. The EPS miss is mostly a SpaceX mark. Revenue grew 25% to $127.6B. Operating income grew 30% to $40.6B, a 31.8% margin. EPS came in at $2.12 against $3.02 expected because OI&E (other income and expense, the non-operating line where investment gains and losses land) swung to negative $8.2B. Most of that was an unrealized loss on the SpaceX stake after its share price fell. My rough math: after tax, that mark costs around $0.55 a share, so it explains more than half the $0.90 gap but not all of it.
3. Q4 Search is already guided softer. Lapping last year's Search acceleration cost about 3 points of growth in Q3 and will cost a couple more in Q4. FX gets slightly worse at current spot rates, mostly in Search and YouTube. The rest of Services won't make up the difference: YouTube ads grew only 9% to $11.2B, and Network grew 1% to $7.5B.
4. The AI Mode ad fix is rolling out but not proven. Ads in AI Mode go international for the first time, and multi-turn ads (ads that stay in a back-and-forth conversation) launch in the U.S. The evidence offered: inside AI Mode, sponsored and organic product listings get similar click-through rates. AI Mode queries run about 3x longer than classic searches, and most are incremental, meaning they would not have happened otherwise. What's missing is any number on how far AI Mode revenue per query trails classic Search. Agentic commerce (AI agents that shop and buy for users) is called additive, and consumer agents arrive in the Gemini app later this year. No revenue is attached to it yet.
5. Cloud doubled. Revenue grew 103% to $30.7B. Operating income was $11.1B, more than triple last year. Margin was 36.2% versus 23.7% a year ago. Backlog (signed contracts not yet booked as revenue) reached $561B, and just over 55% of it, about $310B, converts within 24 months. Two caveats. TPU system sales contributed to the growth but were not sized, and Cloud growth excluding TPUs was described as very strong with no number attached. Also, the rented third-party compute capacity came online only at the very end of Q3, so its margin drag lands in Q4.
6. TPU economics stay private. A TPU is Google's in-house AI chip, sold here as full racks with networking, storage and software. One analyst framed a TPU system at roughly $35B per gigawatt and asked how much of that becomes revenue and at what margin. Neither got an answer, because product-level financials are not disclosed. The cash cycle is clear. A signed contract shows up in backlog, Google builds inventory, and cash arrives on delivery. Inventory is only built against committed contracts. There are nearly 100 TPU customers, including Anthropic (directly and through Broadcom) and Apple, which is scaling pretraining on TPUs. New distribution comes through Oracle and Crux AI, a joint venture with Blackstone. The meaningful revenue ramp is 2027.
7. Free cash flow went negative. Q3 FCF was negative $9.1B on $51.4B of capex. Operating cash flow of $42.3B also absorbed the TPU inventory build and a $5.2B payment on a 2018 EC fine. Trailing twelve-month FCF is $19.7B. The 2026 capex guide is unchanged at $195B to $205B, and 2027 rises significantly, with numbers on the Q4 call. On top of that sit $924B in future purchase commitments and $152B in leases that haven't started and other obligations. Those are contractual maximums, and some have volume flexibility.
8. The return math is stated, not shown. Return on invested capital is said to stay meaningfully above the cost of capital, with no figure. Compute allocation is described as a long-range plan run on ROIC discipline, with no ranking of who gets capacity first. Cloud says it could serve a lot more demand with more capacity. About 65% of Q3 infrastructure spend went to servers and networking depreciated over six years, while TPU 8t claims up to 2.7x the training performance of the prior generation. My view: six years is a long life for hardware that loses that much ground each generation. Shorter lives would mean higher depreciation on top of the capex.
9. First-party model APIs process nearly 32B tokens (units of text an AI model reads or writes) per minute, up 40% QoQ and more than 4x YoY. In its first week on developer platforms, Gemini 4 Argon ramped 3.5x faster than Gemini 3, even with capacity limits. Customers who used Gemini Enterprise in Q3 2025 have more than quadrupled their spend. The Gemini app is at 1B users. This is the clearest evidence that the capex has real buyers.
Bottom line: The operating business beat. Revenue grew 25%, operating income grew 30%, Cloud doubled at a 36% margin, and the headline EPS miss is mostly a paper loss on SpaceX. The stock still fell about 10% after hours, from $322.22 to $288.35, because the market is now pricing cash, not growth. Quarterly FCF is negative, capex runs near $200B and rises again in 2027, commitments total about $1T, and the biggest profit pool is slowing as usage shifts to a surface that earns less per query. Management described every link between the spend and the payoff and did not put any number on them.
Show more
Now we know why $GOOGL and @sundarpichai were on a PR spree this week. Horrific.
ALPHABET $GOOGL Q3’26 EARNINGS HIGHLIGHTS
🔹 EPS: $2.12
🔹 Revenue: $127.6B (Est. $128.22B) 🔴; +25% YoY
🔹 Google Cloud Revenue: $30.7B (Est. $29.15B) 🟢
🔹 Search & Other: $63.4B (Est. $64.99B) 🔴
🔹 YouTube Ads: $11.2B (Est. $11.4B) 🔴
🔹 CapEx: $51.4B (Est. $55.2B) 🟢
🔹 CapEx to Revenue: ~40.3%
Key Updates:
🔹 Equity Securities: $8.7B net unrealized loss; EPS impact $0.55
Google Services:
🔹 Google Services Revenue: $96.3B (Est. $98.42B) 🔴
🔹 Google Services Oper Income: $39.2B (Est. $40.8B) 🔴
🔹 Google Services Operating Margin: ~40.7%
🔹 Google Advertising: $82.0B (Est. $83.6B) 🔴
🔹 Google Network: $7.5B (Est. $7.2B) 🟢
🔹 Subscriptions, Platforms & Devices: $14.3B (Est. $14.85B) 🔴
Google Cloud:
🔹 Google Cloud Oper Income: $11.1B (Est. $9.75B) 🟢
🔹 Google Cloud Operating Margin: ~36.2%
Other Bets:
🔹 Revenue: $372M
🔹 Operating Loss: -$2.0B
Alphabet-Level Activities:
🔹 Oper Loss: -$7.7B; shared AI research and development expenses
Other Metrics:
🔹 Total Oper Income: $40.6B (Est. $43.23B) 🔴
🔹 Total TAC: $16.4B
🔹 Revenue ex-TAC: ~$111.2B
🔹 Employees: 204,757
🔹 Paid Subscriptions: 150M+
🔹 First-Party Model APIs: nearly 32B tokens per minute
Geographic Revenue:
🔹 U.S.: $67.7B; +39% YoY
🔹 EMEA: $32.7B; +9% YoY
🔹 APAC: $19.9B; +12% YoY
🔹 Other Americas: $7.1B; +17% YoY
Financials:
🔹 Operating Margin: 31.8%
🔹 Net Income: $26.5B
🔹 Other Income, Net: -$8.2B
🔹 Gain on Equity Securities, Net: -$8.7B
🔹 Operating Cash Flow: $42.3B (Est. $52.72B) 🔴
🔹 Free Cash Flow: -$9.1B (Est. -$2.7B) 🔴
🔹 TTM Free Cash Flow: $19.7B
🔹 Cash & Marketable Securities: $265.1B
🔹 Long-Term Debt: $125.1B
Capital Return:
🔹 Dividend: $0.22/share
🔹 Stock Repurchases: $0
CEO Sundar Pichai Commentary:
🔸 “Google Cloud revenues grew 103%, powered by huge demand for our AI infrastructure and AI solutions. We’re also seeing strong demand for our TPU systems across a wide range of customers.”
🔸 “Overall, our first-party model APIs are now processing nearly 32 billion tokens per minute, more than four times what we processed a year ago.”
🔸 “On top of that, the Gemini app has reached one billion users—making it the fastest-growing product in Google’s history—and we surpassed 150 million YouTube Music and Premium subscribers globally, including trials.”
Show more
$STRL
ANOTHER BEAT AND RAISED GUIDANCE!!
Record quarter across the board:
• Revenue: $1.56B (+127% YoY)
• Adjusted EPS: $7.58 (+118%)
• Adjusted EBITDA: $337.8M (+117%)
• Organic revenue growth: 70%
The backlog story is getting even stronger:
• Signed backlog: $4.14B (+61% YoY)
• Combined backlog: $6.14B (+78%)
• Nearly $8B in total scoped opportunities, including future project phases
~95% of Sterling’s nearly $7B E-Infrastructure opportunity pipeline is tied to mission-critical projects, including data centers, manufacturing and semiconductor facilities.
CEC is another major growth driver. Management now sees potential for over $1B in revenue in 2026, with operating margins improving both YoY and sequentially.
FY2026 guidance raised:
• Revenue: $4.63B–$4.73B
• Adjusted EPS: $21.70–$21.96
• Adjusted EBITDA: $970M–$981M
At the midpoint, management now expects 88% revenue growth and 101% adjusted EPS growth for 2026.
Show more
$STRL
LET’S GOOO!
Incredible after-hours performance, especially with Google disappointing on earnings.
$STRL
Earnings drop today after market close (conference call is tomorrow morning at 9 AM ET)
the stock compressed to 18.5x forward adjusted EPS
management already flagged softer Q3 awards and a potential sequential backlog decline last quarter due to lumpiness
so expectations have already been reset
I like the R/R here
Show more
$STRL
Earnings drop today after market close (conference call is tomorrow morning at 9 AM ET)
the stock compressed to 18.5x forward adjusted EPS
management already flagged softer Q3 awards and a potential sequential backlog decline last quarter due to lumpiness
so expectations have already been reset
I like the R/R here
Show more
Big news for $IREN
Eases some of the concerns around its capital constraints
$IREN just closed $363M of AI compute asset financing. $MSFT reports tonight.
Both sit on the same chain. Money turns into compute. Compute turns into revenue faster.
This financing lowers the equity pressure on IREN and raises GPU asset turnover.
Same capital can pull more GPUs and more MW. That directly improves the two numbers that matter most for a neocloud: expansion speed and return on capital.
If MSFT keeps stressing AI compute is still tight and capex stays high, the IREN story gets cleaner. Better financing ability + customer demand still there + projects keep delivering.
For neoclouds the gap still comes down to who can turn contracts into revenue faster at lower cost.
Show more
One last rally before $ANT IPO?
This is my highest-conviction idea to date. Viking might be the highest-quality compounder I’ve come across, and the market is pricing it like a broken story on low-water worries. I am long $VIK, here’s why:
(1) Street models ~5% fleet CAGR off the disclosed orderbook, I see 7-8%. The ex-VP of New Builds confirmed shipyards have no delays and Viking can order more ships at any point. With 40-50% ROIC on its ships, it will exercise its options, lifting FY’28 EPS ~15% vs Street.
(2) Street fears low water will squeeze ‘27 and ‘28 demand, but channel checks say otherwise. Travel agents report little to no cancellations, and Viking raised prices on 53 of 125 tracked ‘27 cabin pairings. The Rhine and Danube routes hit by the water held their prices.
(3) The low-water hit is a one-off ~$191m across FY’26–29, not a broken river business. Identical Longships let Viking swap guests past the dry stretch instead of cancelling. Street has yield growth slowing to +4.9% by FY’28, I see it speeding up to +7.7%.
Quick background: Viking is the premium river and ocean cruise line for the 55+ traveler — no kids, no casinos, one language on board, identical ships across the fleet. Guests pay deposits 12–18 months out, 54% of them come back, and half book direct, so the company gets paid before it sails and spends less to fill cabins than anyone else in the industry.
Why it’s down: the stock went from ~$105 at the Aug highs to the high-$70s after Q2. The print itself was a beat (EPS $1.31 vs $0.95, revenue +16.5%), but management flagged that record-low water on the Rhine and Danube would hit Q3, and the market read that as the river franchise cracking. It came in from 31x to 23.5x NTM P/E on what is, in my view, a one-quarter weather problem on a business still 96% booked for 2026 and 53% for 2027.
Based on my extensive channel checks, expert calls with former employees, competitors, travel agents, and Viking's guests I believe VIK to be an incredibly attractive R/R scenario at these levels.
The entry point for this was across 78-83$ levels. Happy to chat about it, you can also find the PPT presentation for it on my website.
Show more
Reminder: $STRL reports tomorrow.
Read-through from $GLW (Corning) Q3 2026 earnings, for the sector.
1. $NVDA: AI fiber is growing faster than GPUs ship. Gen AI product sales more than doubled YoY and outgrew GPU units, so the fiber attached to each GPU is rising. All of that is scale-out, linking GPUs across a cluster, at roughly 16 fibers per GPU. Corning signed a multiyear manufacturing partnership with NVIDIA in May and says it sits inside the scale-up ecosystem. Weeks is "quite convicted that this accelerates in 2028." Read: neutral to bullish. The networking around NVIDIA systems keeps getting richer, but this is not a GPU volume signal. (inferred: NVIDIA not named on this call)
2. $META: hyperscaler capex is being prepaid, not just planned. In January Meta signed a deal worth up to $6B through 2030 for Corning fiber, cable and connectivity, anchoring a cable plant in Hickory, NC. Corning didn't name Meta on this call. It did say hyperscalers now put cash up front to fund capacity, which Corning returns as they buy the committed volume. Net customer deposits reached $639M in the first nine months. Customers prepaying for fiber they'll use in 2027 and 2028 is a harder durability signal than any capex guide. Read: neutral for $META's stock (it's spend), bullish for the networking suppliers selling into that build. (inferred: Meta not named on this call, deal confirmed separately)
3. $VZ $T: carriers are now buying fiber for data centers. Corning signed new long-term agreements with Verizon and AT&T this quarter (plus Zayo, which is private). They center on DCI, data center interconnect, the long-haul fiber linking one data center to another. Corning now says it will beat its old target of a $1B DCI business by 2030 on both timing and size. It also expects fiber-to-the-home builds by the big carriers to keep growing. For the carriers, it means hyperscaler wholesale demand is real enough to lock in multi-year supply. Read: mildly bullish for their wholesale fiber revenue, neutral on the stocks until that capex earns a return. (direct: management named them)
4. $LUMN: the DCI demand behind its fiber build is bigger than Corning planned for. Lumen came up on the call as an existing carrier partner. In 2024 it reserved 10% of Corning's global fiber capacity for two years to more than double its intercity network for cloud data centers. Corning raising its own DCI expectations is a demand check on exactly that bet. The catch is that supply is tight. Corning hands its newest high-density cable first to long-standing customers and those who prepay for capacity, so Lumen's next tranche likely needs a bigger commitment. Read: bullish on demand, neutral on supply terms. (direct on the relationship, inferred on the allocation)
5. $AAPL: cover glass fully onshored, device market soft. By the end of 2026, all cover glass for iPhone (including iPhone Duo) and Apple Watch sold worldwide will come from Corning's Kentucky plant, which makes a tariff-exposed input fully U.S.-made. Separately, Corning sized the device market for Q3: notebooks and handhelds down about 20%, TVs down 4%, with memory prices blamed. That's an industry figure, not an Apple one. Read: neutral, a supply-chain positive against a weak unit backdrop. (direct: management named them)
6. $CRDO: copper holds scale-up through 2027, and 2028 is the clock. Scale-up, the short links between GPUs inside one system, is 100% copper today per Corning. That is where Credo's AECs live (active electrical cables, copper with signal-boosting chips). Corning expects optics to start taking scale-up in 2028, and fully optical scale-up means about 160 fibers per GPU, 10x today. Credo is hedging with its own optical lines (ZeroFlap optics, plus active LED cables slated for fiscal 2028), but the copper franchise now has a stated window. Read: neutral near term, bearish on long-dated copper share. (inferred)
7. $LITE: Corning's scale-up date lines up with Lumentum's. Lumentum has guided to its first scale-up CPO shipments by late calendar 2027 (co-packaged optics, lasers and optics mounted right next to the chip) and calls optical scale-up its biggest growth driver. Corning sits at the fiber layer and independently puts the acceleration in 2028. Two suppliers at different layers pointing at the same window makes the timeline more credible. It is still management conviction, with no adoption rate attached. Read: bullish on timing, not yet on size. (inferred)
8. $APH: Corning's sold-out high-density cable is Amphenol's opening. Amphenol closed its $10.5B purchase of CommScope's connectivity business in January, which makes it a direct competitor in data center fiber connectivity at about $4.1B of 2026 sales. Corning says demand for its newest high-density cable is outrunning its ramp, so it rations by relationship length and willingness to prepay. Buyers who miss Corning's allocation have to source somewhere else. Read: bullish for $APH share gains, until Corning's planned 10x U.S. connectivity capacity build closes the gap. (inferred)
Bottom line: The strong read is fiber demand. Enterprise sales +84%, Gen AI sales more than doubled, and the DCI target beaten early are direct evidence that hyperscalers and carriers are still building, and the deposit model shows that spend is committed into 2027 and 2028. That supports the carriers ($VZ $T $LUMN) and the connectivity competition ($APH). The speculative read is scale-up timing. The $NVDA, $CRDO and $LITE chain rests on Corning's conviction and Lumentum's matching timeline, and nobody has given an adoption rate. $GLW also traded down about 4% on a beat, so the market isn't paying for "demand is strong" alone anywhere in this chain.
Show more
$GOOGL
The only hyperscaler that is suffering, but why?
Sundar Pichai says Google Cloud is deliberately sacrificing near-term revenue to give DeepMind priority access to scarce compute.
That allocation is “non-negotiable.”
The problem? The model performance isn't matching the spending
Gemini 4 Argon still trails Claude Opus 5.5 and sits mid-pack on most independent agent and coding benchmarks
Google even skipped 3.5 Pro after coding weaknesses surfaced during training
Billions in capex prioritized for DeepMind yet no clear model leadership
Show more
If history repeats itself
large caps like $INTC $ARM $AMD will rally first
Before small/mid caps like $NBIS $AEHR $AAOI
The part of the Anthropic filing that is most relevant if you own $IREN or $NBIS is below 👇
Anthropic filed the IPO.
What I care about most is operating cash flow turning positive to $3.929B in the first half.
Weve been asking whether AI applications can actually pay back the infrastructure spend. This filing gives part of the answer.
First half 2026 revenue $16.286B. Same period last year $1.163B. Up about 1,300%, roughly 14x.
Operating cash flow went from a $1.513B outflow to a $3.929B inflow.
That doesnt mean the financing problem is solved. The company explained part of the cash flow improvement came from payment timing. Customers paid first, supplier bills came later. First half still had about $693M operating loss. Future compute spend isnt going away.
The tension I watch is customers pay as they use. Anthropic still has to lock multi-year compute ahead of time.
Commercialization is speeding up. Revenue and cash collection still have to keep up with the purchase commitments. A successful IPO can add funding flexibility.
This matters for the $IREN I hold. Earlier media reports said the two signed a multi-year cloud service agreement. If that gets confirmed, Anthropic running better helps market confidence in the customers ability to pay.
Anthropic is also pushing direct chip leasing with dedicated buildings to try to lower long-term cost. Independent cloud names still have a shot at catching some of the demand.
Show more
Anthropic’s S-1 reveals an interesting opportunity for $NBIS 👀
47% of Anthropic’s 2025 revenue ($2.16B) flowed through $AMZN and $GOOGL cloud marketplaces.
Anthropic paid 16% in fees, far above the typical 1.5–3% marketplace fee.
Why? Claude is embedded directly into Amazon Bedrock and Google Vertex AI, making this more than a standard SaaS listing.
Could $NBIS capture some of that economics by bringing Claude to its Token Factory? 👀
Show more
