← Paulo Macro hub  ·  Research hub  ·  Research library

Paulo Macro — The Worm Has Turned: NVDA, AI, ORCL, Positioning, Uranium

"Once you are on the backside of the narrative and flow mountain, it is a long way down in time and price." The AI/NVDA/OpenAI/datacenter narrative has decisively turned — plus a positioning check and a bullish uranium aside.
2025-NOV-24 · Paulo Macro (Substack, paid) · written note · ↗ Read · note text · actionable insights
One-line take: The core call is that the AI/NVDA/OpenAI/datacenter narrative has turned for real — not a dip to buy. The catalyst was Google's Gemini 3, which "caught up and passed OpenAI" without NVDA GPUs or a Stargate/Oracle/Blue Owl financing morass (Berkshire bought GOOG at ATHs in Q3; "GOOG becomes the long, NVDA the funding short"). On NVDA, Paulo shorted the post-earnings pop and flags a receivables-vs-revenue divergence (rising cash consumption / working-capital strain) as the tell — even a leaked H200-to-China headline couldn't lift the stock (a "news failure" day). ORCL is the datacenter-credit posterchild (CDS blown out; ABS off-balance-sheet disintermediation; a possible "WorldCom of this cycle"). On positioning, fear finally showed but there's "a lot of room to go" (put/call, junk-credit divergence since Sept, margin debt, retail out of ammo, CTAs only partly de-risked) — he leans careful, wary that a violent Value/small-cap rotation could "converge on the downside." The one long: uranium — he was "buying again on Friday" as spot stabilized (Cameco + a utility bought $76-77/lb vs ~$85 term), betting on a Strategic Uranium Reserve + a triple-digit US floor price as the "wedge" that separates uranium from the AI/power selloff. Back-filled post.

1. Stocks & names mentioned

TickerNameResearchViewWhat's saidSource
GOOGLAlphabet (Google)QT · SA · STK · FAPositiveThe winner of the turn: Gemini 3.0 "caught up and passed OpenAI" without NVDA GPUs or a Stargate/Oracle financing morass; ~$70bn FCF/4 quarters, institutional memory, no key-man risk. Berkshire bought GOOG in Q3 at ATHs. Framed as the long leg of a GOOG-long / NVDA-funding-short pair.read ↗
CCJCamecoQT · SA · STK · FAPositiveBullish uranium tell: Cameco (+ a utility) showed up to buy spot at $76-77/lb vs a ~$85 term price with the carry wide open — and Cameco has been borrowing pounds to meet deliveries, "a game that can be played for only so long." Paulo "was buying again on Friday," betting on a Strategic Uranium Reserve + triple-digit US floor-price "wedge."read ↗
URAGlobal X Uranium ETFQT · SA · STKNeutralReference: the "key uranium ETF" whose momentum-factor / OKLO-meme flows dragged uranium miners into the AI/power-adjacent selloff — the crossfire uranium equities need a "wedge" to escape.read ↗
AnthropicAnthropic (private)NeutralReference: cited (via The Information) as poised to out-earn OpenAI this year on API/coding revenue — evidence OpenAI's technological lead has narrowed, part of the AI-narrative turn.read ↗
BRK.BBerkshire HathawayQT · SA · STK · FANeutralReference / confirmation: Berkshire bought Google in Q3 as it made new all-time highs — a "smart money" tell contrasted with SoftBank ("Masaponzi") selling NVDA to buy OpenAI. "These are not the same."read ↗
NVDANVIDIAQT · SA · STK · FANegativeThe funding short: Paulo shorted the post-earnings pop and added at the open. The tell is a YoY/QoQ revenue vs accounts-receivable divergence — rising cash consumption despite revenue growth, "a massive strain on working capital." Even a leaked H200-to-China headline couldn't lift it (a "news failure"); "the worm has turned," a long way down in time and price.read ↗
ORCLOracleQT · SA · STK · FANegativeThe datacenter-credit posterchild: CDS blown out for weeks; debt pushed off-balance-sheet into ABS ("opaque disintermediation"); Deutsche shows ORCL as a huge outlier. "May yet end up as the WorldCom" of this cycle's datacenter-SPV insanity — "the numbers will not math."read ↗
OpenAIOpenAI (private)NegativeAt the center of the turn: The Information's leak of Altman's "rough vibes / economic headwinds" memo after Gemini 3's leapfrog; ~$500bn valuation but projected >$100bn cash burn while inference costs run ahead of revenue. "No culture" vs Google's institutional memory; key-man risk.read ↗
PLTRPalantirQT · SA · STK · FANegativeInsider-selling tell: CEO Alex Karp sold 585,000 shares for $96mn on Thursday — right after his "hubristic defiant" CNBC appearance two weeks earlier. Cited as one of the frothy AI-adjacent risk assets to keep in context.read ↗
OKLOOklo IncQT · SA · STK · FANegativeNamed pejoratively: "OKLO meme-type garbage" driving the URA ETF and dragging real uranium miners into the AI/power-momentum selloff — the froth uranium equities need to separate from.read ↗
BMNRBitMine Immersion TechnologiesQT · SA · STK · FANegativeCrypto-avalanche example: sitting on a ~$4bn loss on its Ether bet; flagged "structural issues" (opaque, costly structure; leadership/advisor comp could extract ~$157mn/yr) — part of the NVDA-and-bitcoin illiquidity/momentum linkage.read ↗

"View" reflects this note's framing (Positive = the two names he's constructively positioned in — GOOG long, uranium via Cameco; Neutral = context/confirmation references; Negative = the turning AI/datacenter/crypto complex he's short or cautious on). OpenAI and Anthropic are private (no ticker). Bitcoin is an asset class, not a security, so it's discussed but not tabled. Written post — no video, so "Source" opens the Substack note. Research: QT Qualtrim · SA Seeking Alpha · STK Stock Analysis.

2. Talking points

The framing question — Risk Off or a fierce rotation?

Uranium — buying the crossfire

The turn — Gemini 3 is the real news

NVDA — the receivables tell

The "average price magnet" diary

ORCL — the datacenter-credit posterchild

Positioning — room to go

Crypto — the illiquidity linkage

Bottom line — be careful; the rotation may not save you

3. In plain English

A jargon-free summary of the thesis behind each name — what it actually is and why that view. (Plain-language companion to the table above; renders on each ticker's consolidated page.)

GOOGL — Alphabet (Google) Positive

For two years the AI story was "OpenAI and NVIDIA win everything." Paulo argues that just flipped: Google's new Gemini 3 model caught up to and passed OpenAI — and Google did it with its own chips and its own cash, without the expensive borrowed-money data-center schemes the others rely on. Google throws off about $70 billion of spare cash every few months, has deep engineering culture, and doesn't depend on one founder.

So he frames it as a pair trade: own Google (the durable winner, which even Warren Buffett's Berkshire bought at record highs) and short the names that need constant new funding to keep the story alive. Google is the "long" side.

CCJ — Cameco Positive

Uranium (nuclear-plant fuel) sold off because it had been lumped in with the AI/power hype and momentum-chasing funds. Paulo thinks that selloff is a buying chance. Cameco, the biggest Western uranium miner, and a power utility both stepped in to buy uranium at $76-77 a pound even though longer-term contracts are priced around $85 — a sign the metal is cheap. Cameco has even been borrowing uranium to meet its delivery promises, which it can't do forever, so it has strong reason to buy.

His bigger bet is a political catalyst: he expects Washington to create a Strategic Uranium Reserve and possibly guarantee a high floor price for US-mined uranium (as it did for rare earths). That would give uranium stocks their own reason to rise, separate from the AI selloff dragging them down. He was buying again.

NVDA — NVIDIA Negative

NVIDIA makes the chips that power the AI boom and has been the market's biggest winner. Paulo is now short it. His main red flag is technical but simple: NVIDIA's revenue keeps growing, but the money customers owe it but haven't paid yet (accounts receivable) is growing even faster, and its actual cash is getting tighter. In plain terms, sales look great on paper while real cash is under strain — often a warning sign, especially when the company is also helping fund the very customers buying its chips.

He thinks the "story" that drove the stock has peaked — and once a story turns, "it's a long way down." A sign of that: normally a bullish headline (letting NVIDIA sell chips to China) would pop the stock; instead it fell, which he reads as the market no longer rewarding good news.

ORCL — Oracle Negative

Oracle went all-in on building AI data centers, largely financed with debt. Paulo sees it as the poster child for the riskiest part of this cycle. The cost to insure Oracle's debt against default (its "CDS") has spiked, and a lot of the borrowing has been moved off Oracle's own books into complex, hard-to-see financing structures. He compares it to WorldCom — a company whose aggressive accounting and debt eventually blew up — and says "the numbers will not math": the promised revenue doesn't justify the spending. A name to avoid, and a warning for the whole data-center-credit complex.


Key points extracted from the paid Substack post (in transcript.txt) for personal study. Not investment advice. © Paulo Macro for source material.