Podcast trade ideas

Prof G Markets

How Meta Could Quietly Win The AI Race

Sep 21, 20263h

The hosts argue the Fed's surprise quarter-point rate hike was the right call for inflation and Fed independence, though they warn higher rates and a 10-year yield above 5% could pressure stocks near term. Their most actionable view is bullish Meta: Ed owns it (bought around $564), expects MetaMuse to rival ChatGPT and Claude on the consumer side, and both note Meta trades cheaply versus Big Tech, while Ed calls Apple a sell. They also flag Canada/EU trade realignment as a shift away from the US and predict a country or bloc will ban US frontier AI models within 90–180 days.

Positions · 6

• 42:19METALongMetaMuse's consumer AI launch leverages Meta's unmatched first-party data and nine-of-ten internet-user reach, and the stock is still cheap versus Big Tech. — Ed Elson0.0% since episode
  Why this trade

  “I bought in June $564 a share. I'm up around 20%... But I still think it has room to run because it's cheap.” — Ed Elson

  1. Ed says he bought Meta in June at $564/share and is up ~20%, arguing it has room to run because it's cheap.
  2. Scott argues Meta's 4B monthly users and first-party data give it a unique advantage in consumer AI via MetaMuse.
  3. Resulting trade: stay long META as the AI product gains adoption and multiples remain below peers.
• 1:04:51METALongMetaMuse will see significant adoption and rival ChatGPT and Claude on the consumer side within six months. — Ed Elson0.0% since episode
  Why this trade

  “My prediction, I think that MetaMuse will see significant adoption. I think it'll be rivaling ChadGBT and Claude in the next six months... on the consumer side.” — Ed Elson

  1. Ed predicts MetaMuse will see significant adoption over the next six months, rivaling ChatGPT and Claude on the consumer side.
  2. Meta leveraged open source and first-party data to ship a cheap, useful consumer agent that tracks tasks users actually want.
  3. Resulting trade: long META as consumer AI adoption drives engagement and transaction take-rate.
• 42:05AAPLAvoidApple trades at 38x earnings, which Ed finds remarkably expensive, and he still doesn't think it's anywhere close to a buy. — Ed Elson+0.2% since episode
  Why this trade

  “Apple, 38 times earnings. I'm just still blown away by the by how expensive Apple stock is right now. I still don't think it's anywhere close to a buy. I think it's a sell, actually.” — Ed Elson

  1. Ed reviews Big Tech PE multiples and notes Apple at 38 times earnings, the richest of the group.
  2. He says he is blown away by how expensive Apple stock is and sees no value at this price.
  3. Resulting trade: avoid/sell AAPL versus cheaper peers like Meta at 25x.
• 42:19MSFTLongEd cites Microsoft among his bullish picks, up more than 20%, as part of his broad AI-linked Big Tech longs. — Ed Elson+0.2% since episode
  Why this trade

  “I think it's pretty clear that I've had multiple bullish picks, including Meta and Microsoft and Salesforce and ServiceNow, which are up all up more than 20% at this point.” — Ed Elson

  1. Ed says he has had multiple bullish picks including Meta, Microsoft, Salesforce and ServiceNow.
  2. He contrasts Big Tech valuations, noting Microsoft's adjusted PE around 28x.
  3. Resulting trade: long MSFT as a cheaper mega-cap AI beneficiary among his picks.
• 42:19CRMLongEd lists Salesforce among his bullish picks, all up more than 20% at this point. — Ed Elson+0.7% since episode
  Why this trade

  “I've had multiple bullish picks, including Meta and Microsoft and Salesforce and ServiceNow, which are up all up more than 20% at this point.” — Ed Elson

  1. Ed says he has had multiple bullish picks including Salesforce, up more than 20%.
  2. He uses these picks to rebut claims he is a permabear.
  3. Resulting trade: long CRM as one of his identified winners.
• 42:19NOWLongEd lists ServiceNow among his bullish picks, up more than 20% at this point. — Ed Elson+0.6% since episode
  Why this trade

  “I've had multiple bullish picks, including Meta and Microsoft and Salesforce and ServiceNow, which are up all up more than 20% at this point.” — Ed Elson

  1. Ed says his bullish picks include ServiceNow, up more than 20%.
  2. He cites these gains to counter criticism that he is a permabear.
  3. Resulting trade: long NOW as one of his identified winners.

Ideas · 5

• 37:58OPENAIAvoidScott warns OpenAI and Anthropic are fragile, hating their doomerism and suggesting frontier labs could lose 70% of value within a year if there's a public AI incident. — Scott Galloway-0.5% since episode
  Why this trade

  “I have never seen 2,000,000,000,000 plus companies, Anthropic and OpenAI, more fragile right now... they could be down 70% in value within twelve months.” — Scott Galloway

  1. Scott says he has never seen $2T+ companies like Anthropic and OpenAI more fragile and that people hate them.
  2. He suggests they could be down 70% in value within twelve months if a frontier-model mishap causes public backlash.
  3. Resulting trade: avoid exposure tied to OpenAI/Anthropic frontier-model economics.
• 37:58ANTHROPICAvoidScott sees Anthropic and OpenAI as fragile and disliked, with Anthropic's doomerism opening a void Meta is filling. — Scott Galloway+0.5% since episode
  Why this trade

  “I have never seen 2,000,000,000,000 plus companies, Anthropic and OpenAI, more fragile right now. I think there's a fairly... People hate them.” — Scott Galloway

  1. Scott says Anthropic and OpenAI are more fragile than ever and people hate them.
  2. Ed cites Ramp data showing Anthropic's enterprise share at 8% versus Astra at 13%, suggesting it is falling behind.
  3. Resulting trade: avoid Anthropic-linked exposure as Meta and others gain share.
• 31:50GOOGLLongEd's dinner test shows Google Gemini passed the task, positioning Alphabet's Gemini as a capable consumer AI agent despite extra confirmation steps. — Ed Elson-0.9% since episode
  Why this trade

  “Gemini presented Dan with some restaurants... Then it asked to confirm again... It passed the test, but with a lot of extra steps and a lot of annoying confirmations.” — Ed Elson

  1. Ed tested four agents (Muse, Claude, ChatGPT, Gemini) on a booking task and Gemini completed the reservation.
  2. The product is relevant to a huge consumer use case, though it required more confirmations than Meta's Muse.
  3. Resulting trade: mild positive for GOOGL's Gemini consumer-agent positioning.
• 14:59TLTLongEd notes the 10-year yield rose above 5% and warns higher rates mean bonds offer higher risk-free returns, transferring capital from equities to credit. — Ed Elson+0.2% since episode
  Why this trade

  “the ten year yield rose again above 5%, and I think that is gonna be the really important thing to keep track of.” — Ed Elson

  1. Ed says the 10-year yield rose again above 5% and that the jury is out on inflation expectations.
  2. Scott notes a transfer of capital from equity markets into bond markets as bonds offer 5-6% risk-free.
  3. Resulting trade: bonds/credit become relatively more attractive than equities in this hiking environment.
• 18:08SPYLongScott notes that historically after a first Fed hike the S&P 500 typically gains about 9% over the following twelve months, favoring staying invested. — Scott Galloway+0.1% since episode
  Why this trade

  “while initially a hike takes the S and P down over the following twelve months, there's typically on average a 9% gain, in the S and P.” — Scott Galloway

  1. Scott cites economic history that a first hike takes the S&P down initially but averages a ~9% gain over the following year.
  2. He argues the market responds well to fiscal discipline and Fed independence.
  3. Resulting trade: stay long broad US equities (S&P 500) over a 12-month horizon.

Prof G Markets

He Warned AI Could Destroy Us. Now The Industry Is Listening — ft. Nick Bostrom

Sep 18, 202675h

Ideas · 4

• 32:12TSMLongHe argues leading-node fab capacity is already nearly maxed out by AI demand, making TSMC a structural bottleneck and beneficiary of the AI buildout. — Nick Bostrom+1.8% since episode
  Why this trade

  “it's a significant fraction of the total production of TSMC in the leading node is going to these NVIDIA chips” — Nick Bostrom

  1. Bostrom says a significant fraction of TSMC's leading-node production already goes to NVIDIA AI chips.
  2. That means new capacity requires building fabs, which takes time and gives TSMC pricing power and strategic importance.
  3. Implication is that TSMC remains a key chokepoint supplier as long as AI compute demand persists.
• 32:12NVDALongHe says AI progress is driven largely by massive compute scaling, with NVIDIA chips consuming much of TSMC's leading-node output, implying continued demand. — Nick Bostrom+1.2% since episode
  Why this trade

  “already now, it's a significant fraction of the total production of TSMC in the leading node is going to these NVIDIA chips” — Nick Bostrom

  1. Bostrom attributes a large part of AI gains to scaling up compute, especially GPU clusters.
  2. He notes NVIDIA chips are the destination for a large share of TSMC leading-node production.
  3. That supports NVIDIA's central position in the AI infrastructure spending cycle.
• 1:11OPENAIAvoidHe highlights that OpenAI will not go public this year, meaning retail investors cannot directly buy the leading private frontier lab. — Speaker 5+4.1% since episode
  Why this trade

  “Sam Altman said he agreed with Amade and added OpenAI will not be going public this year.” — Speaker 5

  1. Speaker 5 states Sam Altman said OpenAI will not be going public this year.
  2. Bostrom discusses OpenAI as a private frontier lab facing intense competitive and safety pressures.
  3. The practical takeaway for public-market investors is that OpenAI equity is unavailable.
• 1:11ANTHROPICAvoidAnthropic is discussed as a private frontier lab whose CEO is calling for a synchronized AI slowdown, but there is no listed vehicle for direct exposure. — Speaker 5+0.7% since episode
  Why this trade

  “Anthropic CEO Dario Amade published an essay calling for the industry to slow down the development of AI models.” — Speaker 5

  1. Dario Amodei of Anthropic published an essay calling for the industry to slow AI development.
  2. Bostrom discusses Anthropic and OpenAI as private frontier labs with intense competitive dynamics.
  3. No public ticker or investable listed proxy is identified in the episode.

Prof G Markets

Fed Hikes Rates For First Time In 3 Years — Here’s Why It Matters

Sep 17, 202699h

Ideas · 4

• 1:38Bank stocks (sector, via major indices)AvoidBank stocks had their worst day since February on concerns that higher rates will slow lending growth, a channel Ed flags as a key market risk. — Speaker 6–
  Why this trade

  “bank stocks had their worst day since February on concerns that higher rates will slow lending growth.” — Speaker 6

  1. Ed notes bank stocks fell the most since February on fears higher rates slow lending growth.
  2. That links Fed tightening directly to bank earnings sensitivity.
  3. The read-through is caution on bank stocks amid a hiking cycle.
• 15:26NVDALongArmstrong argues hyperscaler AI capex is extremely price insensitive, noting builders will absorb even a 50% GPU price jump, protecting Nvidia's demand. — Speaker 7+3.6% since episode
  Why this trade

  “The price of NVIDIA GPU goes up 50%. Fine. I'll pay it.” — Speaker 7

  1. Armstrong says firms building data centers are price insensitive and will pay more for GPUs.
  2. That implies continued demand for Nvidia's AI chips even as financing costs rise.
  3. Supports a constructive stance on Nvidia amid the AI capex boom.
• 15:32Hyperscalers / AI infrastructure spendersLongArmstrong says the AI capex boom is a growth driver keeping nominal GDP ~6% and that an extra 100bps won't derail projects—only worth watching in six months. — Speaker 7–
  Why this trade

  “another 100 basis points on your interest bill are not gonna bug you.” — Speaker 7

  1. Armstrong notes growth is ~3% plus 3% inflation, with AI investment a real support.
  2. He says financing costs 'don't matter until they do' and likely won't change spending now.
  3. Supports staying constructive on AI capex leaders, watching borrowing costs over the next six months.
• 30:38Chinese open-weight model companies (e.g., DeepSeek, Moonshot)AvoidHan warns Congress could next year make it hard for US companies or clouds to use Chinese open-weight models, a real risk to China's AI exporters. — Speaker 9–
  Why this trade

  “make it difficult for US companies to use Chinese open weight models” — Speaker 9

  1. Han says elites see AI politicization distracting from anti-China regulation.
  2. She flags potential US restrictions on Chinese open-weight models as a coming issue.
  3. That regulatory risk pressures Chinese open-weight AI firms.

Prof G Markets

AI Insiders Keep Saying We’re In Danger — Where’s The Evidence?

Sep 16, 2026123h

Ideas · 1

• 20:07ANTHROPICAvoidZittrin expects Anthropic's S-1 to reveal an immensely unprofitable company once trading costs, stock comp, and revenue-share payments to Amazon/Google/Microsoft are included. — Ed Zittrin+1.0% since episode
  Why this trade

  “Well, that 80% gross margin, to be clear, did not include trading costs or stock based compensation. It's kind of like saying I'm profitable if you don't include my costs.” — Ed Zittrin

  1. Zittrin claims the reported 80% gross margin and adjusted operating profitability exclude trading costs and stock-based compensation.
  2. He also says these figures exclude revenue-share payments to Amazon, Google, and Microsoft used to inflate revenues.
  3. Conclusion: the upcoming Anthropic IPO will show accounting shenanigans and deeply negative true economics.

Prof G Markets

Why OpenAI And Anthropic Are Pumping The Breaks

Sep 15, 2026147h

Prof G Markets discusses the AI 'pace the frontier' debate, with Base10's Charlie O'Neil arguing that OpenAI and Anthropic's safety push won't slow compute demand—it may even increase it—so he rejects the bearish AI-slowdown read that hit NVIDIA (-3%), Oracle (-4%), CoreWeave (-7%) and SoftBank (-15%). Host Ed Elson is skeptical of Anthropic's claim of two quarters of adjusted operating profitability, saying he'll 'believe it when I see it,' while noting OpenAI told Fortune it won't IPO in 2026.

Positions · 1

• 27:31ANTHROPICAvoidEd Elson says he won't trust Anthropic's pre-IPO profitability claims because the adjustments exclude revenue-sharing and training costs—'I will believe it when I see it.' — Ed Elson+1.1% since episode
  Why this trade

  “when it comes to the profitability of AI, I stand by what I said last week, and that is that I will believe it when I see it.” — Ed Elson

  1. Anthropic told investors it has been profitable for two straight quarters ahead of its IPO.
  2. Elson notes profitability is only 'adjusted operating,' with >80% gross margins calculated before revenue sharing and model training costs.
  3. Without S-1 clarity, he treats the headline as uninvestable hype.

Ideas · 5

• 5:07NVDALongO'Neil says the labs' safety push requires MORE compute, not less, so the AI-slowdown selloff that took NVIDIA down 3% is misguided. — Charlie O'Neil+6.1% since episode
  Why this trade

  “it's certainly not gonna be a, you know, a a bearish sign for, like, the amount of compute the world is gonna need over the next few years.” — Charlie O'Neil

  1. O'Neil argues OpenAI/Anthropic will allocate up to 20%+ of compute to safety while still scaling model roadmaps.
  2. More compute allocation means labs 'might even see the labs be even more aggressive with compute build outs' for GPUs.
  3. Monday's AI-adjacent selloff (NVIDIA -3%) is therefore a buying opportunity rather than a warning.
• 15:51ORCLLongOracle fell 4% on AI-spending-slowdown fears, but O'Neil argues compute demand keeps rising, making the dip a mispriced reaction. — Charlie O'Neil+4.0% since episode
  Why this trade

  “Running a model on compute has never been more valuable, and I don't think there's any world in which any lab... are going to want to stop spending on compute.” — Charlie O'Neil

  1. The episode lists Oracle down 4% among AI-adjacent names sold off on slowdown concerns.
  2. O'Neil says compute only becomes more valuable ($10M/MW to $20-25M/MW next year), so spending won't stop.
  3. Falling on a thesis the guest rejects implies the selloff overshot fundamentals.
• 5:07CRWVLongCoreWeave dropped 7% on AI-capex worries, but O'Neil's 'more compute, more buildouts' argument suggests demand for GPU cloud capacity remains intact. — Charlie O'Neil0.0% since episode
  Why this trade

  “we might even see the the labs be even more aggressive with compute build outs and and securing compute.” — Charlie O'Neil

  1. CoreWeave sold off 7% Monday on fears an AI slowdown would hit spending.
  2. O'Neil claims labs will secure and build even more compute for safety and training.
  3. That supports demand for third-party GPU capacity providers like CoreWeave.
• 2:52SOFTBANKLongOpenAI → SoftBank fell 15% as a major OpenAI investor on slowdown fears, but O'Neil says OpenAI's compute spend is only going up, and Altman said no 2026 IPO rather than a wind-down. — Ed Elson+2.4% since episode
  Why this trade

  “SoftBank, which is a significant investor in OpenAI closed down 15%.” — Ed Elson

  1. SoftBank closed down 15% because it is a significant OpenAI investor and markets feared an AI pullback.
  2. O'Neil argues the labs won't cut compute spending because each megawatt is getting more valuable.
  3. If OpenAI spend continues, the read-through to its largest backer is positive.
• 25:21OPENAIAvoidElson highlights OpenAI's $20B+ operating loss and Altman's own statement that an IPO in 2026 is 'ill advised,' keeping the core AI business model under a cloud. — Ed Elson+6.8% since episode
  Why this trade

  “we learned that OpenAI racked up more than $20,000,000,000 in operating losses last year.” — Ed Elson

  1. Leaked documents show OpenAI racked up over $20 billion in operating losses last year.
  2. Altman told Fortune OpenAI won't go public in 2026, calling it an 'ill advised moment.'
  3. Together these undermine the AI business-model bull case Elson has questioned repeatedly.

Prof G Markets

The Rate Hikes Are Coming

Sep 14, 2026171h

Positions · 5

• 1:01:16AAPLLongGalloway is cautiously optimistic the foldable iPhone Duo, priced near $2,000, is a hit because Apple is the 'world's best, most profitable second mouse in history'. — Scott Galloway+1.3% since episode
  Why this trade

  “I am cautiously optimistic about this one... Apple is gonna come in and sell more foldable phones in the first sixty days that have been sold in the last seven years” — Scott Galloway

  1. Galloway argues Apple waits, learns and makes a more elegant product than first movers like Samsung, and that iOS's luxury-brand signaling drives pricing power.
  2. He says the foldable can replace both iPhone and laptop, and he personally tried to buy one on launch day.
  3. Resulting trade: bullish Apple hardware cycle, though he hedges with 'cautiously optimistic'.
• 1:02:58AAPLAvoidEd remains bearish on Apple stock, seeing the foldable as unimpressive and Apple's 36x multiple unjustified given weak AI/product innovation. — Ed Elson+1.3% since episode
  Why this trade

  “I have been publicly kinda bearish on Apple for a while... I would be more bullish actually on building a data center than selling a foldable iPhone” — Ed Elson

  1. Ed says the device is 'too large and too cumbersome' and not compelling versus existing foldables, and that he wants to see more from Apple.
  2. He notes Apple trades at 36x earnings vs ~26x for the S&P while spending far less on CapEx than Alphabet.
  3. Resulting trade: avoid/stay bearish Apple despite being wrong on it so far this year.
• 21:07Long-dated US TreasurysAvoidEd says bond investing is 'dying', warning persistent 3-5% inflation eats yields and long Treasuries just posted their worst decade in 100+ years. — Ed Elson–
  Why this trade

  “bond investing is in a lot of ways dying... it does call into question the sixty forty portfolio of 40% bonds because at least in the past decade, that portfolio structure has not been rewarded at all” — Ed Elson

  1. Ed argues structural inflation will require far higher yields for bonds to be worth it, eroding real returns.
  2. He says the 60/40 portfolio hasn't been rewarded on a risk-adjusted basis for a decade.
  3. Resulting trade: underweight/avoid long-dated Treasuries; favor equities.
• 21:07US EquitiesLongEd argues that in a forever-inflation regime investors must 'invest more aggressively in stocks than ever before' since equities are the only asset class being rewarded. — Ed Elson–
  Why this trade

  “you have to invest more aggressively in stocks than ever before because your purchasing power is going down, and the only asset class that is actually being rewarded is equities” — Ed Elson

  1. Ed says purchasing power is declining and structural inflation now looks persistent.
  2. He argues bonds won't compensate, so equities are the only rewarded asset class.
  3. Resulting trade: overweight stocks despite bubble concerns.
• 22:43Investment-grade Corporate BondsLongGalloway says healthy corporate credit yielding 6-8% is now rewarding risk-takers, and a mixed stock/bond portfolio still makes sense. — Scott Galloway–
  Why this trade

  “if you can buy a corporate bond that looks very healthy, you know, and get a six, seven, 8% return, you know, that's not bad... it's to not get poor. The way you don't get rich, you you not get rich, but you don't get poor is with bonds” — Scott Galloway

  1. Galloway argues for the first time debt investors are being paid for the risk they take (6-8% yields).
  2. His strategy is 'not to get rich, it's to not get poor', favoring less-volatile credit with real teeth.
  3. Resulting trade: buy high-quality corporate bonds/credit over Treasuries.

Ideas · 10

• 55:06ANTHROPICAvoidEd doubts Anthropic's claimed adjusted operating profitability and warns the upcoming S-1's $30T TAM headline is likely massaged numbers. — Ed Elson+1.1% since episode
  Why this trade

  “I believe that neither company is even close to profitable. That's my personal belief... I think this profitability question is enormous.” — Ed Elson

  1. Ed says the arguments for AI economics making sense are unconvincing and Anthropic's profitability claim likely reflects 'BS accounting'.
  2. He expects the S-1 total addressable market of $30 trillion to involve 'a lot of massaging of numbers'.
  3. Resulting trade: skepticism/avoid ahead of the Anthropic IPO.
• 51:17OPENAIAvoidGalloway is torn but leans bearish, comparing AI capex needs ($2.5T) versus ~$150-200B of revenue to railroads/Internet and expecting 'an enormous correction'. — Scott Galloway+9.5% since episode
  Why this trade

  “we're gonna have a an enormous correction down... this looks more like the railroads and the Internet, and that is the technology will survive these valuations, but these valuations are are gonna have real volatility” — Scott Galloway

  1. Galloway says AI must reach $2.5T revenue to justify capex versus current ~$150-200B.
  2. He sees parallels to railroads, the electric grid and the Internet, where technology survives but valuations correct.
  3. Resulting trade: cautious/avoid at current AI valuation levels while acknowledging huge ARR growth.
• 41:36ANTHROPICShortEd argues an Anthropic insider's viral 'AI could kill all humans' claim, endorsed by a current researcher, is material public information that could invite securities fraud scrutiny ahead of the IPO. — Ed Elson+1.1% since episode
  Why this trade

  “maybe even dry up the stock price. Well, then we we need to start talking about different issues like securities fraud... we need to talk about issues about public deception” — Ed Elson

  1. Ed argues the alignment lead's statement is a corporate communication, meaning the company must explain or face investigation.
  2. If hyperbolic to stoke IPO hype, it raises public deception/securities fraud issues.
  3. Resulting trade: bearish/avoid setup into the Anthropic IPO until claims are substantiated.
• 23:24SHELLongGalloway cites Shell as a wealth-transfer winner from $100+ oil, up 27%, benefiting Permian/energy landowners and exporters while consumers lose. — Scott Galloway–
  Why this trade

  “Some of the winners with 100 plus dollar oil, Shell's up 27%. ExxonMobil, 34%. Chevron, 37. Norway's, Equinor is up 84.” — Scott Galloway

  1. Galloway says high oil prices transfer wealth from consumers at the pump to energy producers.
  2. He lists Shell among major oil gainers (up 27%) in this environment.
  3. Resulting trade: bullish read-through to big oil equities as a macro inflation beneficiary.
• 23:24XOMLongGalloway lists ExxonMobil (up 34%) as a direct beneficiary of $100+ oil and the wealth transfer from diesel-dependent consumers to producers. — Scott Galloway-3.6% since episode
  Why this trade

  “ExxonMobil, 34%. Chevron, 37. Norway's, Equinor is up 84. They supply 30% of all the natural gas consumed in The EU.” — Scott Galloway

  1. Galloway says the US being a net energy exporter means higher energy prices enrich producers like Exxon.
  2. He names ExxonMobil up 34% among oil winners.
  3. Resulting trade: bullish oil-major equity exposure on sustained $100 oil.
• 23:24CVXLongGalloway cites Chevron (up 37%) as another high-oil winner from the consumer-to-producer wealth transfer. — Scott Galloway-3.6% since episode
  Why this trade

  “ExxonMobil, 34%. Chevron, 37. Norway's, Equinor is up 84.” — Scott Galloway

  1. Galloway says energy producers are among the few winners from $100+ oil while consumers lose purchasing power.
  2. He names Chevron up 37% in the winners list.
  3. Resulting trade: bullish big-oil energy equities.
• 23:24EQNRLongGalloway notes Equinor is up 84%, benefiting as the supplier of 30% of the EU's natural gas amid the oil conflict. — Scott Galloway–
  Why this trade

  “Norway's, Equinor is up 84. They supply 30% of all the natural gas consumed in The EU.” — Scott Galloway

  1. Galloway cites Equinor as a major winner up 84% on higher energy prices.
  2. He notes it supplies 30% of EU-consumed natural gas, tying its fortunes to the European energy squeeze.
  3. Resulting trade: bullish Equinor as a European gas beneficiary.
• 23:39UNPLongGalloway lists Union Pacific (up 23%) as a railway winner because rails are far more energy-efficient than trucking when diesel hits records. — Scott Galloway–
  Why this trade

  “railroads, which are much more efficient from an energy standpoint, Canadian Pacific, Kansas City, plus 21%. Union Pacific, 23%. CSX, up 34%. So a transfer of wealth” — Scott Galloway

  1. Galloway says $100 oil and record diesel make energy-efficient rail more attractive versus trucking.
  2. He cites Union Pacific up 23% among rail winners.
  3. Resulting trade: bullish rails on fuel-cost-driven modal shift.
• 23:39CSXLongGalloway cites CSX (up 34%) among energy-efficient rail winners as diesel hits record highs. — Scott Galloway–
  Why this trade

  “Union Pacific, 23%. CSX, up 34%.” — Scott Galloway

  1. Galloway says railroads benefit from being far more energy-efficient than trucks amid $100 oil.
  2. He lists CSX up 34% in the rail winners group.
  3. Resulting trade: bullish CSX as a macro rail beneficiary.
• 23:39CPLongGalloway lists Canadian Pacific Kansas City (up 21%) as a rail winner due to energy efficiency advantage over diesel trucking. — Scott Galloway–
  Why this trade

  “Canadian Pacific, Kansas City, plus 21%.” — Scott Galloway

  1. Galloway notes rails outperform when fuel costs spike, citing CPKC up 21%.
  2. This follows from the wealth transfer to energy-efficient transport from record diesel prices.
  3. Resulting trade: bullish CPKC in a high-oil regime.

Prof G Markets

Why The Bond Market Is Starting To Revolt — ft. Katie Martin

Sep 11, 2026243h

Positions · 1

• 1:06:45US Federal Reserve rate path (Fed funds)ShortMartin explicitly expects the Fed to hike rates before year-end: 'I do. Yeah... If inflation stays sticky and payrolls keep doing what they're doing, I don't see how Walsh has a choice.' — Katie Martin–
  Why this trade

  “Do you think we see a rate hike before the end of the year? I do. Yeah.” — Katie Martin

  1. Blowout 162k jobs report moved market odds to a two-in-three chance of a hike
  2. Martin says sticky inflation and strong payrolls force Warsh to hike this month or before year-end
  3. Higher rates mean higher borrowing costs and further pressure on bonds

Ideas · 10

• 5:27TLTAvoidMartin: governments are 'borrowing too much damn money' and bond investors now demand a higher return, so long-end Treasuries are unattractive. — Katie Martin-3.9% since episode
  Why this trade

  “The twenty nine year Treasury yield recently reached its highest level since before the financial crisis... bond investors say, I'll buy these bonds, but it's gonna cost you.” — Katie Martin

  1. Martin says US debt passed $40tn and debt-service costs exceed defense spending
  2. Bond investors now demand higher yields to lend, pushing 30-year above 5.25%
  3. Long-duration Treasury exposure faces further yield-driven losses, especially if Warsh hikes
• 52:41METALongMartin notes investors view Meta's corporate bonds as competitive with Treasuries, saying 'I'm pretty comfortable with Meta as a credit risk' and buying its paper instead of government debt. — Katie Martin+5.5% since episode
  Why this trade

  “I'm pretty comfortable with Meta as a as a credit risk, I'm gonna buy the Meta bond instead.” — Katie Martin

  1. Martin says hyperscalers now borrow heavily in bond markets to fund AI capex
  2. She cites an investor choosing Meta's bond over US government debt on same-day issuance
  3. Implies Meta is perceived as a strong credit with an attractive spread, supporting sentiment
• 54:09QQQAvoidMartin warns the AI trade dominates stocks, private markets and corporate credit and is 'bending everything out of whack,' so everyone's portfolio is effectively an active tech bet. — Katie Martin+2.4% since episode
  Why this trade

  “When people talk about the AI trade being really dominant in stock markets, that is not even half the story... It's bending everything out of whack.” — Katie Martin

  1. Martin says AI exposure is now shot through equities, private credit and corporate bonds
  2. Every index investor, even EM or Korea/Taiwan/Japan, is really buying a tech fund
  3. If the AI trade falls over, 'a lot of things would go wrong at the same time'
• 55:34OPENAIAvoidMartin flags that OpenAI and Anthropic are in talks with rating agencies to get investment-grade ratings, calling the strategy 'a little bit iffy' and part of the concentration risk. — Scott Galloway+3.5% since episode
  Why this trade

  “Anthropic and OpenAI are now having conversations with credit agencies about getting an investment grade credit rating, and it sounds like their strategy is just to ask them nicely to give them a low rate... another red flag.” — Scott Galloway

  1. FT reporting: Anthropic and OpenAI talking to credit agencies for investment-grade ratings
  2. Scott asks them to give a low rate, which he sees as a red flag
  3. Martin acknowledges it looks 'iffy from the outside' and AI credit is underappreciated risk
• 55:34ANTHROPICAvoidScott calls Anthropic's and OpenAI's approach to rating agencies a 'red flag,' and the hosts flag their upcoming IPOs as a key AI-concentration risk to watch. — Scott Galloway+3.0% since episode
  Why this trade

  “Anthropic and OpenAI are now having conversations with credit agencies about getting an investment grade credit rating... it feels like this is extremely big deal.” — Scott Galloway

  1. Scott flags FT report that Anthropic/OpenAI are seeking investment-grade ratings
  2. He says it's 'iffy' and an underappreciated red flag for markets
  3. Both IPOs cited as critical events for second-half 2026 AI sentiment
• 20:21UK giltsAvoidMartin says the UK had to 'really pay up' for new debt with borrowing costs highest since 1998, and the UK is a 'shit show' that hasn't grown in ten years. — Katie Martin–
  Why this trade

  “The UK issued some government debt today... the borrowing costs on that were the highest since 1998.” — Katie Martin

  1. UK issued government debt at highest borrowing costs since 1998
  2. Martin attributes weak UK growth to Brexit's lasting damage
  3. Higher yields plus stagnant growth make gilts unattractive
• 29:19French government bonds (OATs)AvoidMartin warns France's polarized politics risks 'a real loss of investor confidence' going into next year's presidential election, though the ECB backstop limits the downside. — Katie Martin–
  Why this trade

  “There's a serious risk that when we have presidential election next year that this could could lead to a real loss of investor confidence in France.” — Katie Martin

  1. France cannot pass a budget due to a hollowed-out political middle
  2. Martin sees serious risk of loss of investor confidence in next year's election
  3. ECB support programs make an outright debt crisis unlikely but OATs remain vulnerable
• 16:40SPYLongMartin says it's 'very hard for stocks to lose at the moment because companies are just making so much damn money' and the Fed would likely cut to stop the bleeding. — Katie Martin+0.8% since episode
  Why this trade

  “It's very hard for stocks to lose at the moment because companies are just making so much damn money.” — Katie Martin

  1. US companies are making 'bucket loads of cash' with great earnings beyond AI
  2. Moral hazard: the Fed would likely cut rates if stocks really took a hit
  3. That dynamic keeps equity downside limited for now
• 24:10Italy government bonds (BTPs)LongMartin notes Italian borrowing costs are now below France's and 'the market likes what Italy's been up to' on fiscal consolidation. — Katie Martin–
  Why this trade

  “Italian borrowing costs are a little bit lower than France's... Market likes what Italy's been up to. France is your problem right now.” — Katie Martin

  1. Italy has done a good job with fiscal consolidation relative to peers
  2. Italian yields have fallen below French yields, a notable shift
  3. Relative fiscal improvement supports Italian debt
• 57:17VGKLongMartin says clients want non-tech exposure and 'the UK and Europe look pretty good from that regard as diversifiers. So they're actually doing surprisingly well in this environment.' — Katie Martin–
  Why this trade

  “The UK and Europe look pretty good from that regard as diversifiers. So they're actually doing, you know, surprisingly well in this environment.” — Katie Martin

  1. Asset-management clients want tech stripped out of international allocations
  2. UK and Europe are relatively under-owned for non-tech exposure
  3. Rising demand for ex-tech international diversification could support these markets

Prof G Markets

Anthropic Whistleblower Says AI Could “Kill Us All”

Sep 10, 2026267h

Positions · 3

• 11:41AAPLLongMcGee says the new iPhone 18 Pro will 'do just fine' and he's 'probably going to buy that,' even though he's skeptical of the $2,000 foldable Duo. — Patrick McGee+5.7% since episode
  Why this trade

  “I think the I 18 Pro, the the new version of the best iPhone will do just fine, and people will buy that. I'm probably going to buy that.” — Patrick McGee

  1. McGee dismisses the foldable Duo as a low-volume halo product with no clear use case for him.
  2. He explicitly separates it from the flagship iPhone 18 Pro, which he expects to sell well and plans to buy himself.
  3. Net read: core iPhone demand intact, so Apple stays a hold/buy despite Duo skepticism.
• 22:47OPENAIAvoidKharazian warns of extreme customer concentration: 80% of OpenAI/Anthropic enterprise revenue from 1% of businesses, and those top spenders are cutting AI spend ~10% month-over-month. — Ara Kharazian+1.2% since episode
  Why this trade

  “1% of customers driving 80% of that spend... the market is relatively underpricing.” — Ara Kharazian

  1. Ramp data shows 80% of AI lab enterprise revenue comes from 1% of customers, mostly correlated high-growth tech and AI startups.
  2. Those top 1% spenders cut per-employee spend from $8k to $7.2k MoM even as token volume rose, aided by price wars toward cheaper models.
  3. Kharazian says the market is 'underpricing' this concentration risk, a caution flag on the AI trade ahead of the OpenAI IPO.
• 23:44ANTHROPICAvoidSame Ramp data: Anthropic's enterprise revenue is dangerously concentrated in 1% of customers who are now pulling back, with spend shifting to lower-margin standard/light models. — Ara Kharazian-0.9% since episode
  Why this trade

  “you start to see a couple other negative trends... increasing share of that volume of AI spend volume by businesses going toward the standard and light models” — Ara Kharazian

  1. Kharazian: Anthropic and OpenAI derive ~80% of enterprise revenue from 1% of businesses that are highly correlated.
  2. The shift toward cheaper standard/light models (Sonnet, Terra) over frontier models pressures margins even as volume grows.
  3. He flags this as underpriced concentration risk into Anthropic's confidential IPO filing.

Ideas · 2

• 10:48AAPLAvoidMcGee doubts the $2,000 foldable iPhone Duo becomes a generational hit, comparing it unfavorably to the iPad's clear new-category positioning. — Patrick McGee+5.7% since episode
  Why this trade

  “Is the iPhone Duo trying to position itself like that? That's the sense I got from the new CEO, John Ternis, but I'm not really convinced.” — Patrick McGee

  1. McGee says the Duo is impressive hardware but he isn't 'convinced' it addresses a real need beyond an iPad with keyboard.
  2. He notes Apple may already be baking in low sales expectations, similar to the iPhone Air's weak sales.
  3. Conclusion: don't underwrite the Duo as Apple's next growth engine.
• 26:15MSFTAvoidOpenAI → you have Anthropic and OpenAI, which are contributing to, like, 70 or 80% of the AI revenues of these, big data center companies like Microsoft — Ed Elson+0.7% since episode
  Why this trade

  “you have Anthropic and OpenAI, which are contributing to, like, 70 or 80% of the AI revenues of these, big data center companies like Microsoft” — Ed Elson

  1. Kharazian documents that OpenAI/Anthropic enterprise spend is concentrated and slowing.
  2. Elson notes those labs contribute 70-80% of AI revenue for data-center companies like Microsoft, Amazon, Google.
  3. If lab spend decelerates, the mega-cap AI revenue narrative faces the same concentration risk.

Prof G Markets

Canadian Economist: Trump’s Tariffs Are A Gift To Mark Carney

Sep 9, 2026291h

Economist Mike Moffett discusses the escalating Canada-US tariff war, predicting a modest GDP hit for Canada but noting higher oil prices and Carney's popularity provide some cushion. FT's John Byrne-Murdoch analyzes the OECD's declining global education scores, linking them to AI use and rising inequality, with uncertain economic implications. Ed Elson highlights the growing US debt and interest burden as a systemic risk.

No actionable calls in this episode

Prof G Markets

OpenAI Says “AGI” Is Here — What Does That Actually Mean?

Sep 8, 2026315h

Gary Marcus argues that OpenAI's GPT-6 Astra is not true AGI but a marketing label to boost stock prices, echoing a pattern of hype that can lead to meme-like trading and eventual losses for retail investors. Catherine Ann Edwards says the strong August jobs report is overstated and the labor market is actually weak, with wage growth slipping and long-term unemployment high. No specific stock picks are made.

No actionable calls in this episode