Ed Zitron's worst enemy, an AI summarization of his anti-AI content. Get the squeeze of the Where's Your Ed At newsletter.This feed was made with Podkey. It's based on the content from the following sources:- Ed Zitron's Where's Your Ed AtCreate your own with Podkey at https://podkey.fm
Reading the feed…
A tool that helps debug a crash log can be useful. A trillion-dollar industry built around the promise that it will eventually think, work, and decide for us deserves a far harder audit. The question isn’t whether language models can do anything. It’s whether their demonstrated value matches the certainty, spending, and social pressure surrounding them. Useful tools and the boundary of judgmentOutsourcing work versus thoughtThe financial story under the hypeInference costs and productivity proofGrowth, power, and human investmentFuture tense and accountabilityThe maintenance burden Source: Ed
The core question here isn’t whether people want AI. It’s whether the biggest AI company can sell more of it without losing even more money. And if the answer is no, that problem doesn’t stay inside one startup. It leaks into cloud giants, chip vendors, lenders, and a whole lot of investor storytelling. Unsustainable economicsCompute obligations and circular demandFunding constraintsRun-rate headlines versus durable revenueAnthropic, IPO optics, and valuation logicWhat failure would actually look likeSystemic risk or overheated narrative Source: Ed Zitron's Where's Your Ed At — What Happens If
A lot of the AI story still gets told through stock charts, capex totals, and vibes. The harder question is whether there are enough real customers, at real prices, to justify what everybody is building. Revenue concentrationDemand versus capacityHow the growth story got framedCircular financing questionsWhat breaks if the labs slow down Source: Ed Zitron's Where's Your Ed At — Don't Look Up This podcast was created with Podkey. Make your own at https://podkey.fm
A lot of the AI story gets told as pure technology progress. But if the customers only exist because the vendor helped finance them into existence, you’re not just looking at demand. You’re looking at a system. The core claimWhat Jensen Huang admittedHow dependent the neoclouds lookThe financing loopTechnology versus hype economicsWhy the GE Capital comparison mattersWhat this protects and what it risksBottom line Source: Ed Zitron's Where's Your Ed At — Premium: The Hater's Guide To NVIDIA (Part 2) This podcast was created with Podkey. Make your own at https://podkey.fm
The headline here isn't just that AI is expensive. It's that one customer-partner relationship may already be large enough to shape how we read Microsoft's entire AI business. And once you look at the disclosed revenue mix next to the scale of capital spending, the obvious question is whether this is proof of strength, or a warning about concentration. What the revenue disclosure actually saysWhy concentration mattersWhat this could mean strategicallyThe capital spending numberHow to connect the two stories carefullyWhat would strengthen either case Source: Ed Zitron's Where's Your Ed At — New
A lot of the AI economy starts looking strange once you ask a boring question: who is actually paying for all this? If the answer is basically two companies, then the story shifts from explosive broad demand to concentrated financial dependency. Revenue concentrationCapex versus actual demandThe circular funding questionIs there real demand beyond the labsDisclosure and narrative framingDebt, balance sheets, and second-order riskWhy NVIDIA is part of the same storyWhat actually seems most supported Source: Ed Zitron's Where's Your Ed At — The AI Demand Bubble This podcast was created with Podk
A lot of the AI conversation is being carried by vibes, projections, and fundraising headlines. The awkward part is that some of the people closest to the data are also admitting the biggest labor-market claims still are not showing up in reality. AI job impact narrativesEconomic indices and narrative incentivesRevenue versus funding mismatchThe scale of commitmentsInfrastructure costs and profitabilityVenture capital exposureWhat actually seems supported Source: Ed Zitron's Where's Your Ed At — Premium: AI Is Getting Way Too Expensive This podcast was created with Podkey. Make your own at htt
A lot of the AI story still sounds like physics and destiny. But the pressure is showing up somewhere less glamorous: bond spreads, memory prices, lease obligations, and the fact that nobody wants to say exactly how much money this is making. Blackwell efficiency claimsThe capex mountainWhat bond markets are sayingThe memory cost spiralCircular financing and missing foundationsThe startup business model problemOpaque revenue and the hype machineWhat could break, and what might not Source: Ed Zitron's Where's Your Ed At — The More You Buy, The More You Lose This podcast was created with Podkey.
Oracle's story right now sounds simple if you only read the hype: huge AI demand, huge data centers, huge future. The harder question is whether the business underneath that story is actually getting stronger, or whether scale is masking fragility. The baseline problemFree cash flow and what changedThe AI GPU spending surgeThe OpenAI contract questionCerner and the margin mixAccounting and the story management problem Source: Ed Zitron's Where's Your Ed At — Premium: The Hater’s Guide To Oracle (Part 2) This podcast was created with Podkey. Make your own at https://podkey.fm
A lot of the AI story is being told as a technology race. But underneath it, there's a financing story, and that story matters because booms usually break where the money got weird. Why the CDO comparison existsOff-balance-sheet debt and what it hidesOverbuild versus actual demandCustomer concentration and contract cliffsNVIDIA's backstop and circular demandWho ultimately holds the riskWhy so much money flowed inHow bad could it get Source: Ed Zitron's Where's Your Ed At — The Subprime Data Center Crisis This podcast was created with Podkey. Make your own at https://podkey.fm
The core claim here is bigger than one startup maybe overspending on GPUs. It’s that OpenAI has become the story holding together a huge chain of spending, lending, construction, and stock-market optimism, and if that story breaks, a lot of supposedly separate bets may turn out to be the same bet. The load-bearing narrativeOpenAI’s spending and obligationsWho actually holds the riskOracle as the sharpest edgeMemory chips and knock-on inflationPhantom deals and media amplificationSoftBank and the funding chainAnthropic as a mirror imageWhat would actually happen in a failure Source: Ed Zitron's
A lot of AI spending gets talked about like it's abstract, inevitable, almost weightless. Then you look at memory costs, who controls supply, and how much of this buildout depends on debt, and suddenly the whole thing looks much more fragile. Why memory changes the mathConcentration and the collusion questionNVIDIA as the demand centerThe hyperscaler spending loopWhat a pullback would actually breakConsumers eventually payWhat holds up best Source: Ed Zitron's Where's Your Ed At — Premium: The Hater's Guide To The Memory Crisis This podcast was created with Podkey. Make your own at https://pod
The hard question here isn't whether generative AI is impressive. It's whether the money pouring into it reflects durable demand, or a financing loop mistaking spend for traction. The core bubble thesisWeak commercial demandGPUs, data centers, and overbuildPrivate credit and hidden exposureSoftBank and concentrated betsToo big to fail, or just very expensiveAGI narratives and fandomWhat a bust would actually hit Source: Ed Zitron's Where's Your Ed At — Let AI Burn This podcast was created with Podkey. Make your own at https://podkey.fm
SoftBank’s story sounds simple if you hear it fast enough: huge AI exposure, huge upside, visionary founder. Slow it down and the real question is whether that upside can arrive before the financing pressure does. The structure problemHow much debt really mattersWhat the market is signalingThe OpenAI wagerVision Fund track recordNAV versus the gooseMasayoshi Son's pattern Source: Ed Zitron's Where's Your Ed At — Premium: The Hater's Guide To SoftBank This podcast was created with Podkey. Make your own at https://podkey.fm
A lot of the AI story still gets told like demand is infinite and financing is somebody else's problem. The more interesting question is what happens if the spending assumptions are even a little bit wrong. The trillion-dollar premiseAre data centers being overbuiltWhat AI revenue actually meansOpenAI and Anthropic economicsOracle and SoftBank concentration riskMonetization versus infrastructure spendSystemic risk and second-order effectsThe human cost of always-on AI Source: Ed Zitron's Where's Your Ed At — The AI Industry Is Losing This podcast was created with Podkey. Make your own at https
A lot of tech right now feels like people acting out the shape of innovation without the substance underneath it. You can see the same pattern across AI branding, strange product launches, and giant partnerships that sound important before you ask what problem they're actually solving. Cargo-cult innovationAI hype and the search for a storySide bets that look like signalsPR narratives versus actual strategyExecutive detachment and product judgmentWhat actually follows from all this Source: Ed Zitron's Where's Your Ed At — Premium: Notes From The Bubble, Volume 1 This podcast was created with P
A lot of tech stories right now ask you to trust the scale before you understand the product. Billions in GPUs, endless token burn, weird hardware demos, and growth metrics that sound huge until you ask what actually got better. Cargo-cult AI spendingAI token loopsFinancial opacity and cost shiftingSaaS slowdown and monetization rotVC drift and zombie companiesProduct theater and founder modeThe hype loop in tech media Source: Ed Zitron's Where's Your Ed At — Cargo Culture This podcast was created with Podkey. Make your own at https://podkey.fm
There are two stories around AI right now. One is that the future is arriving at godlike speed. The other is that a lot of this depends on accounting, incentives, and people being too emotionally invested to ask basic questions. OpenAI financialsThe pseudo-cult problemHype, doom, and inevitabilityThe infinite-growth promiseSub-bubbles feeding each otherWhat actually holds up Source: Ed Zitron's Where's Your Ed At — Premium: The Silicon Valley Bubble (Part 2) This podcast was created with Podkey. Make your own at https://podkey.fm
A $38.5 billion loss sounds like catastrophe until you ask what kind of loss it actually was. With OpenAI's 2025 numbers, the most important question isn't whether the figure is big. It's what inside that figure reflects the business, and what reflects a one-time accounting shock. The headline numberWhy the conversion changed everythingWhat the operating business looks likeMicrosoft, SoftBank, and dependenceThe cash cushion and the real riskWhat holds up and what doesn't Source: Ed Zitron's Where's Your Ed At — Exclusive: OpenAI Losses Increased Nearly 8X in 2025, With Spending Hitting $34 Bil
One story here looks like safety theater colliding with geopolitics. The other looks like a business model colliding with arithmetic. Anthropic breach and export controlsTokenomics and the pricing crisisCorporate cost discipline arrivesOpen source alternatives and market structureSilicon Valley monoculture Source: Ed Zitron's Where's Your Ed At — AI's Brokenomics This podcast was created with Podkey. Make your own at https://podkey.fm
An OpenAI IPO sounds like a simple milestone story until you put the financing next to the spending. Then it stops looking like a normal growth company and starts looking like a test of how much strain the market will absorb for AI. OpenAI and public-market pressureAnthropic's financing structureRecursive self-improvement as an explanationThe AI bubble and the Rot-Com theoryWhy venture capital changedWhat actually holds up Source: Ed Zitron's Where's Your Ed At — Premium: The Silicon Valley Bubble (Part 1) This podcast was created with Podkey. Make your own at https://podkey.fm
The core question here isn't whether AI is useful. It's whether the industry's spending plans make any economic sense at the scale being promised. And once you look at the debt, the token billing, and the revenue assumptions together, the stress points start to line up. Compute commitments versus plausible revenueData center debt and the financing bottleneckToken billing, visibility, and vendor incentivesInternal token burn and ROI scrutinyWorkforce strain and the human layerThe coming exposé and what to watch Source: Ed Zitron's Where's Your Ed At — AI Is Slowing Down This podcast was created
The loudest AI story says inevitability: massive demand, massive disruption, massive value. The quieter story is about losses, subsidies, weird billing, concentrated customers, and a lot of people treating demos like proof. Manufactured hypeFinancial realityToken billing and enterprise realityHallucinations and task limitsCompute overbuild and NVIDIA exposureJob displacement claims Source: Ed Zitron's Where's Your Ed At — Premium: The Hater's Guide To The AI Bubble 3.0 This podcast was created with Podkey. Make your own at https://podkey.fm
A lot of the AI argument right now comes down to a weird mismatch. The public story is abundance, efficiency, and inevitability. The financial story looks a lot murkier once you ask who pays, what gets measured, and whether any of it actually holds up without subsidies. ROI and the accountability problemToken billing reveals the real priceWhy subscriptions look fragileCompute costs are not falling the way the story saysIs this worse than Dot ComDark output and the job replacement storyWhat leadership still hasn't answered Source: Ed Zitron's Where's Your Ed At — AI Doesn't Have ROI This podcas
A trillion dollars in GPUs sounds like inevitability until you ask who is actually using them, where they’re going, and how any of this gets paid back. The interesting question isn’t whether AI is real. It’s whether the current scale story survives contact with construction timelines, customer concentration, and basic unit economics. NVIDIA demand versus installed realityConstruction bottlenecks and the backlog problemRevenue concentration in AICash burn and the path to profitabilityToken billing and ROI opacityCredit markets and supply-chain exposureWhat holds up and what doesn't Source: Ed Z
