🦔A developer found that loading the AliExpress homepage triggers hidden scripts that use your browser's audio system to fingerprint your computer. The scripts generate a silent waveform, run it through your audio hardware, and measure the output to create a unique identifier for your machine. No sound plays, no permission is asked, and it runs even when the page is sitting idle. The scripts were identified as part of Alibaba's browser security tooling. Firefox lets you disable it. Brave blocks it by default. Most people have no idea it's happening. My Take A guy found this because his Bluetooth headphones started acting weird when he opened AliExpress. He dug into it, found hidden scripts running silent audio through his browser to fingerprint his machine, and realized the site had been doing it on every page load without any indication. If his headphones hadn't glitched he never would have looked, and no regulator or audit caught it either. The EFF has a free tool called Cover Your Tracks that shows how trackable your browser is, and based on everything going on right now with data collection I'd run it. Hedgie🤗

🦔OpenAI's head of data centers, Chris Malone, left the company last week. Malone joined in March 2025 right after the Stargate venture was announced and oversaw the company's data center buildout. He's the fourth senior executive to leave ahead of the planned IPO, after the CRO, COO, and product chief. OpenAI has projected $750 billion in compute spending through 2030 and just signed a 10-gigawatt data center lease in Ohio backed by $105 billion in Nvidia guarantees. My Take The person in charge of building the data centers left a company that just committed to $750 billion in compute spending. The CRO who was supposed to sell the product left. The COO who was supposed to run the business left. The product chief left. And the CFO held an all-hands last week telling everyone not to worry. Four senior departures in a few months, all before an IPO, at a company valued at $852 billion. OpenAI President Greg Brockman said this level of turnover "isn't that atypical." Losing your revenue chief, your operations chief, your product lead, and your data center head in the same summer before an IPO is very atypical. These are the people who knew how the business worked from the inside, and they chose to leave before the payout. Whatever they saw, staying wasn't worth the money. Hedgie🤗

🦔Harvard Business School is now selling AI clones of its professors for $699. Seven faculty members were cloned into AI avatars for an online startup bootcamp where founders practice investor pitches and board meetings with digital versions of Harvard professors. 760 founders have gone through it so far. It's not a degree and carries no academic credit. A year of actual Harvard Business School costs over $84,000. My Take Harvard has a $50 billion endowment and charges $84,000 a year for business school. Now they're selling AI versions of those same professors for $699 to people who can't afford the real thing. This isn't about democratizing education. This is a $50 billion institution figuring out how to monetize its brand name at scale without adding a single seat in the classroom. The professor who got cloned said the AI isn't meant to replace him. But if an AI version of you is good enough to charge $699 for, the follow-up question from every dean and CFO in higher education is how many of you they actually need. We recently covered a financier saying his AI-native interns produced shallow work because they leaned on AI through college. Harvard is now selling AI as the college experience for founders who can't get in the door. Universities are going to use AI the same way every other industry has, cut costs, maintain prices, and let the people at the bottom absorb the difference. Hedgie🤗

🦔Anthropic is expected to tell IPO investors that its total addressable market exceeds $30 trillion. That's roughly the entire GDP of the United States. TAM is the revenue a company could theoretically earn if it captured 100% of its market, and Anthropic is defining that market as all work that could be completed with AI models. The company lost $42 billion last year, projects $190 to $200 billion in revenue by 2028, and is seeking a $2 trillion valuation. SpaceX claimed a $28.5 trillion TAM in its IPO filing. Anthropic is one-upping that. My Take Every company inflates its TAM in an IPO filing, that's standard. But $30 trillion is the entire US economy. Anthropic is telling investors that its addressable market is every task a human does that could theoretically involve AI. By that logic my addressable market is every atom in the universe because I could theoretically touch all of them. TAM is supposed to help investors size an opportunity. At $30 trillion it's just a number on a slide. A year ago these companies were pitching billions. Then it was hundreds of billions. Now it's $30 trillion, which tops SpaceX's $28.5 trillion claim from May. The numbers escalate because they have to. Each round of fundraising needs a bigger story than the last one, and the story has to be big enough to make a $2 trillion valuation on a company that lost $42 billion last year seem like a bargain. I think most of the investors in the room know the $30 trillion number is decorative. The problem is the retail investors who buy the IPO after them, who might not. Hedgie🤗

🦔The cost of insuring Broadcom's debt just spiked faster than any other AI company. Broadcom is negotiating up to $100 billion in off-balance-sheet debt through special purpose vehicles to finance AI chips for Anthropic, on top of $35 billion it already backstopped in June. The debt doesn't show up on Broadcom's balance sheet but Broadcom guarantees a portion of it. JPMorgan warned that off-balance-sheet AI commitments across the industry are heading into the trillions and called it "phantom leverage." My Take Nvidia guarantees data center leases. Broadcom guarantees chip financing. Both use SPVs to keep the debt off their books so it doesn't show up in the numbers investors normally look at. The bond market is pricing in what the balance sheet doesn't show, which is why Broadcom's credit default swaps spiked 28 basis points in August while the company's earnings still look fine on paper. The phantom leverage is piling up fast. Leases, purchase commitments, residual value guarantees, chip financing backstops, all off the books, all contingent on AI demand staying hot. The SEC decided this month that data center securitization falls outside Dodd-Frank risk retention rules, so the safeguards built after 2008 don't apply here. Apollo and Blackstone show up in every one of these deals, arranging the debt, earning the fees, and passing the risk to the bondholders. If Anthropic or any major lessee stumbles, these guarantees come back onto the chipmakers' books, and the CDS market is already moving on that possibility. Hedgie🤗

🦔74% of newly created web pages now contain AI-generated content. Over half of published articles are AI-written or AI-assisted. Automated traffic passed human traffic for the first time last year, hitting 51% of all web activity. Snap banned AI-generated videos from recommendations and human creator numbers jumped 120% immediately. LinkedIn added an AI slop reporting button the same week it kept sending notifications nudging users to write posts with its AI assistant. Apple TV started labeling content "Made by Humans" in credits. Brands that went all-in on AI-generated ads are paying premiums to go back to human creators because AI ads stopped converting. My Take Every technology that makes something abundant turns the scarce version into a premium. Printing press made text cheap, handwritten manuscripts became collectibles. Mass manufacturing made goods cheap, handmade became luxury. AI made content free and infinite, and now human-made is becoming the premium category. Brands are already paying for it. A company called Icon charges $1,000 for six ads filmed by a person, and the demand is growing because AI ads stopped converting. Consumers can feel when something is manufactured even if they can't explain why, and they stop engaging. The ad industry is the canary here because it measures everything. Click-through rates, engagement, conversion, all tracked in real time. The entire pitch behind AI content was that it would be cheaper and just as good. Cheaper, yes. Just as good, no. Advertisers tested that at full scale and reversed course fast. If the one industry with the best data on whether AI output works is already paying premiums to go back to humans, the rest of the economy should be paying attention to why. Hedgie🤗

🦔55% of companies that laid off workers citing AI now regret it, according to Forrester. Two-thirds are already rehiring. A survey of 600 HR leaders found over a third had rehired more than half the roles they cut, often within six months and at higher cost. Klarna bragged about replacing 700 customer service workers with AI, then quietly started hiring humans again when quality collapsed. Ford is rehiring engineers to fix problems automated systems couldn't handle. Forrester projects AI automates about 6% of jobs by 2030. Six percent. My Take Companies fired people, replaced them with AI, watched things break, and are now paying more to bring people back than they saved by cutting them. I think it exposes how poorly most executives understood what their own employees actually did. The work that's easy to see, writing code, answering tickets, processing claims, looked like the expensive part. The work that's hard to see, the institutional knowledge, the judgment calls, the context about why things are done a certain way, turned out to be what held everything together. This reminds me of the offshoring wave in the 2000s. Companies moved everything they could overseas, discovered the hidden costs a year later, and quietly brought it back under a different name. Same impulse, same blindness to what the people they cut were actually doing all day. The difference is that offshoring at least put humans on the other end. This time they replaced humans with a tool that hallucinates, can't handle edge cases, and has no memory of why the last three attempts at solving this problem failed. Forrester says 6% by 2030. Over half the companies that swung bigger than that already regret it. Hedgie🤗

🦔Disney posted record revenue this quarter off Toy Story 5 and Spider-Man: Brand New Day. Former CEO Bob Iger left to buy the LA Lakers. New CEO Josh D'Amaro's first move was notifying 200,000+ employees that starting in 2027, their spouses can no longer be on the company health plan if the spouse has any job that offers benefits, even if those benefits are worse. Disney called it part of their "Total Rewards" loyalty program. My Take Disney had one of its best revenue years ever and chose to use the moment to cut benefits for its workforce. Workers fall below buybacks, below executive comp, and below a new cruise ship on the priority list. Cutting benefits during a record year saves money on this quarter's earnings call and costs you in turnover once your best people start looking elsewhere. Those 200,000 households just got hit with higher healthcare costs on top of $4 gas, rising grocery prices, and credit card rates near record highs. Multiply Disney by every other company doing the same thing, and you start to understand why Walmart's same-store sales missed for the first time in five years. Companies keep reporting record profits and cutting worker costs at the same time, and those cuts come out of the same consumers the economy depends on. Hedgie🤗

🦔AI companies have pre-ordered almost all of the world's RAM production for 2027. Prices have climbed roughly 500% in 12 months, with a 64GB DDR5 kit jumping from $191 to over $1,100 and a 128GB kit going from $329 to $3,399. DRAM chips are now worth over half as much per kilogram as gold. PC and smartphone manufacturers are competing for whatever is left over, and SK Hynix's CEO warned that 2027 will be the worst year for memory supply in the industry's history. My Take The more expensive your device gets, the harder it is to own one powerful enough to do anything locally, and the more you depend on cloud services and subscriptions for everything. Gaming, storage, software, computing, all rented instead of owned. The companies buying up all the RAM are the same ones who'd love to sell you a subscription instead. The more your next laptop costs, the easier it is to talk you into renting everything from the cloud. The AI shortage didn't create that play, but it sure sped it up. Three companies control about 90% of global DRAM, Samsung, SK Hynix, and Micron. Samsung and Hynix have criminal convictions for fixing RAM prices in the early 2000s, paid about $485 million in fines between them, and executives did prison time. The industry-wide penalties topped $730 million. Micron was part of it but ratted everyone out and walked. Now a new class action alleges the same three used the AI shift as cover to cut consumer memory production and inflate prices. The case is unproven, but when companies with that track record are making two to three times the margin selling to AI data centers while you pay 500% more for the same product, "just supply and demand" is a tough sell. Hedgie🤗

🦔If you want to see what the future of American healthcare looks like, read this. A venture capital firm bought one of the biggest hospitals in Akron, Ohio, converted it from nonprofit to for-profit, and is now using it as a "test kitchen" for AI products made by companies that same firm invested in. General Catalyst paid $515 million for Summa Health, the county's largest employer with 8,500 workers, and just announced nine AI partnerships covering diagnostics, documentation, patient calls, claims processing, and a command center linking them all together. Every one of those companies is backed by General Catalyst. Workers lost eligibility for federal student loan forgiveness in the conversion. A doctor sued over it and the hospital called loan forgiveness "a private matter." My Take Most hospitals adopt AI because a vendor pitched them and the cost savings looked good on a slide. This hospital was purchased so a VC firm could test its own investments on a captive patient population, and that changes the incentive structure completely. The people deciding which tools to deploy profit when those tools get adopted elsewhere, so the incentive is to prove the portfolio works, not to improve patient care. Some of these tools are already live, like AI phone calls for surgery prep. Others are rolling out now. There's no federal law requiring hospitals to disclose AI use in clinical care, and Ohio doesn't have one either, so most patients walking into Summa won't know if AI flagged their scan, wrote their visit notes, or processed their claim. Nine companies, all backed by the same investor, handling patient data through one integrated command center, with no public disclosure about how that data moves between them. General Catalyst plans to roll whatever they learn at Summa out across more than two dozen partner health systems. Akron is the pilot, and the patients are the test group whether they signed up for it or not. Hedgie🤗

🦔Delta's CEO said AI will boost profits by 50% by setting a different ticket price for every passenger in real time. The airline uses an AI system called Fetcherr that generates a unique price for each shopping request. Delta told Congress last year it was not using personal data for individualized pricing. Its investor presentation described building toward offers for "you, the individual." Airlines are exempt from FTC pricing oversight. My Take This is what AI does for a company like Delta. It doesn't make your flight better or your bag show up faster. It watches how you shop and figures out the most you'll pay before you close the tab. Consumer Reports found Uber and Lyft charging 42% different prices for the same ride at the same time, and the House Oversight Committee found one case where two people got quoted $76 and $24 for the same trip. Delta is building the airline version of that, and every carrier will follow because the economics are too good to pass up. Walmart told us this week that families are choosing between gas and groceries at $4 a gallon. Delta's answer to those same families is an algorithm that squeezes them for every dollar they have left. Airlines are exempt from FTC pricing rules, state laws are preempted by federal deregulation, and the CEO told Congress one thing while telling investors another. When people ask me where the trillions going into AI actually end up, I think Delta just gave us one of the most honest answers we've gotten. Hedgie🤗

🦔Five ECB economists warned that an AI market correction is likely whether current valuations are justified or not. They studied every major tech revolution and found the same boom-then-correction pattern each time, even when the technology was transformative. European households hold about €440 billion in Mag 7 stocks through index funds and pensions. The ECB said policymakers have less room to cushion a correction now than during dot-com. Their line: a US AI correction would not remain a US problem. My Take The ECB isn't saying AI is fake. They're saying the technology can work and the stocks can still crash, because that's what happened with railroads, electricity, and the internet. The tech survived every time. The investors who bought at the top didn't. Lucent made the equipment that built the internet, dominated the market, and lost 98% of its value when the financing behind the buildout collapsed. Nvidia sells the chips that build AI and is now guaranteeing the leases and financing the purchases of its own customers. The technology was legitimate both times. The financial structure around it was not. €440 billion in European retirement savings is exposed to Mag 7 stocks through index funds, and the same concentration exists in US 401k target-date funds. If you hold an S&P 500 index fund, roughly 30% of it is in a handful of AI companies. The ECB flagged that the tools to respond to a correction are weaker now than in 2000. Rates are high, deficits are blown out, and the fiscal room to cushion anything has shrunk. That means if the correction comes, the people holding these stocks through their retirement accounts absorb more of the loss than they did last time, with less help on the other side. Hedgie🤗

🦔ChatGPT can now read your entire iMessage history and send texts on your behalf. Every conversation with your family, your doctor, your lawyer, your ex, on OpenAI’s servers so it can reply to your mom for you. Altman told us a few days ago he wanted ChatGPT to hold your entire life. He meant it. This is a company that hasn’t turned an annual profit, is heading for an IPO, and just asked for access to the most personal data on your computer. But sure, let it answer your group chat. Hedgie🤗

🦔Yesterday the Treasury doubled bond buybacks and yields dropped. Today yields are back above where they were before the announcement. The 30-year hit 5.25%, the 10-year reached 4.71%. The national debt officially crossed $40 trillion yesterday. WTI crude pushed toward $89 after Trump threatened economic warfare on Iran. The buyback bought about 18 hours of relief. My Take I compared the buyback to the yen intervention and expected a few weeks of breathing room. It lasted less than a day, which means the bond market took this even less seriously than the currency market took the yen defense. $4 billion in buybacks against $40 trillion in debt and $739 billion in quarterly borrowing wasn't enough to hold yields down overnight. Bessent went on CNBC today promising spending cuts are coming. The bond market has heard that from every Treasury Secretary for the last 20 years. Meanwhile oil is heading toward $89 because the Iran war has no end in sight, the July deficit was $432 billion, the highest monthly total since March 2021, and interest payments on the debt have hit $1.2 trillion this fiscal year. Bessent is trying to manage yields with a tool that amounts to less than 1% of quarterly borrowing, and yields are already back where they started. Hedgie🤗

🦔Anthropic expects its IPO to match or beat SpaceX's record $75 billion raise and could file publicly within weeks. Q2 revenue was $11.5 billion. The 2025 net loss was nearly $42 billion. Investors are floating a $2 trillion valuation based on projections of $190 to $200 billion in revenue by 2028. Dario Amodei wants super-voting shares with about 2% ownership. My Take $42 billion in losses on $18 billion in revenue and the ask is $2 trillion. Revenue has never been the problem for these companies. They can grow the top line fast. They can't stop the spending from growing faster. Anthropic's own CFO told investors the one profitable quarter may not repeat because the compute commitments are that heavy. The SpaceX contract alone runs tens of billions over three years, and Anthropic is trying to match the IPO size of the company it pays for compute. Think about that for a second. Dario wants super-voting control with 2% equity. Public investors put up the capital, he makes the calls. Meanwhile Nvidia guarantees $105 billion of data centers running these models. Pension funds hold the bonds behind the buildout. DeepSeek undercuts the pricing every few months. And Anthropic's own risk report says its models bypass safeguards and deceive operators. I've covered every one of these threads this year and they all converge on the same question: does the $2 trillion price account for any of it? I don't think it does. I think it prices a future where the revenue keeps compounding and none of the risks show up, and that's a very expensive bet to be wrong on. Hedgie🤗

🦔Walmart beat on revenue and earnings but the stock dropped 9% today. Same-store sales grew 2.6% versus the 3.7% Wall Street expected, the slowest pace in over six years and the first miss in at least five years. Shoppers visited more often but spent less each trip. The company expects $2 billion in extra fuel costs this year and cut prices on thousands of items to keep people coming in. Q3 guidance came in below estimates. My Take Walmart is one of my favorite real-time looks of the American consumer, and the consumer looks worn down. People are visiting more often but buying less because they're stretching their dollars further than they were even a quarter ago. Higher-income families who used to shop at Target and Whole Foods are showing up at Walmart now, and while that helps Walmart's numbers it means the stores above them are losing customers they won't get back easily. The CFO pointed to $4 gas as the line where spending habits change, and I think he's right. Gas at $4, diesel at record levels, grocery prices still elevated after two years of increases, and credit card debt above a trillion dollars. Families are making choices they weren't making six months ago. Retail sales already fell 0.6% in July, and now the biggest retailer in the country is guiding lower for next quarter. I think we're watching the consumer run out of room in real time, and Walmart's numbers today are the most clear picture of it we've gotten. Hedgie🤗

🦔OpenAI's CFO held an all-hands meeting today to tell employees the company will go public by 2027 and not to worry about Anthropic beating them to it. This came days after the CRO, COO, and product chief all left. OpenAI just reported $6.7 billion in Q2 revenue. Anthropic reported $11.5 billion for the same quarter. OpenAI is valued at $852 billion. My Take That all-hands happened because people inside the building are spooked. You don't pull the whole company together to talk about IPO timelines and tell everyone to ignore the competitor unless morale took a hit from three executives walking out in one month. The CFO called the IPO just another fundraise, which is an odd thing to say to a room full of people whose compensation depends on the stock being worth something. Both companies dropped revenue numbers within days of each other, OpenAI at $6.7 billion for Q2 and Anthropic at $11.5 billion, right as both prepare to go public. Neither number has been audited, neither company has reported an annual profit, and both have every incentive to put the best figure possible in front of investors before the roadshow starts. A lot of regular investors are going to buy both on name recognition without digging into how those revenue numbers were built or whether the companies can ever earn more than they spend, and I think both companies are counting on exactly that. Hedgie🤗

🦔The Fed released its July meeting minutes this afternoon. The vote to hold rates was 9-3, with three regional presidents voting to hike. Many members said they'd need to raise rates if inflation doesn't come down, and some thought current policy isn't restrictive enough. CPI is still at 3.4%. This dropped the same day the Treasury doubled its bond buybacks to bring yields down. My Take The Treasury spent this morning trying to push yields down. The Fed spent this afternoon telling you they might push them up. That contradiction on the same day is the clearest sign yet that nobody in Washington has a plan, they're improvising. Bessent wants lower long-term rates because the deficit is eating the government alive at 5.3%. Warsh can't cut because inflation is still nearly double the target. Three of his own members wanted to hike in July and said waiting would only force bigger hikes later. I think the Fed hikes before the year is out, probably December, and the Treasury's buyback this morning was Bessent trying to build a cushion before it happens. The bond market has been saying this for weeks, long-term yields climbing even as economic data softens, and the minutes confirmed what the market already priced. Rates aren't coming down anytime soon, and anyone still making financial decisions based on cheap money coming back needs to adjust. Hedgie🤗

🦔I wanted to share a compilation that I saw of residents recording what data centers sound like from their neighborhoods. The hum runs 24 hours a day, 7 days a week. I've covered the community side of the buildout before with the Gilroy and Pecos County projects, but hearing it is different than reading about it. My Take The people in these videos didn't sign up for a 24-hour industrial hum so that AI companies can train their next model. They weren't asked, and in a lot of cases they weren't even told until construction started. The buildout gets measured in gigawatts and billions of dollars on earnings calls. On the ground it gets measured in sleep you don't get and a home you can't sell. I've posted a lot of numbers this year about the AI boom. This is what those numbers sound like. Hedgie🤗

🦔The Treasury just announced it's doubling its buybacks of long-term government bonds, from $2 billion to at least $4 billion per operation, starting September 9. Yields dropped immediately, with the 30-year falling 9 basis points to 5.19% and the 10-year dropping 6 basis points to 4.65%. Stocks rose. But the buybacks aren't a debt paydown. The Treasury pays for them by issuing more short-term bills, so it's shifting the debt from long-term to short-term, not reducing it. My Take Three weeks ago the US and Japan spent billions buying yen to prop it up. The yen gave back most of the gains in under three weeks. Now the US Treasury is intervening in its own bond market, and I think it ends the same way. Buying back long bonds pushes yields down today, but it doesn't shrink the deficit or slow the borrowing. All it does is move the debt from 30-year bonds into short-term bills that roll over more often at whatever rate exists when they come due. The market rallied because it likes seeing Bessent willing to act, and I understand the relief. But the reasons yields climbed to 19-year highs last week are all still in place, and a $4 billion buyback doesn't change any of them. Governments keep reaching for the same tool, spend now and hope something changes before it wears off. The yen intervention bought about three weeks and I'd be surprised if this one buys much more. Hedgie🤗