On August 3, Jim Cramer told CNBC viewers he's selling every last one of his Bitcoin. His reasoning: quantum computers will crack it within about three years.

He'd interviewed IBM CEO Arvind Krishna a few days before that, who said investors should be "paranoid" about quantum computers breaking modern cryptography in 3-4 years.

Obviously, Jim Cramer's track record tells us he's not someone we should be taking all that seriously (Inverse Cramer exists for a reason)… That said, enough PRO members asked about the story that John Gillen (our crypto analyst) decided to write a tidy little breakdown on 'the quantum threat.' His advice, in short: pay attention to it, but don't freak out about it.

The actual threat looks like this…

Your wallet has a private key (your secret password), and that key generates your public key (the identifier everyone else can see). On a normal computer, following the breadcrumbs to reverse engineer a user's private key from their public key would take billions of years. But a big enough quantum computer running something called Shor's algorithm could do that reversal in minutes.

Nowadays, your public key only touches the Bitcoin blockchain when you send funds out - so if you were to move all of your funds to a fresh wallet with no outgoing transfer history, you'd be safe.

But this wasn't always the case.

Coins sitting in 2009-2010 wallet formats are exposed, because those formats wrote their public keys straight to the chain. When you add the BTC held in 2009-2010 era wallets to the BTC held in wallets that have sent Bitcoin in the past, it adds up to about a third of every Bitcoin ever mined. Roughly 1.1M of those older coins are Satoshi's, and none of those keys can ever be un-exposed.

The simplest solution to protect against quantum: send your coins to a fresh address you've never spent from, and then leave them alone. Though as John puts it, anyone with a quantum computer powerful enough to hack Bitcoin would have to walk straight past Satoshi's million coins to come after yours (unlikely).

… so when does a computer like this actually turn up? That estimate has moved twice this year.

Public miners have unloaded 28,000 BTC this year — down from 127,000 to 99,000 in holdings — injecting $1.78 billion in selling pressure that has gone largely undiscussed while ETF outflows and treasury company sales have dominated the narrative. Price is set at the margin, and steady selling into weak demand hits harder than the headline numbers suggest. The irony: many large miners are pivoting to AI and leaving the network, which has eased difficulty 18% from November's peak, meaning the miners that stayed are now earning 18% more BTC per block — a classic free-market reset that could attract new entrants back in.

Goldman paid up to $2.25 billion to acquire NEOS Investments — rather than launch the covered-call bitcoin ETF it had already filed for in April — instantly gaining a $1.1 billion fund (BTCI) yielding 27% and a $30 billion options-based ETF platform across 19 funds. The deal lifts Goldman's total ETF assets above $130 billion and puts it in direct competition with BlackRock's BITA, which launched two months before Goldman's own filing and had a head start Goldman clearly decided it was cheaper to leapfrog than race.

Kalshi is adding Solana-based DoubleZero's low-latency data infrastructure to its prediction market order book — giving institutional traders a machine-readable view of Kalshi's most active contracts over dedicated fiber, simultaneous to all connected firms. The same distribution model underpins NYSE, Nasdaq, and the CME. It's a signal that prediction markets are being built for the same institutional audience that demands co-location and microsecond data edges in traditional finance, not just for retail speculators.

Inflation was benign in July, a second month of relief for American consumers and economic policymakers contending with stubborn price pressures.

Driving the news: The Consumer Price Index rose just 0.1% in July after falling 0.4% in June, while the year-over-year inflation rate edged down to 3.4%, from 3.5%.

• Core CPI, which excludes food and energy, advanced 0.2% after a flat reading the prior month and was up 2.5% from a year earlier, down from 2.6%.
• On a three-month annualized basis, core inflation was running at roughly 1.6%, down from 2.4% in June.

Why it matters: The tame reading might give the Fed some breathing room to hold off raising interest rates in September. Still, the Iran war and the whiplash in energy prices it has unleashed leave lingering questions about the staying power of the inflation relief.

Between the lines: The report shows the war-induced energy shock continuing to fade. Energy prices declined 1.5% in July, with a nearly 3% drop in gasoline prices.

• But that masks a sharp reversal as July progressed. The average price of regular gasoline climbed to $4.10 a gallon by the end of the month — up from $3.78 just weeks earlier.
• Gas prices have retreated some this month, though oil prices have spiked higher again, with the U.S. benchmark, West Texas Intermediate crude oil, back above $80 a barrel as renewed Middle East fighting raises fresh concerns about global energy supplies.

When investment on the scale of the current AI boom occurs, it inevitably has to come at the expense of something. All the resources devoted to building data centers and developing AI models would otherwise go to something else.

The big picture: This crowding out is smaller than you might expect, Goldman Sachs economists find in a new note. But it does exist, they say, and takes the form of displacing other tech investment and construction, as well as raising corporate borrowing costs.

By the numbers: AI investment will be about $600 billion this year, some 2% of GDP, accounting for 10% of business fixed investment and 15% of equipment investment.

State of play: The first crowding-out channel identified is the displacement of other tech spending at the hyperscalers themselves and at the companies that spend on AI services. Corporate IT budgets facing new, big costs for AI tokens may seek to cut back on other software and tech spending. That doesn't have much impact on overall GDP, however, as it amounts to shifting spending around.

Zoom out: The Goldman team also sees the data center boom crowding out other building activity, as construction labor and equipment is devoted to the AI buildout. Gross margins on data center construction are more than twice as high as margins on non-tech projects, which has resulted in data centers pulling resources away from other projects.

Zoom in: The hyperscalers' bottomless demand for capital has created a surge in AI-related debt issuance. The rest of the corporate sector faces higher borrowing costs as a result — which may hem in their own investment. But the Goldman team finds that this impact has been limited so far, raising corporate borrowing costs by only 0.05 percentage point and perhaps reducing non-AI investment by a modest $10 billion.

On August 10, Nvidia announced partnerships with six major Wall Street firms: Apollo, BlackRock, Blackstone, Brookfield, Goldman Sachs, and KKR.

Together, they're looking to mobilize up to $500B for AI infrastructure.

For reference, AWS, Microsoft Azure, Google, Meta, and Oracle are expected to spend roughly $750B on CapEx in 2026.

That's a huge new pool of capital entering the AI infrastructure race.

Both CoreWeave and Nebius crushed earnings in a span of 24 hours.

CoreWeave's backlog went from $25.9B in Q1 2025 to $104.2B in Q2 2026 which is roughly a fourfold increase in five quarters.

While Nebius delivered strong results as well showing just how quickly demand for AI compute is expanding.

Neocloud companies like Nebius and CoreWeave need enormous amounts of capital to buy GPUs and build data centers. Historically, that meant having a massive balance sheet.

But now, institutional financing can help bridge that gap.

In simple terms: Wall Street provides the capital → neoclouds build the infrastructure → Nvidia provides the GPUs → AI companies rent the compute.

The ability to build AI infrastructure is no longer limited to companies with hyperscaler-sized balance sheets.

Spotify announced it will explicitly label AI accounts, excluding them from the company's playlists and algorithmic recommendations.

Anthropic said it will mark text and images generated using its Claude large language models with readable metadata that's invisible to the human eye. This will allow people to more easily verify if something is AI-generated. In the case of written responses, Claude will "weave an imperceptible watermark directly into the text itself" that won't impact its readability or meaning.

On one side: OpenAI and Anthropic, championing a closed-model ecosystem in the name of safety. On the other side: just about everyone else in the tech world, including Mark Zuckerberg's Meta, and perhaps most importantly, Nvidia. In March, Nvidia CEO Jensen Huang spent much of Nvidia's developers conference positioning Nvidia's chip technology as foundational to the open-source AI movement.

XRP briefly touched 99 cents on Tuesday — the first time it has been below $1 since November 2024, when Donald Trump won the presidential election and sent the whole crypto market surging. It quickly bounced back to just above $1, but the bounce stalled. Meanwhile open interest in XRP futures has climbed to its highest level since October, meaning a lot of leveraged bets are stacked up right at a price that, psychologically and technically, matters enormously.

The U.S. inflation report lands this morning — a hot reading could push XRP and the broader market through the floor; a soft one could be the catalyst crypto has been waiting for all summer.

A hacker drained $200,000 from an XRP bridge by depositing nothing and withdrawing real tokens — the software thought the fake deposits were real.

Fidelity just filed to add staking and quarterly cash payouts to its nearly $900 million ether ETF, a move that turns a passive index fund into something that more closely resembles a dividend-paying asset.

The long-awaited US inflation print landed at an exact 3.4% headline year-over-year. With Core CPI matching consensus at 2.5%, Kevin Warsh's data-dependent Fed gets a clean disinflation reading. This in-line print takes immediate September rate hike pressure off the table.

Elevated crude oil prices and Middle East geopolitical risks continue to linger, but risk assets are breathing a sigh of relief. Bitcoin remains tightly coiled near $64,000 while traditional markets digest the news. Macro friction is simply failing to push digital asset prices lower.

Crypto market analysis shows institutional capital quietly absorbing supply while long-term holders stack through the noise. On-chain data confirms liquidity is quietly building underneath, laying the groundwork for a rotation into high-throughput settlement rails.

NASA has debunked a viral conspiracy theory that the Earth will lose gravity for seven seconds today.

Europe will have its first total solar eclipse since 1999.

A key inflation reading comes out this morning.

Stocks fell yesterday as oil prices climbed due to doubts that a deal with Iran will materialize. Big Tech in particular had a bad day, with Alphabet, Amazon, and Apple all dipping.

Claude is watermarking the content it generates.

Michigan has implemented an unusual new grading policy.

Nvidia has lined up six Wall Street giants to create more than $500 billion in "dedicated pools of capital," presumably long-term institutional investors like insurers and pension funds, to finance projects for Nvidia's customers.

In some cases, Nvidia said it "may provide a residual-value support mechanism for up to 25% of an opportunity, assessed carefully on a project-by-project basis." In other words, the chipmaker could provide some type of guarantee for some projects, potentially cutting interest rates for borrowers, and helping to keep the boom going.

U.S. stock futures are up this morning after AI "neocloud" company CoreWeave's stock surged in after-hours trading on the back of a positive earnings report.

Norway's sovereign wealth fund — the largest in the world — reported record half-year profits, crediting a rally in Asian tech stocks with its growth.

The exchange operator CME Group said yesterday that it plans to launch two compute futures contracts on Oct. 5, "pending regulatory review." The group is working with Silicon Data, a company that publishes indexes tracking compute pricing.

Each futures contract would represent a month's worth of rent for the Nvidia H100 or its newer Nvidia Blackwell B200. The idea is to turn compute into a "standardized, tradable commodity," Pete Keavey, global head of energy and environmental products at CME Group, said in the release.

Another company, New York-based OneChronos, is also working on a compute futures marketplace and is awaiting approval from federal regulators. The company expects it to be operational this year, CEO Kelly Littlepage tells Axios. OneChronos is using "combinatorial" auctions to address the fact that compute is not fungible, storable, or transportable like traditional commodities, working with economist Paul Milgrom who won a Nobel Prize for developing a similar system to auction wireless spectrum.

The whole financial universe is holding its breath as we count down to today's critical inflation print. Bitcoin is sliding toward the $64,000 level in very thin trading. The market is completely frozen by macroeconomic anxiety. Traders are waiting for the latest Consumer Price Index data before putting new money to work.

Broader equities remain fragile with the S&P 500 still trying to digest the recent volatility. Crude oil futures hover near $79 on continued geopolitical risk. The crypto market has been in its most constricted compression for years.

Anthropic overtook OpenAI in April in terms of revenue, reaching $30 billion ARR.

With food-safety concerns swirling, CAVA CEO Brett Schulman says the industry must hold suppliers to high standards and adhere to rock-solid safety protocols. Recent outbreaks of salmonella and cyclospora have sickened thousands of Americans, calling attention to the vulnerability of the food supply chain. Schulman acknowledged that "there's just broader concerns around leafy greens" that temporarily dampened CAVA sales in recent weeks — but he said traffic is bouncing back and that the Mediterranean fast-casual chain's food has not been affected. In its second quarter — which ended around the time the cyclospora outbreak began in July — CAVA showed significant growth, delivering a 9% same-store sales increase (higher than the 7.2% growth expected by S&P Global Capital IQ). Customer traffic at comparable restaurants grew 5.3%. Schulman confirmed that CAVA sources certain products from Taylor Farms — the supplier of the since-recalled iceberg lettuce linked to the cyclospora outbreak — but said "we do not source any ingredients from Taylor Farms-impacted facilities or farms, and we do not source from Taylor Farms Mexico."

Men can now make the first move on Bumble, ending the longstanding rule that only women can initiate a conversation on the dating app. Bumble has been struggling to get its groove back in the ever-evolving dating landscape, with its stock down 56% over the last year. Bumble — which is under pressure to win back Gen Zers who are exhausted with dating apps — said today that it's making the change in part to appease younger folks. Many women "appreciate the flexibility for men to start the conversation," the company said in a statement, citing internal research. The move comes after Bumble CEO Whitney Wolfe Herd said in May on "The Axios Show" that the company would eliminate the swipe feature and hinted at a shift away from the women-must-message-first rule, stating: "We will not force one gender over another to do something first."

Super Micro Computer cited strong demand for its AI-optimized servers and forecast fiscal 2027 revenue above expectations, sending shares up over 8% this afternoon in extended trading.

In 1913, the Federal Reserve was created in part to ensure that a dollar would always be accepted as a dollar. For much of the previous century, bank-issued dollars carried the credit risk of the banks that issued them — and therefore traded at different prices in different places. The Panic of 1907 led directly to the Federal Reserve Act of 1913, which put the Fed at the center of a national payments system, settling payments between member banks with reserves they were required to hold at the Fed. The Fed was also tasked with promoting "par clearing": A $100 check drawn on one member bank should be worth $100 when presented for payment at another.

Economists call this the singleness of money, and they say it's important. Without it, every payment comes with an exchange rate, the way dollars once did. Stablecoins still come with an exchange rate — at any given moment, one USDC might be worth more than one USDT, or vice versa. The BIS says this disqualifies stablecoins from being used as money: "Money-like claims that are not able to circulate with no questions asked cannot really function as money."

Two startups aim to change that. The Better Money Company is, essentially, a clearinghouse for stablecoins: It promises to exchange any two compliant stablecoins 1:1, no questions asked. A second startup extends the fungibility of stablecoins to US dollars. Ubyx promises to allow fintechs to exchange any approved stablecoin for a US dollar, 1:1, helping "stablecoins to exhibit singleness of money."