Who pays for the buildout, who profits, and what that financing structure commits everyone to.
Loading market prices…
Market data by TradingView.
Stablecoin issuers froze about $3bn of the funds stolen from Binance within hours, according to Binance. Their freezing powers are written down. What outsiders cannot see is how those powers are applied, and who can challenge a decision.
A new policy report says AI data centres are diverting memory-chip supply, and consumers are already paying for it.
American AI policy begins from the premise that there is a race with China and that America must win it. Pope Leo XIV's first encyclical begins from the premise that treating AI as a race is itself the danger. Peter Thiel thinks that makes the pope an ally of Beijing. Underneath the row sits a question neither side can avoid: who gets a say in what these machines are for?
Nvidia's $20bn Groq deal delivers a takeover's substance while avoiding the word, and the antitrust scrutiny that comes with it.
American technology companies will spend around three quarters of a trillion dollars this year on infrastructure they are no longer financing entirely from their own cash flow. What fills the gap is a financial system, and the question of which country has the better one has quietly become a question about which country gets the future it is paying for.
Alphabet has written $43.8 billion of backstops on data centres it does not own, against which it carries a fair-value liability of $815 million. Nvidia has just answered with $500 billion of financing platforms of its own. The AI contest is still reported as a race between models and chips. The more consequential competition is now over who can supply the cheapest credit, and where the risk settles once the arranging is finished.
The gap between what frontier AI earns and what is being spent to make it possible is enormous, and it is usually reported using the most flattering measure available. Read it with the right numbers and the interesting question is no longer whether AI is a bubble. It is that one lab has just shown the revenue can arrive, and that the structures built while everyone waited for it will outlast the waiting.
SoftBank sold its entire Nvidia stake and part of its T-Mobile holding to help fund its multibillion-dollar commitments to OpenAI and chip startup Ampere, deepening its concentrated AI bet.
Meta's 6.6GW nuclear deal mostly funds reactors that won't deliver power until the 2030s, with only a fraction available immediately.
In Lanzarote, the residual heat of an eighteenth-century eruption cooks lunch for tourists. A few hundred kilometres west, an Icelandic-led consortium has begun drilling beneath Tenerife to find out whether the same kind of heat can carry an island's electricity system — with roughly half the exploration bill paid by the public.
Amazon's cut of roughly 14,000 corporate jobs is less a cost-saving measure than an early demonstration of AI-driven headcount reduction.
By converting its CHIPS Act grants into an Intel stake, the US government has crossed from subsidizing the chip industry to owning a piece of it.
Chips, models and cloud capacity are no longer ordinary exports but instruments of statecraft—granted to allies, withheld from rivals, traded for investment and alignment. Three developments in recent weeks suggest the transactional era has arrived; the harder question is how long such leverage survives being used.
A handful of firms are financing each other's AI ambitions, and the same dollars increasingly appear to travel a short loop between supplier, customer and investor.
Data centers are driving grid costs sharply higher, and the largest capacity market in the US is quietly passing the bill to households.