Stay informed with free updates
Simply sign up to the Technology sector myFT Digest — delivered directly to your inbox.
Tech has always needed its storytellers. By turning impersonal electronics into objects of high style and personal empowerment, Apple’s Steve Jobs once set the direction for an entire industry.
These days, it also needs people who can mobilise almost unimaginable amounts of capital. Meta, Alphabet and Amazon stunned Wall Street in recent days with plans to spend more than half a trillion dollars on AI data centres this year, far more than expected.
If anyone was made for this moment, it is Elon Musk. His ability to persuade investors to suspend their disbelief and back his sci-fi adjacent dreams (rockets to Mars? Fleets of robot taxis?) is unrivalled. It also helps, of course, that he has pulled off impressive feats of engineering, like pushing electric cars into the mainstream with Tesla and virtually monopolising the market for launching rockets into orbit with SpaceX.
Even for Musk, however, the financial demands of the moment are daunting. That is the main message from the news this week that he is about to fold xAI, his private AI company, into SpaceX. At $250bn, it will be the most expensive corporate acquisition ever mounted.
Regardless of the industrial logic he has claimed for the deal, it is, first and foremost, about laying the ground for a massive wave of financing to come. The Muskverse — the constellation of companies over which Musk exerts control — is in serious need of recapitalisation.
Consider the stretched finances of Musk’s most prominent ventures. xAI is burning through $1bn a month but is well behind rivals like OpenAI in building a business around its hugely expensive AI models. Tesla’s car operations have hit a ceiling as competition from China has mounted and, although it has a big financial cushion, it is about to turn cash flow negative for the first time in a decade as it pours money into driverless taxis and humanoid robots. SpaceX has found a money-spinner in its communications network, Starlink, but it still faces hefty investments after a patchy series of test launches for its giant new rocket, Starship.
Things are about to get much more expensive. Along with all the AI supercomputers, rockets and robots he wants to build, Musk is also eyeing huge solar plants and chip fabs, which rank among the most complex and costly bets in all of tech.
xAI raised $20bn last month (that is about as much as Tesla has raised in its entire history). SpaceX is hoping for $50bn later this year from what would be the biggest IPO ever. But with one of today’s biggest AI data centres costing about that much to build, he will need much more.
Musk is likely to mobilise the money. After all, he has conjured up an astonishing $2.5tn in value (what the public and private markets think his main ventures are worth), despite his companies’ relatively modest cash flow.
Combining SpaceX and xAI, whose ability to keep raising ever larger amounts of money is unclear, echoes previous moves by Musk to merge companies whose outlook looks uncertain. It is also a way to reward investors who have backed his deals through thick and thin. More than anything, though, it sets up the next storyline in the audacious, always changing narrative Musk has used to beguile the financial markets.
In his telling, Tesla was going to be the first mass producer of electric cars — but that was before it pivoted to become an alternative energy conglomerate, then an operator of self-driving taxis. Now, he talks about it building “an enormous robot army”.
Even by Musk standards, the logic claimed for this week’s deal sounds a stretch: to make it easier to launch data centres into orbit, where they will be able to harness the Sun’s power more efficiently to generate AI that can “understand the Universe and extend the light of consciousness to the stars!”. For aficionados of the Muskverse, it is the perfect combination, a story yanked from the pages of science fiction that has just enough of a connection to reality to seem plausible.
Musk, after all, is far from alone in the tech industry in thinking that flinging data centres into space could be the answer to the energy shortages the AI world is facing. If so, SpaceX could be a trump card.
If this proves to be the key to unlocking the next trove of capital, then why stop there? Tesla is already looking to ramp up its production of solar panels to power SpaceX’s fleet of 1mn orbiting AI satellites, while Musk has talked about using space-borne AI to organise fleets of robotaxis back on Earth.
That doesn’t make a merger of SpaceX and Tesla the inevitable next move. It will only happen if it makes a better story for the capital markets. Also, if SpaceX pulls off its planned IPO this year, Musk will be in charge of two giant public companies, leaving him with less room to manoeuvre than he has enjoyed with his private ones.
There is also financial fashion to consider. Eventually, investors will get over their infatuation with the AI spending boom and demand dependable cash flow again. At that point, we will get to assess the durability of the Musk Exception — the leeway investors have been willing to grant a supposed visionary engineering genius. Until then, he will have to keep the plates spinning and the deals coming.