TheChek News Flow

The fast-tracked deal is premised on SpaceX’s satellite prowess combining with xAI’s technology

Elon Musk took a gulp of water on stage at the World Economic Forum in Switzerland last month and started talking up one of his latest passions: data centers orbiting Earth.

“The lowest-cost place to put AI will be space,” Musk said. “That will be true within two years, maybe three.”

Behind the scenes, the groundwork was already under way for a megamerger of SpaceX and xAI—his artificial-intelligence startup—intended to make the sci-fi vision a reality.

Within days, bankers from Morgan Stanley provided an estimated valuation of both companies that could be used to structure a merger, according to people familiar with the matter and investor disclosures viewed by The Wall Street Journal. SpaceX’s board decided as of Jan. 30 the company was worth $1 trillion, and xAI’s board decided it was worth $250 billion, the documents show.

SpaceX and xAI signed a merger agreement creating a $1.25 trillion company on Jan. 31. The deal closed two days later, making it the biggest corporate tie-up by value in American history.

The merger follows a busy year in dealmaking for Musk, from the combination of his social-media company X and xAI to Tesla’s $2 billion investment in the AI firm. It also promises a windfall for investors in xAI.

The billionaire entrepreneur has envisioned making xAI’s Grok the most popular artificial intelligence in the world, and building out its capabilities requires a lot of money. And for the startup, being hitched to SpaceX ahead of an initial public offering of stock gives it more financial muscle to compete with the likes of OpenAI and Anthropic.

However, longtime investors in the satellite builder and rocket operator have had to make room for Musk’s AI shareholders, and they face the prospect of supporting an artificial-intelligence company through what some believe to be a bubble reminiscent of the dot-com boom and bust.

Merging SpaceX and xAI represents a level of risk-taking and ambition that is novel even for Musk. The $1.25 trillion value assigned to the new SpaceX is a bet on Musk’s plan for the convergence of AI and space. It is also a wager on unproven technology deployed at a vast scale.

In a memo sent to employees Monday, Musk said starting launches of AI satellites is the company’s immediate focus. SpaceX, xAI and Musk didn’t respond to requests for comment.

Work on the deal gathered steam around the time Musk won a key victory in November, when Tesla investors endorsed his moonshot $1 trillion pay package, the largest ever for a public-company boss.

Not long after that, members of his inner circle set their sights on an even more audacious target: merging SpaceX and xAI. Combining them had emerged as a priority, people familiar with the deal said, given xAI’s capital needs and SpaceX’s coming IPO.

Some investors in xAI became convinced Musk’s companies should be more intertwined, given the ties the AI firm already had established with them.

Soon enough, SpaceX executives were selling the company’s new focus on AI satellites in discussions with banks and large investors, people familiar with those discussions said. The space company previously told investors it would generate ​​roughly $8 billion in adjusted earnings on $16 billion in revenue in 2025.

For investors, Musk’s corporate maneuvers have often been lucrative: Banks that held underwater debts from his buyout of Twitter were ultimately repaid at around 100 cents on the dollar plus years of high-interest payments.

“You invest with Elon and that is what you get, and if you don’t like it, you leave,” said Ross Gerber, an independent investment adviser and longtime investor in Musk companies, including X and Tesla.

Musk has previously united disparate corners of his business empire. Last year, he merged X, the social-media company formerly known as Twitter, and xAI into one company valued at an estimated $113 billion. Before the deal, X had also owned a stake in xAI and they did business with each other.

The transaction would ultimately serve as a blueprint of sorts for the tie-up between SpaceX and xAI.

SpaceX was one of xAI’s earliest enterprise customers. It has been scaling up the internal use of Spok, a tailored version of xAI’s Grok chatbot that has been trained on the space company’s data, according to people familiar with the matter.

Last summer, SpaceX invested in xAI while pressing forward with a busy launch schedule and expanding its Starlink satellite-internet service.

At xAI, contractors broke ground on a second data-center project around Memphis, Tenn. But several executives left Musk’s artificial intelligence startup because of concerns about its management and financial health. Leaders from Valor Equity Partners, which is also a SpaceX investor, got more involved with running the company, the Journal reported.

Behind the scenes, SpaceX had been studying how orbital data centers could work and had a technical breakthrough associated with the technology last fall, the Journal has reported.

Musk began talking more, including on X and at events, about harnessing the sun’s energy to power orbital data centers. On Jan. 21, SpaceX filed documents in Nevada, where xAI is incorporated, to create entities overseen by its finance chief, Bret Johnsen.

Those entities, later used for the merger, carried the name K2, an apparent homage to the Kardashev scale—a theoretical framework that illustrates a civilization’s advancement through energy usage, including from stars.

The next day, Musk touted his vision for the future of AI data centers in space on stage in Davos, while professionals involved with making the deal happen were getting to work, according to people familiar with the matter.

Bankers at Morgan Stanley, long Musk’s bank of choice, worked on both sides of the deal, a somewhat unusual arrangement, as each company in an M&A deal typically hires its own advisers to ensure it is getting a fair shake. But in this case, Musk controls both companies.

At the high end, the bank said, SpaceX could be valued at as much as $1.26 trillion, far north of the $800 billion valuation it targeted as part of a recent secondary stock sale. They said xAI could be worth as much as $294 billion.

Both companies went for clean, round numbers: SpaceX’s board decided as of Jan. 30 the company was worth $1 trillion, and xAI’s board decided it was worth $250 billion, the investor disclosures show. Morgan Stanley didn’t provide a “fairness opinion” on the valuations, a common feature in some M&A deals to ensure they were reasonable, although it is less common in transactions between private companies.

Morgan Stanley declined to comment.

Steve Kaplan, a professor of finance at the University of Chicago’s Booth School of Business, said the transaction reminded him of deals in the dot-com boom, like the $180 billion merger between AOL and Time Warner.

“That’s the thing that’s very hard here,” Kaplan said. “They’ve come up with their valuations and their share exchange rates, and you don’t know how accurate those numbers are, particularly since neither one is public.”

The rationale behind merging two very different companies has been the subject of intense discussion among people close to SpaceX.

Many are supportive of the new vision. They expect the company to frequently launch orbital data centers on Starship, a rocket still under development, similar to how it deployed the largest satellite fleet in history with Starlink on a different SpaceX vehicle.

However, some SpaceX investors believe xAI investors gained a larger stake in the combined company than they would have preferred. The transaction is handing xAI investors a roughly 20% stake in the combined company.

On a call with investors on Monday, SpaceX representatives including CFO Johnsen confirmed plans to take the company public this summer, according to people familiar with the discussions.

Puma’s new Chinese backer will give Nike a run for its money

Anta’s sales have grown on average 17 per cent a year for the past five years, with the US maker of the famous swoosh managing 4 per cent

Companies buying into sports brands often themselves sound like coaches. Anta Sports, about to become Puma’s largest shareholder with 29 per cent, is no different: it wants to help the ailing German brand get stronger and perform better. Unlike the typical coach, though, it’s Anta that wants to be the real star.

Anta boss Ding Shizhong has been clear that he doesn’t want his Hong Kong-listed group to be just the Chinese version of Nike. Judging by his current model, that means taking his own-label sportswear global while managing or being involved with brands such as Puma and outdoor wear specialist Jack Wolfskin. A successful turnaround of Puma’s fortunes in China would help his pitch.

For now, Anta’s sales are almost entirely in China, and its $10bn of revenue compares with $46bn at sportswear leader Nike. But in growth terms, it has already surpassed some of its peers. Its sales have grown on average 17 per cent a year for the past five years, with Nike managing 4 per cent. Anta’s shares have delivered a total return of 450 per cent over the past decade, according to LSEG, where the US maker of the famous swoosh returned just 19 per cent.

Anta has its work cut out with Puma. The German sports brand’s sales in China dropped by more than a fifth between 2023 and 2025, according to Nomura analysts, who blamed slow updates to its stores, weak online channels and a reliance on a few big retail partners for the bulk of its sales.

In theory, Chinese consumers want what western ones do: to be fit and well. Both regions have a population that is ageing but would presumably rather not feel or dress like it. But it has proved difficult for western brands to fully accept the degree of localisation needed to succeed there. More than half the clothes Adidas sells in China are now designed and developed there specifically for that market, the German brand said recently.

Nike, too, wrestles with the same trends. It admitted in December that its China turnaround was behind schedule and that it needed to operate more locally, as well as follow boss Elliott Hill’s global push back into high-end performance and away from casual gear. “The reality is we have become a lifestyle brand competing on price in China,” Hill told investors in December.

That sounds dangerously like the place Puma finds itself in, notwithstanding its strength in football and basketball. It is an opportunity for Anta though, if it can use its local expertise to boost Puma in its home market. Asia Pacific in total provides about a fifth of Puma’s sales, so its overall recovery still rests largely on its actions in Europe and the US. But a strong game in China would boost Anta’s standing in global leagues too. 

Platform offers discounts for weight-loss and fertility medicines sold by Novo Nordisk, Eli Lilly and others

Weight-loss and fertility drugs sold by Novo Nordisk, Eli Lilly and EMD Serono are among the medicines available on Donald Trump’s direct-to-consumer drug platform that launched on Thursday evening.

TrumpRx.gov, which directs people to websites where they can find drugs at a discount, offers more than 40 medicines as part of the US president’s drive to bring down consumer costs ahead of November’s midterm elections.

Certain AstraZeneca and Pfizer drugs are also available on the website, and the White House said more would become available.

“Dozens of commonly used drugs available today are just the beginning,” Trump said on Thursday. “You’re going to save a fortune.”

All five companies last year agreed to lower US drug prices in exchange for tariff reprieves and expedited reviews for new products at the Food and Drug Administration.

Pharmaceutical companies have recently disclosed that the price negotiations with Trump have blunted their sales. Pfizer this week said its 2026 sales guidance was hurt by the “unfavourable impact” of its deal.

Novo and Lilly also said their sales had been hit by lower prices for weight-loss drugs. The duo agreed pricing deals with Trump in November.

“The launch of TrumpRx will further extend patients’ reach to Ozempic and Wegovy, including the newly approved Wegovy pill,” Novo said in a statement.

Rival Lilly said its obesity shot Zepbound and one for insulin would be listed on TrumpRx. 

The launch of TrumpRx comes amid a price war between Novo and competitors in the weight-loss market.

Telehealth company Hims & Hers said on Thursday it would sell a version of Novo’s weight-loss drug Wegovy for $49 a month in the US, compared with the Danish company’s $149 monthly rate.

Novo threatened legal action on Thursday, accusing Hims & Hers of “illegal mass compounding” and alleging that the company posed “significant risk to patient safety”.

TrumpRx also lists drugs for in vitro fertilisation sold by EMD Serono, which is owned by Germany’s Merck and is one of the world’s largest providers of fertility assistance. The company, which announced a pricing deal with Trump in October, said the cost of the drugs for a standard IVF cycle would drop 84 per cent.

“Behind every fertility treatment is a person’s hope and dream,” Miguel Fernández Alcalde, president of EMD Serono, said in a statement. “Each step and effort towards easing this burdensome journey makes a real impact for these patients.”

Big Tech’s ‘breathtaking’ $660bn spending spree reignites AI bubble fears

Stocks tumble despite strong earnings reports from most Silicon Valley heavyweights

Big Tech stocks sold off heavily after unveiling plans to spend $660bn this year on AI, as investors fret that the “breathtaking” capital expenditures are outpacing the earnings potential of the new technology.

Amazon, Google and Microsoft are set to lose a combined $900bn in market value since filing their quarterly earnings over the past week.

Shareholders balked at the sector’s gargantuan capex plans — more than the GDP of Israel — which overshadowed strong revenue growth at the companies’ cloud arms.

Along with social media giant Meta, their proposed outlay on data centres and specialised chips needed to train and run advanced AI models would mark a 60 per cent rise from the $410bn they spent in 2025 and a 165 per cent increase from $245bn in 2024.

“The capex is breathtaking,” said Jim Tierney, head of the concentrated US growth fund at AllianceBernstein.

Even a 14 per cent boost to their combined annual revenue to $1.6tn was not enough to overcome the pessimism. Apple, which has sat out the AI capex arms race, was the only Silicon Valley behemoth to emerge unscathed, with shares up 7.5 per cent since it reported record sales.

Amazon fell 11 per cent after market on Thursday after saying its capex will reach $200bn this year — $50bn more than expected — surpassing already eye-watering numbers from Google and Microsoft.

Chief executive Andy Jassy said such large sums were needed to position the company for a boom in AI, chips, robotics and satellites. He pointed to a 24 per cent growth in revenue at Amazon Web Services as evidence investment was starting to pay off.

Worst hit was Microsoft, which has fallen 18 per cent since it reported last Wednesday. Revenue at Microsoft’s cloud division rose 26 per cent to $51.5bn. But this was slightly slower than expected and the market reacted to a 66 per cent surge in quarterly data centre spending.

Microsoft also laid out for the first time its exposure to OpenAI. It disclosed that 45 per cent of its $625bn book of future cloud contracts was from the start-up, leading analysts to flag its over-reliance on one customer.

Anna Nunoo, a senior analyst at AllianceBernstein, said this quarter’s earnings had brought a “shock in terms of the increased capex”.

“The onus is on Microsoft and Amazon to prove out the attractive returns on all the spending,” she added.

Even record earnings at Google were not enough to override these concerns. Parent company Alphabet surpassed $400bn in annual revenue for the first time and made $132bn of profit in 2025. But plans to double capex to $185bn still knocked its shares.

“AI bubble fears are settling back in,” said Brent Thill, an analyst at Jefferies. “Investors are in a mini timeout around tech, and nothing the companies say fundamentally matters.”

The spending plans also indicate that more time and money will be required to deliver the full promises of AI.

Higher capex “telegraphs that it may take longer for AI strategies to play out”, said Dec Mullarkey, managing director of $300bn asset manager SLC Management. “Not welcome news for investors that are already fixated on when AI-related revenue will start to show up.”

Meta also said last week its capex would double to $135bn, but the stock rose 10 per cent as the social media network showed how AI was improving advertising efficacy. However, it has since given up those gains as it was caught in a wider market rout that has pushed the tech-heavy Nasdaq down 4 per cent in the past five days.

Software stocks were hit due to fears about new AI coding tools from Anthropic and OpenAI disrupting their businesses.

Markets have also been rattled by confirmation that OpenAI’s $100bn investment and infrastructure deal with Nvidia has not gone ahead.

Oracle, which relies on OpenAI for a large share of its future cloud business, dropped 18 per cent over five days even as it raised $25bn in debt and insisted it was “highly confident in OpenAI’s ability to raise funds and meet its commitments”.

Apple stood out as the winner from the round of earnings. 

The company reported a record $144bn in quarterly revenue driven by a surge in iPhone 17 sales in the US and China. Capex fell 17 per cent to $2.4bn in the final three months of the year, giving it an annual total of about $12bn.

In January, Apple struck a deal to use Google’s Gemini to overhaul its AI features, including its Siri voice assistant.

“Apple’s tiny capex is the AI dividend of partnering with Google for compute and frontier models,” said Dan Hutcheson, vice-chair of market intelligence firm TechInsights. “This shifts Apple’s AI capex to a pay-as-you-go model,” with the iPhone maker outsourcing most of the underlying infrastructure costs to Google.

The partnership “absolutely” explained some of Google’s increased capex plans for 2026, Hutcheson added.

Chipmaker Nvidia, the world’s most valuable public company, now faces a volatile market as it prepares to announce earnings later this month. After more than three years of being asked to stomach escalating capex, investors are looking for an imminent end to spending based on faith in AI.

“These are wild times,” said Drew Dickson, founder of Albert Bridge Capital. “We’ve evolved from an environment where capex alone was enough to trigger euphoria to one where the market expects it to translate into revenue growth in a time horizon that makes little sense.”

‘I thought we’d be faster and I was wrong’: Sanofi chief defends R&D pivot

Paul Hudson struggles to convince investors he can plot a more profitable future for the French pharma group

Paul Hudson has yet to solve a conundrum raised with him when he took charge of Sanofi in 2019: finding a successor to its hit drug Dupixent.

The eczema and asthma treatment made up more than a third of the French pharmaceutical company’s sales last year, but trial setbacks last year have increased pressure on its British boss as the drug nears the end of its patent protection in the early 2030s.

Making up for the impending loss of Dupixent profits is all the more urgent for Hudson as Sanofi’s board weighs whether to renew his mandate, due to expire this year. Advisers to the company, former senior employees and analysts told the FT that Hudson had yet to show he could deliver future growth.

Defending his record in an interview with the FT, Hudson said more time would be required for his research and development-led push to bear fruit after a “difficult 2025”.

“My first conversation in September 2019 was ‘how are we going to manage the end of Dupixent?’,” he said. “The next 12 to 24 months will tell us whether the choices we’ve made — I’ve made — will put us on track.”

In the more than six years since he arrived from Novartis, Hudson has sought to transform Sanofi from a vast business spread across multiple pharmaceutical sectors and countries to a more focused, research-led operation that can discover more of its own medicines.

One of the former employees said Hudson had a “negative view” of the company’s past performance and its lack of focus: “We were weak, indecisive, caught in-between: we did everything and he wanted to do less.”

The Mancunian has invested in R&D and sought to streamline the company, including by selling off half of Sanofi’s consumer healthcare business in 2024, following similar moves by rival pharmaceutical groups.

Sanofi has also successfully expanded sales and treatment areas for immunology treatment Dupixent, underlining its historic strengths in marketing treatments. The drug was jointly developed by US biotech Regeneron, with which Sanofi splits the profits.

In recent results, Sanofi reported €43.6bn in sales and €12.1bn in operating income, showing strong growth from 2024 and continued strong profit margins for the group under Hudson. But 2025 has also brought home how efforts to foster in-house innovation have yet to materialise.

Since Hudson announced a poorly received strategy to focus on more R&D spending in 2023, research costs rose to €7.8bn last year — but there is little to show for it.

Nicolas Dumas, a partner at Roland Berger and former Sanofi employee, said Hudson had been “rather courageous” in cutting back management and reducing investment in underperforming areas. But he added: “In 2025, they had very bad news . . . there is nothing to replace Dupixent.”

In September, the company had almost $13bn wiped off its market value in one day after its experimental eczema drug amlitelimab underwhelmed compared with Dupixent in treating a severe form of the disease. The stock is down more than a fifth over the past 12 months, leaving it worth about €100bn.

Tolebrutinib, an experimental drug for multiple sclerosis, also faced a double setback in December. The US Food and Drug Administration refused to approve its use in an advanced form of the autoimmune disorder and the drug failed a late-stage trial in another form of MS.

While trial failures form part of the vicissitudes of the pharmaceutical industry and R&D investment takes time to deliver results, the setbacks have underlined a lack of depth in Sanofi’s pipeline, which had just six blockbuster drugs excluding vaccines in 2025 — far fewer than most European pharma rivals.

Several people said that management had overhyped potential drugs, with overly confident projections that have riled investors and analysts, adding to pressure on Hudson.

One of the people familiar with the board said Sanofi’s “stagnating” share price and uncertain outlook had put the question of Hudson’s future “on the table” ahead of a general assembly in April, while several other people familiar with the company said that Hudson’s role ought to be in question.

Asked about criticism of the company’s presentation of pipeline drugs, Hudson said: “On occasion, the data has been very encouraging and then surprised to the negative. Then it’s very easy for people to say, well, you were too optimistic.”

But the trial failures have intensified doubts over whether Hudson’s research-focused turnaround will bear fruit.

Hudson said he had “underestimated” the difficulty of turning the group into a more innovative company. “I thought we’d be faster and I was wrong because I think our starting point was further back than I expected,” he said.

He suggested Sanofi would turn to more acquisitions to replace Dupixent if further drugs failed. The company’s $9.5bn purchase of Blueprint Medicines last year was its largest under Hudson’s tenure and gave it a drug already approved, pushing up the acquisition price.

But with just $725mn in pro forma sales for this immunology treatment Ayvakit last year, Sanofi will need to develop or buy in more new treatments to address concerns over impending lost income from Dupixent.

Hudson pointed to upcoming trial results in Crohn’s disease, chronic obstructive pulmonary disease and oncology as tests of the company’s improved R&D output.

However, Peter Verdult, a pharmaceutical analyst at BNP Paribas, said there appeared to be few significant trials in 2026 to win back investors.

Sanofi was in the “proverbial doghouse” among investors, with several pointing the finger at Hudson’s leadership, he added. “It’s noticeable and in some ways it’s understandable because of how poorly the shares have performed.”

Still, Hudson is far from solely responsible for the company’s struggles, with several people also questioning the strength of the board and its chair, Frédéric Oudéa.

Oudéa struggled to steer French bank Société Générale through a string of scandals during 15 years as chief executive. He had no experience of the pharmaceutical industry before his appointment in 2023.

One of the bankers familiar with the business said: “Paul Hudson does not have, unfortunately for him, a board who will accompany him as it should.”

Sanofi said it would not comment on speculation regarding Oudéa, adding that the board had “deep expertise across pharmaceuticals, science, technologies, finance and global business”.

Beyond his renewal, Hudson also faces a testing 2026 in charge of leading industry interactions with the Trump administration when he takes the rotating chair of trade body PhRMA later this year.

Sanofi’s vaccine sales have already taken a hit after changes to vaccination schedules by the administration. The company also joined other drugmakers in striking an agreement to lower some US drug prices at the end of last year, although this did not have a meaningful impact on the 2026 outlook.

Asked about the future of his own leadership, Hudson replied: “It’s really a board decision on whether they think we’re making enough progress . . . we need the pipeline success for people to say ‘you know what? We’re finally an R&D-led organisation’.”

New emails reveal the convicted sex offender’s ability to fulfil the wants and needs of some of the world’s richest and most powerful people