TheChek News Flow

Trump administration says it wants to review national security implications of $1.5bn project that is 80% complete

The US has halted construction on a $1.5bn offshore wind farm off Rhode Island, dealing a new blow to Danish energy group Ørsted as it tries to repair its balance sheet.

The Bureau of Ocean Energy Management issued a stop-work order on Friday, citing the need “to address concerns related to the protection of national security interests”. The agency did not provide further details.

Revolution Wind, which Ørsted is developing with Skyborn Renewables, a unit of BlackRock’s Global Infrastructure Partners, is 80 per cent complete. The project has finished installing foundations and 45 of the 65 planned turbines. 

It was due to begin operations next year and supply enough power for 350,000 homes in Rhode Island and Connecticut under 20-year contracts.

It is the second time the Trump administration has intervened in a big US offshore wind project. In April, the White House paused, and then later approved Equinor’s $5bn Empire Wind project, which was at an earlier stage of development than Ørsted’s. 

On Saturday, Ørsted said it has obtained the necessary permits and was “evaluating its options”, including legal action. The company said it would update the market on the financial implications of the order “in due course”.

The potential delay, or loss, of revenues from the project piles greater pressure on the world’s largest offshore wind developer, which launched a $9.4bn emergency rights issue this month after struggling to attract partners for its Sunrise Wind project off New York. The share sale triggered a sharp sell-off in Ørsted stock.

Revolution Wind, the halted project, is part of 8.1GW of wind projects globally that Ørsted said it was relying on to put its finances back on track. It is one of five offshore wind projects in the US that are under construction. 

“This is the kind of stuff that happens in third-world countries and instead it’s happening in what is supposed to be the bastion of the free market. It’s just not serious,” said a spokesperson for the Global Wind Energy Council, an industry group. 

“It has an extremely chilling effect in the sense that the US is not a safe place for investment.”

The stop-work order represents the latest setback to Washington’s Nato ally Denmark. US President Donald Trump has repeatedly said he wants to take control of the giant Arctic island of Greenland from Denmark, and has not ruled out using force.

He has also trained his sights on Novo Nordisk, the pharmaceutical group that has reduced prices on its weight-loss drugs in the US sharply in recent weeks.

Denmark, which is buying F-35 fighter jets from the US, has offered to increase its focus on Arctic security.

But on Friday, Danish foreign minister Lars Løkke Rasmussen was in California where he signed a new partnership with the state’s governor Gavin Newsom, who is the self-appointed head of the “resistance” to Trump. 

It has been a wild ride for Intel over the past few weeks. Eight times the stock has seen a one-day move of plus or minus 3%.

The volatility stems from all of the issues roiling markets these days: geopolitics, trade, artificial intelligence, U.S. industrial policy, and a micromanaging executive branch.

In Intel’s case, everything is amplified by a decadelong struggle with technology.

The headlines whipping the stock have focused on Intel’s financing, but money alone won’t solve the company’s problems, which are rooted in manufacturing.

Those challenges have led to turnover in Intel’s executive suite, undermining a well-respected longtime engineer in Pat Gelsinger, who stepped down in December after failing to mount a turnaround.

The new CEO, Lip-Bu Tan, has limited options and time to fix the situation. On Aug. 7, President Donald Trump kicked off the latest round of Intel’s stock volatility when he wrote on social media, “The CEO of INTEL is highly CONFLICTED and must resign, immediately. There is no other solution to this problem.” Intel shares declined more than 3% that day.

But another solution emerged after Tan visited the White House to gain the president’s favor. On Aug. 11, Trump wrote on Truth Social, “Mr. Tan and my Cabinet members are going to spend time together, and bring suggestions to me during the next week.” Intel stock was up more than 3% that day.

This launched two weeks of stories surrounding Intel’s financing through the U.S. Chips Act and private investors. Late Friday, Intel agreed to a deal in which the U.S. government gets a 9.9% stake in the company, in return for fulfilling its full payout under the U.S. Chips Act, a total of $11.1 billion.

The law passed by Congress and signed by President Joe Biden in 2022 never called for equity stakes for that funding. But political dynamics have shifted. Sen. Bernie Sanders (Ind., Vt.), a self-described democratic socialist, had called for the government to receive equity in exchange for funding, but his policy wasn’t included in the final legislation.

The Biden administration granted Intel $11.1 billion from the Chips Act, spread over five projects, including a $3.2 billion special contract with the U.S. Department of Defense. The grants were disbursed based on capital investment progress. Intel had received only $2.2 billion, leaving most of it up for grabs. What had been free to Intel now comes with a heavy price.

To complicate matters, on Monday Intel and SoftBank Group announced a $2 billion cash infusion in exchange for about 2% of Intel, another seemingly dilutive deal. The price of $23 per share is a small discount to Intel’s closing price earlier that day. On Wednesday, CNBC reported that Intel is seeking similar deals at discounted share prices with other private investors.

Intel needs this funding. At the end of June, the company had $21 billion in cash and short-term investments. On the other side of the ledger, subtracting capital expenditures from operational cash flows, Intel lost $45 billion over the past 3½ years, $6 billion of that in 2025. New financing is crucial, but it won’t be enough to get Intel back into the lead.

As of Friday, the stock is down 64% since its last peak in early 2020.

It’s tempting to look at the decline and see a cheap stock. Indeed, the stock trades right around its book value, which means that investors don’t think it’s worth more than what’s on the balance sheet—assets minus liabilities. Back in 2020, Intel traded at 3.8 times book value.

A rebound would probably provide a giant windfall for shareholders. But Intel has been trying to right the ship for four years, and in 2025 it doesn’t seem any closer to success.

In 2021, with the company already in crisis, new CEO Pat Gelsinger announced an ambitious plan to launch five new manufacturing processes in four years, an unheard-of pace. The goal was to get Intel’s technology back on track quickly after it had lost leadership to Taiwan Semiconductor Manufacturing and, like Taiwan Semi, provide service to outside customers.

So far, only the first of these five “nodes” can be considered a success, having made five different lines of Intel chips from 2021 to 2023. There have been three subsequent nodes, and each was a different sort of failure, allowing Intel’s main competitor, Advanced Micro Devices, to surpass its technology. AMD uses Taiwan Semiconductor to manufacture its chips.

The third of the five nodes was meant to launch Intel as a manufacturer for other chip companies. That didn’t happen, and it was used only sparsely by Intel. The fourth node was supposed to feature Qualcomm as the launch customer, but that went nowhere, and the node looks to have been scrapped.

In the first half of 2025, Intel has earned $53 million from outside customers for its manufacturing segment, which featured an operating loss of $5.5 billion.

While Tan has said that the fifth of the five nodes will make three generations of Intel chips, the company’s foundry still doesn’t have a large external customer. Intel hit its four-nodes-in-five years timeline, but only the first node, and maybe the fifth, have proved to be very useful.

Since 2021, Intel has spent over $100 billion building these nodes, with little to show for it.

Sure enough, Tan has a new approach. Instead of his predecessor’s “build it and they will come” attitude, he is essentially saying, we’ll build it, if they come.

Now, it’s a question of who shows up.

The world’s largest hedge fund made changes to its portfolio in the second quarter that culminated in the sale of several major Chinese technology stocks.

Bridgewater Associates exited its investments in a handful of companies, including Alibaba, Baidu, and NIO, by the end of the second quarter, filings with the U.S. Securities and Exchange Commission show.

The fund sold 5.7 million shares of Alibaba, 2.1 million shares of Baidu, and 1.7 million shares of Nio. The firm also exited auto maker Li Auto through the sale of 12,826 shares it held in the first quarter, and exited JD.com through the sale of 2.8 million shares.

Bridgewater didn’t respond to a request for comment. As of mid-July, the fund had roughly $136.5 billion discretionary assets under management.

Since July 1, Alibaba’s ADR has risen 6.4%, outpacing a 4.1% gain for the S&P 500. Baidu and NIO’s ADRs have gained 4.4% and 37%, respectively.

The portfolio adjustments come against the backdrop of a cooling economy in China. New yuan loans fell unexpectedly in July, representing the country’s first contraction in two decades and sparking concerns about a worsening economic downturn.

While Alibaba posted solid earnings in February, it underperformed estimates in May. A myriad of rivals, including JD.com, Meituan, and Temu parent PDD Holdings, have heaped pressure on China’s largest e-commerce platform.

Baidu, an artificial-intelligence heavyweight and a leader in autonomous driving, has expanded its footprint outside China through partnerships with Lyft and Uber. While Baidu broke a three-quarter streak of falling revenue in May, uncertainty persists amid competitive pressure from rivals like DeepSeek and questions over the company’s ability to find new growth drivers.

NIO, the maker of electric cars, has had a similarly bumpy ride. The company reported a wider-than-expected loss in the first quarter against a backdrop of intensifying competition in the premium EV market. Deliveries came in at 21,017 vehicles for July, up 25% from last year but down from 24,925 in June.

Japan sustained a one-two punch in late July. The perennially governing Liberal Democratic Party lost its majority in parliament’s upper house, throwing politics into turmoil. Three days later, the U.S. imposed 15% tariffs on exports that exceed 3% of Japanese gross domestic product.

Investors reacted...positively. The iShares MSCI Japan exchange-traded fund has climbed 9% since the trade deal with Washington was announced on July 23, consolidating a 20% gain year to date.

Job prospects for Shigeru Ishiba, the prime minister who led the LDP to a third straight election debacle, are looking up, too. “The likelihood of Ishiba surviving this political crisis has increased,” says Tobias Harris, founder of consultant Japan Foresight.

The 15% levy could spell relative advantage for automobiles, Japan’s No. 1 export to the U.S., notes Drew Edwards, head of Japan value equities at GMO. “With U.S. auto makers paying 50% tariffs on imported steel, the Japanese may stand to pick up market share,” he says.

Further gains in Japanese stocks could come harder. “The latest rally has taken us to our limit on Japan,” says Alex Wolf, head of macro investment strategy at J.P. Morgan Private Bank.

Other managers are shifting focus from industrial heavyweights to domestic-facing stocks, particularly financials. Monetary conditions are favorable for Japanese banks, argues Daniel Hurley, a portfolio specialist in international equities at T. Rowe Price. The Bank of Japan is holding its prime rate at 0.5%, while the yield curve for longer-term borrowing tips sharply upward, offering generous interest margins. His picks in the sector include Mitsubishi UFJ Financial Group and Resona Holdings, a regional player based in Osaka.

The yen could appreciate sharply in the medium term, adds Aaron Hurd, senior currency portfolio manager at State Street Global Advisors. Japanese interest rates should inch up while the Federal Reserve cuts and U.S. growth slows, reducing the pull of U.S. assets. The yen, currently at 147 to the dollar, “will be in the low-130s by the end of next year and 120s after that,” Hurd predicts.

A stronger currency would depress earnings for global manufacturers like Toyota Motor and Sony Group, but spell relief and looser wallets for consumers coming to grips with Japan’s first inflation in decades. “Japanese authorities will be pretty happy with a stronger yen,” Hurd expects.

The LDP’s electoral defeat, and need to entice new coalition partners, increases chances for some crowd-pleasing fiscal stimulus that could juice consumer spending in the meantime, Edwards says. “I wouldn’t be surprised to see some populist measures like a cut in taxes on fuels,” he says.

He is focusing on healthcare stocks, which have a surprisingly low weighting in the world’s oldest country. Picks include surgical supplier HOGY Medical and laboratory network H.U. Group Holdings.

A longer-range tailwind for Japanese stocks remains a governance sea change toward returning more capital to shareholders and reducing the deadweight of cross-shareholdings. Share buybacks jumped by more than half last year to $108 billion, according to Nikkei Asia. “Investors expect Japanese companies to allocate more money to active investments and shareholder returns,” says Kazunori Ito, director of Japanese equity research at Morningstar.

Japan cannot escape being a trade-reliant nation in a trade-war era. Exports have exceeded 20% of GDP for the past few years, twice the U.S. proportion. Other strengths are keeping some investors hopeful, though. “Japan remains the most overweight position in our portfolio,” GMO’s Edwards says.

A Bet on QXO Stock Is a Bet on This Billionaire. Why It’s Time to Buy.

Building-products supplier QXO is led by Brad Jacobs, who made two other companies leaders in their sectors.

If Brad Jacobs builds it, the gains will come. That was the case with his previous companies, including United Rentals and XPO, and it should remain the case for his latest, the building-products supplier QXO.

Jacobs, who has a net worth of $16.9 billion according to Bloomberg, isn’t a household name, but maybe he should be. The 69-year-old founded United Waste Systems —now part of Waste Management — United Rentals, and logistics firm XPO. United Rentals, which Jacobs ran from 1997 to 2007, and XPO, where he was CEO from 2011 through 2022 and remains executive chairman, have continued to thrive. Shares of United Rentals are up more than 400% over the past five years, more than quadrupling the S&P 500’s gain in the same time frame, while XPO has surged nearly 340%. XPO also subsequently spun off warehousing firm GXO Logistics and trucking brokerage firm RXO as separate businesses.

Investors are now banking on QXO being the next big Jacobs success story. Going public through a reverse merger in June 2024, QXO recently completed an acquisition of Beacon Roofing Supply for $11 billion, giving it an actual business to run. And business appears to be going well: The company reported a second-quarter profit of 11 cents a share on sales of $1.91 billion, beating analyst forecasts for earnings of 4 cents a share on revenue of $1.87 billion. The stock has gained 30% so far this year.

Analysts expect QXO will continue to generate strong revenue and earnings from Beacon. William Blair analyst Ryan Merkel, who thinks $40 is a fair value for the stock, says QXO can post mid-single-digit organic growth even without M&A activity thanks to the U.S. housing shortage, the data-center buildout, and the need to replace aging infrastructure.

But a bet on QXO is also a wager on Jacobs doing more deals, using a combination of the $2.3 billion in cash it has on its balance sheet and the expected cash flow it will generate from Beacon to make more acquisitions. CEO Jacobs notes that Beacon has annual sales of around $10 billion from nearly 600 branches, and a repair business that is rarely reliant on the whims of market forces. “When your roof is leaking, you fix it,” he says. “Beacon was the right first step.”

The expectation of more deals is priced into forecasts, with analysts predicting that sales will grow to more than $20 billion by 2028. Wolfe Research analyst Trevor Allinson notes that the building products industry “is ripe for disruption and tailor-made for Brad Jacobs.”

He added that the “largely domestic supply chain provides stability in an increasingly uncertain economic environment” and that he expects more acquisitions during the next few years. Allinson has a $44 price target on QXO, more than double Monday’s close.

Jacobs’ goals are even more ambitious. He says the goal is to eventually get to $50 billion in revenue within the next decade. That might seem extreme, but portfolio managers at Orbis Investment Management are optimistic—so much so that the company bought shares in QXO even before it acquired Beacon. Orbis invested in 2024, shortly after the company was formed via an investment by Jacobs in publicly held software and consulting firm SilverSun Technologies, which was renamed QXO. Orbis now owns a 14% stake in QXO.

Orbis had previously invested in XPO in 2012 and likes the potential for Jacobs to hit another home run with building supplies via QXO. “Fragmented competition should yield ample M&A opportunities, where QXO can create substantial value by improving the operations of acquired businesses,” John Christy and Eric Marais of Orbis noted in a report from September 2024, adding that “disruption risk is low” and that “it would be hard to come up with a better fit for the Brad Jacobs playbook.”

Dealmaking may not be as easy as expected. QXO has already walked away from a proposed deal to buy French electrical components maker Rexel after Rexel turned down a bid from QXO in the fall of 2024. QXO also seems unlikely to make a new offer for drywall distributor GMS after Home Depot subsidiary SRS agreed to buy GMS for $110 a share. QXO had offered $95.20 a share for GMS, and Jacobs suggested that QXO isn’t going to get in a bidding war with Home Depot.

“It is incredibly important to be disciplined on price,” Jacobs told Barron’s about the possibility of any future acquisitions. “We don’t fall irrationally in love with any particular company and we are not at all tempted to overpay.”

That discipline should pay off in the long run. Mike Dahl, an analyst with RBC Capital Markets who initiated coverage on QXO in July with an Outperform rating and $33 price target, says investors shouldn’t worry about QXO having to walk away from the GMS deal because QXO has many more options. “There are a significant number of large private companies—both family and private-equity owned—and an even larger number of small to midsize private players that we believe could likely be acquired at discounts to public multiples,” Dahl says.

QXO isn’t cheap—it trades at 39 times 2026 earnings estimates, nearly double that of the S&P 500—but the valuation doesn’t take into account the possibility of future acquisitions. QXO trades at 1.6 times book value, a better metric given its strategy. Orbis’s Christy and Marais argue that the stock should trade between two and five times. At the 3.5 times midpoint, QXO would fetch $46 a share, up nearly 125%.

With Jacobs just getting started, that seems like something investors can build on.

The Technical View

In early August, QXO successfully retested a cup with handle breakout that occurred on June 9, and is now constructing the right side of a potential double-bottom base, with a potential pivot point at $22.38. It looks likely to fill in the upside gap from the June 24 session, when the company announced the pricing of stock offering, by year end. Sell stops should be placed near $18.75. —Doug Busch

Equal-weighted portfolio has outperformed main index in recent days after long period lagging behind

Investors are warming to the S&P 500 index’s “forgotten 493” stocks, as stretched valuations and a sell-off in the AI sector trigger renewed warnings about outsized exposure to megacap companies.

The so-called Magnificent Seven US tech stocks dominate Wall Street’s blue-chip index and have carried it to record highs in recent years.

But this week the tech-heavy Nasdaq Composite index sold off on three consecutive days, prompted by worries over the future profitability of AI. Falls in some of the US market’s biggest, most heavily weighted names — Nvidia fell 3.5 per cent on one day — meant the S&P 500 was pulled lower, although so-called value stocks such as consumer staples, healthcare and utilities advanced at the same time.

“We’re seeing a bit of a rotation towards the forgotten 493,” said Damian McIntyre, head of multi-asset solutions at Federated Hermes. “The Magnificent Seven have rallied so much, it makes sense to take a breather here and reallocate.

“There’s a lot of nervousness around this AI spend,” McIntyre added, while “valuations are a lot more reasonable on the equal weight index”.

The S&P 500 index is weighted by market capitalisation, meaning the biggest stocks have an outsized impact on index-wide moves, which has grown as megacap tech stocks have soared in recent years. Concentration in the S&P 500 is at historic highs, with the biggest 10 stocks now making up 40 per cent of the value of the index.

Concerns about overexposure to a handful of big stocks led some investors to allocate instead to equal weighted funds, in which the smallest stocks are given the same weight as the biggest, limiting investors’ vulnerability to a sell-off in the largest companies.

But this equal weighted approach has badly lagged the main index in recent years and the performance gap between the two has grown to its biggest since 2003. That had caused investors to pull money out of funds like Invesco’s S&P 500 Equal Weight ETF.

In the past two weeks, however, the equal-weighted index has beaten the size-weighted index as some of the biggest names have suffered.

High valuations mean that “it looks attractive to go into the equal weighted index, shying away a bit from the tech space — or at least leaving it on the side for the time being,” said Wolf von Rotberg, equity strategist at J. Safra Sarasin Sustainable Asset Management.

Savita Subramanian, head of US equity and quantitative strategy at Bank of America, said she was bullish on the equal-weighted S&P 500 relative to the size-weighted index, partly because of stretched valuations in the biggest names.

“I would be shocked to see as much multiple expansion [in big tech stocks] from here as we’ve seen so far,” she said. At the same time, she pointed out that growing capital spending by the big AI firms could translate into a broadening out of the market rally, as sectors such as manufacturing start to enjoy the benefits.

“Suddenly you’ve got tech companies spending hundreds of billions of capex, and some of that goes to manufacturing,” said Subramanian. “That could impair megacap tech, and help the broader market. As you see that grow out into other areas, that might benefit the equal weighted index.”

As well as carrying a huge chunk of total value, the index’s biggest stocks are currently responsible for a record proportion of the index’s risk, as measured using volatility.

This is largely due to the biggest stocks today — such as Nvidia or Tesla — being more volatile than the megacap stocks of the past, such as banks or energy companies, said Laurent De Greef, head of portfolio strategy at D.E. Shaw Investment Management.

Such risk can offer bigger returns when conditions are good, but leaves investors more exposed to a downturn in the biggest stocks.

“With heightened valuations you also get heightened downside. Now is the time to get more selective on these megacap winners,” said Arun Sai, multi-asset portfolio manager at Pictet Asset Management.

“Equal-weighted exposure is a great way to hedge your bets in terms of where the current [economic] cycle is going.”

Is it time to sell your AI stocks?

Avoid the hype — most companies are seeing no return on generative AI spending

Managing investments has three principal parts: deciding which to own, which to ditch and — crucially — implementing those decisions.

Over the years I have heard too many tales of woe from investors who had decided what to do but had not got around to doing it until it was too late. 

Many global equity investors will be sitting on decent gains from the rise of the so-called Magnificent Seven stocks. In dollar terms the MSCI All World Index is up over 15 per cent annually over three years. A former colleague used to say that when investors get rich they often think they have become more intelligent. 

If they are intelligent, they will be looking nervously at those gains and asking if now is one of those moments to act. For me the best herald of danger is valuation.

Those megacap Mag 7 companies that have driven so much of this growth now make up more than 20 per cent of global equity indices. It’s worth examining which are flashing red on valuations. 

The ratio of price to earnings is a crude measure that shows how many years’ earnings it would take to pay for one share. By this measure, it would take nearly two centuries to pay for your Tesla shares.

Earnings — or profits — can be understated if a company is investing heavily for growth, building AI data centres, for example. Looking at the ratio of a company’s share price to its sales gives you a sense of how much these revenues must grow to generate the earnings the share price merits when that capex is done. Few shares trade at more than five times sales for long. Nvidia shares would either have to fall a lot or their sales grow at this speed for many years to come to sustain this valuation. 

Revenues are growing strongly in most, but this growth appears to be slowing. AI might improve profitability for a company such as Microsoft — but only if its customers value it sufficiently to merit paying more.

That is a gamble. I am not surprised by recent research showing that 95 per cent of organisations get zero return from their investments in generative AI. On the other hand, the application of AI to diagnostics and factory automation looks like it is raising productivity.

​I would not own Nvidia, Tesla or Alphabet now. If I held Apple, I probably would hang on to it as it is not an AI stock and it’s the AI hype I’m trying to avoid.

Beyond these megacap stocks, not everything related to AI is expensive. Taiwan Semiconductor still looks reasonable value, and we bought cyber security specialist Fortinet on its recent stumble.  

There are still attractive places to have your money. Take Thermo Fisher. This is the world leader in scientific equipment for medical research and diagnostics. It is priced at 21x earnings and 4x sales. Yaskawa makes the blue robots that have the largest share in semiconductor manufacturing and many other areas of factory automation. It is priced at 18x earnings and just 1.5x sales.

Both businesses have seen their growth hampered in recent years by the slowdown in Chinese demand, but this now seems to be moderating and even turning. These are the sorts of companies active fund managers tend to favour. 

Those most exposed to the hot stocks are investors in low-cost global equity trackers. These have outperformed most active funds, including my own. They may continue to rise. I have no fear of missing out.

Over the years I have built a record of selling too early. It means my portfolio will underperform the index when it gets frothy, as now. However, my priority is to ensure every investment I have is underpinned by value for money.

I have no idea when a correction might come. I also had no idea in 2000, but we sold most of our tech stocks anyway because the valuations told us to. It was the same selling British banks in 2007, when they were trading on a stretched multiple of book value. You didn’t need to know that a financial crisis was coming, you could avoid the worst of it by being disciplined on valuation, and we did.  

No great flare went up at the top to tell you to sell. You just had to follow your investment disciplines and not let timing questions get in the way.

Partnership with Midjourney marks shift away from internal product development

Meta will license technology from artificial intelligence image and video generation start-up Midjourney, as the social media group shifts towards working with third parties as its struggle to keep pace with rivals.

Alexandr Wang, Meta’s new chief AI officer, said in a post on X on Friday that the company planned to license Midjourney’s “aesthetic technology for our future models and products, bringing beauty to billions” in a “technical collaboration” between their research teams.

“To ensure Meta is able to deliver the best possible products for people it will require taking an all-of-the-above approach,” he added. “This means world-class talent, ambitious compute road map, and working with the best players across the industry.”

The tie-up will allow Meta to develop and integrate multimedia AI generation features into its apps, as chief executive Mark Zuckerberg has indicated that he expects AI-generated content to become more prominent on the platform.

The move comes as Zuckerberg pours billions of dollars into developing “superintelligence” that surpasses human intelligence. In recent months he has aggressively poached top AI researchers from competitors, doubled down on his investment in AI infrastructure and acquired AI voice company Play AI. Meta also took a stake in data labelling group Scale AI.

This week, Meta announced it was restructuring its AI group — recently renamed Meta Superintelligence Lab — into four distinct teams, the fourth overhaul in six months as it has struggled to solidify its organisational structure.

The Midjourney partnership marks a shift by Meta away from building all of its AI models and products in house, after its existing ones began to lag rivals. 

In 2024 Meta rolled out an image generation tool called Imagine, which allows users to generate images from text prompts. Last October it shared a research paper on a movie generation model, Movie Gen, that will generate and edit videos based on text prompts. Meta promised to integrate it fully into Instagram in 2025. 

However, the integration has yet to happen and industry insiders say the model already appears antiquated compared with Google’s Veo 3 and OpenAI’s Sora models, which have been released to consumers. 

The social media company had also abandoned plans to publicly release its flagship Behemoth large language model, according to people familiar with the matter, focusing instead on building new models.

Meta had started using third-party models internally for tasks such as coding, according to multiple people familiar with the matter, as faith in its Llama models has waned.

San Francisco-based Midjourney, founded in 2021 by David Holz, has become one of the most popular image generation companies, despite its chief executive refusing venture capital and instead opting to self fund. In June, it released its video model V1, which allows users to generate a short video from an existing image.

Holz said in a post on X on Friday that “bringing sublime tools of creation and beauty to billions of people is squarely within our mission”, adding that Midjourney remained an “independent, community-backed research lab, with no investors”.

The budget air carrier is on shaky financial ground after its recent reorganization fell short of providing a long-term fix

Spirit Airlines is exploring strategic alternatives after its recent financial restructuring failed to put the budget carrier on a sustainable path, people familiar with the matter said.

The budget airline has brought on financial adviser PJT Partners as it navigates a cash crunch that it previously said raises doubt about its ability to continue as a going concern. It is also working with consulting firms FTI and Seabury Airline Strategy Group.

Spirit is once again on unsteady financial footing. It emerged from bankruptcy in March having eliminated $795 million in debt in a deal with bondholders, who agreed to receive equity in exchange. The bankruptcy process didn’t sufficiently resolve problems such as high lease expenses.

Spirit on Thursday announced moves to shore up its position. It borrowed the full $275 million available under its revolver. It also extended a credit-card processing agreement for two years in a deal that allows U.S. Bank National Association to hold back up to $3 million a day.

The airline has said it is exploring ways to bring in cash such as selling planes, real estate or excess gate capacity. It also has new leaders in place who have restructuring and turnaround experience, including Chief Executive Dave Davis. The company also continues to work with the law firm Davis Polk & Wardwell, which served as counsel through the recent bankruptcy.

Spirit’s emergence from bankruptcy came at a tough time for the airline industry, which has struggled much of this year with tepid domestic travel demand and a glut of seat capacity that weighed on fares. Discounters like Spirit have faced particular challenges, and Spirit has been trying to attract new customers with higher-end offerings.

In court filings last year, the airline had projected a profit of $252 million for 2025, but it has lost $256 from mid-March through the end of June.

The airline was overly optimistic on its strategic turnaround coming out of bankruptcy and ran into a more difficult environment, said Fitch Ratings analyst Joe Rohlena. Fitch downgraded Spirit last week and said the carrier is “increasingly vulnerable to a default scenario in the near term.”

Moody’s Ratings, which downgraded Spirit this week, forecast the airline will burn through more than $500 million in cash this year, violating minimum liquidity covenants by year-end.

Trump’s reversal had sparked a buying spree from Chinese tech firms, who ordered a total of 700,000 H20 in the subsequent weeks, The Information reported last week.

Semi-finished chips—the individual processor dies ready for the final packaging step—are now piling up at Amkor, creating uncertainty about their future. Samsung also received notices from Nvidia that H20 chip production has been suspended, said one of the two people cited above.