WSJ : The Sudden Fall of OpenAI’s Most Hyped Product Since ChatGPT

The Sudden Fall of OpenAI’s Most Hyped Product Since ChatGPT
Even Disney’s Bob Iger signed on to the vision of allowing users to make their own videos starring Homer Simpson and Darth Vader. Then Sam Altman pulled the plug.

  • OpenAI abruptly shut down its Sora video-generation tools, which were consuming too many AI chips and were not profitable.
  • Disney’s $1 billion investment in OpenAI for Sora never materialized, and its relationship with the company is dormant.
  • OpenAI will redirect its focus to a new “superapp” with agentic AI tools and longer-term bets like robotics.

When Sam Altman arrived in Los Angeles to attend Vanity Fair’s Oscar afterparty earlier this month, his company was just weeks away from being able to license its Sora video-generation tools to Hollywood studios.

After the smash success of ChatGPT, Sora was hyped as AI’s next consumer-friendly frontier—a simple app that allowed users to put themselves and their friends in whatever video settings they choose, from dribbling like the Harlem Globetrotters to clashing lightsabers with Darth Vader.

Disney’s DIS -2.46%decrease; red down pointing triangle Bob Iger signed on to the vision, agreeing to have his company invest $1 billion in OpenAI and allowing the studio’s Marvel, Pixar and other characters to appear in Sora videos. Just as importantly, he put Disney’s valuable imprimatur behind the nascent technology amid widespread fears about protecting the industry’s creative work from AI.

Then OpenAI abruptly decided to shut Sora down.

Disney executives, many of whom learned about the decision less than an hour before it was announced, were shocked. What they didn’t know was that Sora had quietly turned into a liability for OpenAI in the months after its release, particularly as the startup tightened its focus ahead of a looming IPO.

OpenAI was weeks away from finishing work on a new AI model, code-named Spud, and needed to free up more computing resources to power the coding and enterprise products that would run on it. AI chips are the most precious commodity at any leading research lab, and at OpenAI, Sora was eating up far too many of them.

The product wasn’t profitable, and every user who spliced themselves into a World War II newsreel or Hollywood chase scene drew down a finite resource.

Sora now looks like an expensive strategic miscalculation, one that was led by key employees who were at the center of the AI talent war raging across Silicon Valley.

Altman represented the move as a difficult but necessary sacrifice toward the company’s larger goals, writing in a note to employees that he was encouraged by how they were willing to make “difficult trade-offs” for the good of the company.

The decision to kill Sora marked a stunning end to a project that Altman once dreamed would turn OpenAI into the creative pioneer of the AI era and could have been a lucrative new source of revenue.

The company first previewed Sora to the world two years ago, showing dreamlike landscapes conjured up by the technology that invoked the fantastical worlds of Hayao Miyazaki, or perhaps the surrealism of Salvador Dalí. When OpenAI launched a stand-alone Sora app for consumers last September, Altman likened it to the moment when the company first released ChatGPT.

But the app never took off in the ways its creators imagined—it was more AI slop than AI magic. Usage flatlined by the end of the year. With OpenAI’s purse strings tightening ahead of its IPO, company executives began taking a more critical look at Sora—and didn’t like what they saw.

The research team there was about to begin a training run for a new model meant to power video-generation in ChatGPT. Unlike language models, which learn from text, video models have to make sense of entire moving worlds, making them far more expensive to create. After running the numbers on how much it would cost, OpenAI decided to cancel it.

OpenAI expects to redirect its focus toward a new “superapp” the company is building that incorporates so-called agentic AI tools that can autonomously execute tasks for users, such as writing software, analyzing data and booking travel. Such productivity-focused products are becoming widely adopted in the workforce, and OpenAI has so far lagged behind its startup rival Anthropic in winning this market, imperiling its lead in the AI race.

Altman told staff the Sora team will now focus on longer-term bets such as robotics.

An OpenAI spokeswoman said the company is ruthlessly giving priority to its computing resources based on where it drives the most long-term economic value. “This disciplined focus on where we apply that compute allows us to grow, innovate faster, and deliver more efficiently to enterprises and developers,” she said.

Meta’s talent raid
Sora was the brainchild of Tim Brooks and Bill Peebles, two researchers who became close friends while completing doctorate degrees at the University of California, Berkeley. The pair joined OpenAI in early 2023, and set their sights on building models that could simulate the physical world by creating high-quality video from text.

They previewed their research to the public in February 2024, naming the system they developed after the Japanese word for sky. Sora transfixed the tech world by creating seemingly realistic videos of everything from woolly mammoths trekking across a snowy field to a stylish woman walking down a Tokyo street filled with glowing neon signs. Altman asked users on X to submit text descriptions for Sora before sharing their creations.

That December, OpenAI released Sora to the broader public.

The company housed Sora under its world simulation team, led by Aditya Ramesh. The division worked separately from OpenAI’s core research team, which built the language models that powered ChatGPT.

Last spring, Meta CEO Mark Zuckerberg launched a full-scale talent raid of OpenAI, personally reaching out to dozens of the company’s top researchers and luring them to join his new AI lab with giant pay packages. One of his targets was Peebles, who received an offer and briefly considered joining the company.

OpenAI managed to keep Peebles after giving him a raise, according to people familiar with the matter. Soon after, his responsibilities with Sora also expanded. Peebles oversaw training of a new video-generation model, and the creation of the Sora consumer app.

OpenAI’s researchers are able to track how AI chips are allocated between different groups through an internal dashboard. Some of them were surprised by the amount of computing resources the company gave to the Sora team, given that video-generation tools didn’t make much money, nor improve the capabilities of its language models.

Sora’s work was closely guarded from the rest of the company, leading some former employees to describe the project as a startup within a startup.

As the year progressed, there were growing signs that OpenAI was falling behind competitors in key parts of the AI race. Google’s Gemini grew popular with consumers. And Anthropic’s coding tool, Claude Code, was winning over software engineers across Silicon Valley thanks to its ability to write software programs without much oversight. OpenAI scrambled to release a new version of its own coding product, Codex.

But Altman wanted OpenAI to also be the AI company that used the technology to reshape popular culture and entertainment. In early 2025, he asked former Twitter CEO Parag Agrawal to informally consult a team inside OpenAI that was working on a separate social-media project similar to X, people familiar with the matter said. Altman also was working with then-Disney CEO Iger on a deal to enable the entertainment giant’s fans to bring their favorite characters to life with AI through Sora.

OpenAI previewed a version of the new Sora app to its employees in late September before releasing it to the public at the end of that month. The product garnered mixed reviews internally. Some employees felt launching a social-media app built around engagement would hurt the company’s brand. Others had concerns about the safety implications of allowing users to create AI-generated videos, even with guardrails.

Sora shot up to the top of the App Store in the week after its launch, despite an invitation-only user base. Users who made it found it to be a marvel: type in anything they wanted—Homer Simpson doing Riverdance—and a 10-second video of it would appear in a few minutes. And since the app allowed people to upload their own faces, they were suddenly making short, crazy films starring themselves. Altman himself volunteered his likeness, leading to absurd, sometimes violent or disturbing short films, that he didn’t seem to mind.

Since the copyright guardrails were fairly loose, videos soon appeared in the feed that pushed the boundaries of taste. A big featured player was Martin Luther King Jr., who could be instructed to share his dream of anything, from a new season of Fortnite to changing Sora’s content-violation policy. King’s estate complained about the images, leading the platform to announce that it had removed his likeness.

The worldwide user count peaked at roughly a million soon after the app’s launch, but never reached that level again. In the subsequent months, it dwindled to less than 500,000, according to data from Similarweb.

Sora was losing roughly a million dollars a day, according to a person familiar with the matter.

Disney’s AI dream
Still, OpenAI tried to find a way to make Sora work. In December, it announced a multiyear deal with Disney to license more than 200 characters from the entertainment giant’s cinematic universe. As part of the deal, Disney agreed to become a major customer of OpenAI and invest $1 billion into the startup.

In a CNBC interview, Iger said the deal gave Disney the opportunity to play a part in the fast growth of AI and new forms of media and entertainment. Altman said he hoped the partnership would give users a new way to be creative with AI.

For Disney, the deal represented proof that there was a business model for licensing its intellectual property for the use of AI. The day before it announced its deal with OpenAI, Disney sent a cease-and-desist letter to Google accusing the tech giant of “infringing Disney’s copyrights on a massive scale.”

In February, Iger said on an earnings call that short-form videos created with Sora would soon appear on the streaming service Disney+, which was preparing to launch a vertical video feed. Disney was also in talks with OpenAI to use ChatGPT across the company, according to people familiar with the matter.

In recent weeks, OpenAI began piloting an enterprise version of Sora for companies like Disney to use with safeguards.

The tool, which Disney expected to introduce as soon as this spring, would have allowed select Disney executives to use Sora for everything from designing marketing campaigns to special effects without giving OpenAI access to the work.

But OpenAI was already looking at ways to pull back from Sora. The company fell further behind Anthropic, whose recent advancements stirred fresh fears that AI could supplant traditional software and services, briefly triggering a stock-market selloff. OpenAI realized it needed to put more resources into building these so-called productivity tools, and began looking at areas to deprioritize.

After initially planning to continue selling video-generation features through ChatGPT, the company decided to cut Sora altogether.

Disney’s $1 billion investment in OpenAI never went through, and its relationship with the company is effectively dormant.

Under new CEO Josh D’Amaro, Disney is in active discussions with more than a dozen partners about ways to implement other AI tools, according to a person familiar with the situation.

“As the nascent AI field advances rapidly, we respect OpenAI’s decision to exit from the video generation business and to shift its priorities elsewhere,” Disney said in a statement. “We appreciate the constructive collaboration between our teams and what we learned from it.”

On the X account for Sora, the team posted what felt like a digital eulogy. “To everyone who created with Sora, shared it, and built community around it: thank you.”

WSJ : Iran’s Wealth Is Parked on London’s Billionaires’ Row

Iran’s Wealth Is Parked on London’s Billionaires’ Row
Years of Western sanctions haven’t prevented money flows out of Tehran: ‘They probably learned from the Russian oligarchs’

On a leafy residential stretch of North London dubbed Billionaires’ Row, Saudi royals and wealthy Russians have long come shopping for high-end real estate. The Iranians were a more surprising addition to the neighborhood.

Several sprawling mansion plots along the thoroughfare more formally known as The Bishops Avenue make up a chunk of the extensive London property holdings funded by Iranian money that Western officials have linked to the country’s Islamic Revolutionary Guard Corps.

Iranian banker Ali Ansari acquired the Bishops Avenue plots and other nearby property in 2018 for around 90 million pounds, equivalent to around $120 million today, according to people familiar with the deal. The transaction took place offshore through an entity registered in the Isle of Man, the people said.

Now frozen by British sanctions, the properties have fallen into disrepair, the overgrown lots a blight on an upscale street lined with manicured gardens and iron gates. Their desolation contrasts with hives of construction activity nearby, where cranes tower over designer redevelopments.

But they are nonetheless an illustration of how Iran has been able to move money internationally for years despite Western sanctions aimed at choking off the regime’s ability to do so.


The Middle East war has heightened scrutiny on the global financial networks alleged to fund the Iranian regime’s military, domestic oppression and foreign proxies. It is a web that Western officials say stretches from the United Arab Emirates to Toronto, Hong Kong and Singapore.

London has emerged a key part of that nexus. The proliferation of offshore vehicles, combined with a laissez-faire attitude toward sources of foreign wealth, have made it an attractive place for well-to-do Iranians to park cash, real-estate experts and wealth managers say.

“You name a jurisdiction where people have earned a lot of money through questionable means, it ends up in London property,” said Steve Goodrich, who leads research and investigations at Transparency International UK, focused on rooting out corruption.

London is hardly the lone destination for Iran’s dirty money.

The U.S. Treasury’s financial-crimes enforcement arm said last year it identified some $9 billion in potentially illicit Iranian money passing through foreign accounts at U.S. banks in 2024 alone, much of it oil and gas proceeds routed through accounts and shell companies in Asia and the Middle East. The U.S. has highlighted Iranian cash flows it says support arms purchases and bolster the ruling regime. Officials also cite outflows of personal fortunes belonging to connected Iranians, described as the spoils of classic kleptocracy.

Cryptocurrency analysts say they have tracked billions in Iran-linked outflows in recent years. In February, Treasury Secretary Scott Bessent told lawmakers the U.S. had seen Iranian leaders “wiring money out of the country like crazy.”

London has stables of lawyers and accountants experienced at disguising sources of wealth and who help clients obtain visas that grease the wheels of business.

British anticrime officials and the financial-sector watchdogs estimated in 2024 that more than £100 billion in illicit money is laundered inside or through the U.K. every year. Government officials pledged to boost barriers to dirty money following Russia’s 2022 invasion of Ukraine.

Ansari’s real-estate portfolio in the British capital includes a North London mansion that U.K. property records show cost him £33.7 million in December 2014. His name is also attached to two luxury apartments in London’s high-end Kensington neighborhood, purchased in 2014 and 2016 for a combined £36 million.

The apartments are just steps from the Israeli Embassy and the nearby royal residence at Kensington Palace. The proximity has prompted a spate of British media reports about security in the area and the potential for Iran to have monitored consular activities from close range.

Ansari’s close working relationship with senior regime officials “is a well-known secret in Iran,” said Farzin Nadimi, senior fellow with the Washington Institute for Near East Policy, who has studied Iran’s military and international flows of regime money.

The layers of shell companies and use of foreign conduits into luxury real estate help leaders of the Islamic Republic maintain an image back home of embracing modesty and rejecting lavish lifestyles, researchers say.


As for where the regime has chosen to move money, Nadimi said, “They probably learned from the Russian oligarchs that investing in London real estate was a good idea.”

London assets bearing Ansari’s name were frozen when the U.K. imposed sanctions in October 2025. British officials accused Ansari of financially supporting Iran’s Revolutionary Guard, the paramilitary group that protects the regime and serves as its dominant political and economic entity. A British minister in a government statement called Ansari a “corrupt Iranian banker and businessman.”

Ansari was the dominant shareholder in Iran’s Ayandeh Bank, a private lender that collapsed in late 2025. The failure helped trigger massive protests that turned deadly. He hasn’t been sanctioned in the U.S.

Ayandeh was run by regime cronies, The Wall Street Journal earlier reported, and sank under a multibillion-dollar pile of bad loans. In a statement in October, Ansari blamed the bank’s failure on “decisions and policies made beyond the bank’s control.”

Ansari, through a London lawyer, said he plans to challenge the U.K. sanctions and denied any financial relationship with the Revolutionary Guard. The lawyer also denied reporting by Bloomberg in January describing Ansari as an established money man working for Mojtaba Khamenei—who became Iran’s new supreme leader earlier this month—helping move oil proceeds and other business profits abroad.

Khamenei is the second-eldest son of Ayatollah Ali Khamenei, the previous supreme leader, who was killed on the first day of the war in February. The newly elevated son was sanctioned by the U.S. in 2019.

Elsewhere, U.K. and overseas entities have been used to route Iran-linked cash into the European market, including into high-end hotels in Germany and Spain, corporate filings show.

A few minutes’ drive from Ansari’s Kensington apartments, a boutique London hotel called the Gainsborough Hotel sold for £6.5 million in 2018 to an entity controlled by Salim Ahmed Said, according to U.K. property records. The hotel, a short walk from world-renowned museums, is formed of conjoined Victorian townhouses with British and Irish flags displayed out front.

Said is an Iraqi-British national who since at least 2020 has operated a network of companies helping Iran dodge international sanctions on sales of oil, according to U.S. Treasury Department sanctions imposed in July 2025.

Robinbest Limited, a U.S.-sanctioned entity, is the Gainsborough’s registered owner and directly linked to Said, according to U.S. sanctions and U.K. property records. The records list a villa in Dubai’s Palm Jumeirah as his address.

Said, 48 years old, holds two British passports, one that expired this month and a second expiring in 2029, and has used several aliases, according to the U.S. Treasury. He couldn’t be reached for comment. Said has previously said his oil trades were legitimate.

Ansari, 57 years old, was born in Ghazvin, Iran, and holds four passports, two from Iran, one from Cyprus and the other from the tiny Caribbean nation of St. Kitts and Nevis, a former British colony, according to an October U.K. sanctions filing. It lists Dubai as a primary address.

Ansari identified himself as Iranian and provided a Tehran address on early U.K. corporate filings, in 2007. Years later on filings tied to the mansion plots on Billionaires’ Row, he listed his nationality as Cypriot.

As a result of the recent sanctions, a developer who worked with both Ansari and previous owners of the North London properties has gone unpaid for work he did for the Iranian, according to people familiar with the relationship.

The plots sit vacant, with faded signs beside locked gates and security signs warning “Guard dogs in use. Trespassers will be prosecuted.”

FT : Donald Trump says US could ‘take the oil in Iran’

Donald Trump says US could ‘take the oil in Iran’
US president tells the FT he is considering seizing strategic Kharg Island even as negotiations continue

Donald Trump has said he wants to “take the oil in Iran” and could seize the export hub of Kharg Island, as the US sends thousands of troops to the Middle East.

The US president told the FT in an interview on Sunday that his “preference would be to take the oil”, comparing the potential move to Venezuela where the US intends to control the oil industry “indefinitely” following its capture of strongman leader Nicolás Maduro in January.

The president’s comments come as the US-Israeli war against Iran has thrust the Middle East into crisis and sent the price of oil surging by more than 50 per cent in a month. Brent crude rose above $116 a barrel on Monday morning in Asia, near its highest level since the conflict began.

Trump said: “To be honest with you, my favourite thing is to take the oil in Iran but some stupid people back in the US say: ‘why are you doing that?’ But they’re stupid people.”

Such a move would involve seizing Kharg Island through which most of Iran’s oil is exported.

Trump has been beefing up US forces in the region, with the Pentagon ordering the deployment of 10,000 troops trained to seize and hold land. About 3,500 troops arrived in the region on Friday, including roughly 2,200 Marines. Another 2,200 Marines are en route, while thousands of troops from the 82nd Airborne Division have also been ordered to the region.

But an assault on the export hub would be risky, raising the chances of more US casualties and extending the cost and duration of the war.

“Maybe we take Kharg Island, maybe we don’t. We have a lot of options,” Trump told the FT. “It would also mean we had to be there [in Kharg Island] for a while.”

Asked about the state of Iranian defence on Kharg Island he said: “I don’t think they have any defence. We could take it very easily.” 

The conflict has broadened in recent days, with an attack on an air base in Saudi Arabia on Friday wounding 12 American troops and damaging a $270mn US E-3 Sentry surveillance aircraft. Houthi rebels in Yemen also fired a ballistic missile at Israel, threatening a new phase of escalation that analysts said could worsen the global energy crisis. 

However, despite his threats to seize Iranian oil production, Trump stressed that indirect talks between the US and Iran via Pakistani “emissaries” were progressing well. Trump has set a deadline of April 6 for Iran to accept a deal ending the war or face US strikes on its energy sector.

When asked whether a ceasefire deal could be reached in the coming days that would reopen the Strait of Hormuz, the waterway through which a fifth of the world’s oil usually flows, Trump declined to offer specific details.

“We’ve got about 3,000 targets left — we’ve bombed 13,000 targets — and another couple of thousand targets to go,” he said. “A deal could be made fairly quickly.”

Last week, he said that Iran had permitted 10 Pakistan-flagged oil tankers through the Strait of Hormuz as a “present” to the White House. The number of tankers had now been doubled to 20, he told the FT, which was not possible to immediately verify.

“They gave us 10,” he said. “Now they’re giving 20 and the 20 have already started and they’re going right up the middle of the Strait.”

Trump added that Mohammad Bagher Ghalibaf, Iran’s parliamentary speaker and one of the country’s top wartime leaders, had authorised the additional tankers.

“He’s the one who authorised the ships to me,” Trump said. “Remember I said they’re giving me a present? And everyone said: ‘What’s the present? Bullshit.’ When they heard about that they kept their mouth shut and the negotiations are going very well.”

Trump also claimed that Iran had already had “regime change” after Iran’s longtime supreme leader Ayatollah Ali Khamenei and many other senior officials were killed at the start of the war and in strikes that followed.

“The people we’re dealing with are a totally different group of people . . . [They] are very professional,” Trump said.

Trump also reiterated his claims that Mojtaba Khameini, Khamenei’s son and Iran’s new supreme leader, could be either dead or severely injured.

“The son is either dead or in extremely bad shape,” Trump said. “We’ve not heard from him at all. He’s gone.”

Tehran has insisted the head of state is safe and well after his absence from the public eye fuelled speculation he had been badly injured.

FT : Distressed-debt funds target private credit downturn as ‘greatest opportuni

Distressed-debt funds target private credit downturn as ‘greatest opportunity’ since 2008
Investors anticipate money-making bonanza as sector comes under strain

Investors who specialise in scooping up distressed assets at bargain prices have identified a downturn in private credit as their best opportunity since the 2008 financial crisis.

These funds, which typically invest in companies with bad balance sheets but viable underlying businesses, have been largely sidelined for a decade as markets surged but are now betting on making money from strains in private credit.

“Biggest opportunity since 2008,” said Victor Khosla, founder of Strategic Value Partners, which manages $21bn in assets.

Andrew Milgram, founder of Greenwich-based Marblegate Asset Management, said: “This is not about a few bad loans . . . We’re out in the market right now raising a new fund because this is the greatest opportunity I’ve ever seen in my lifetime. I couldn’t imagine God would smile on me like this.”

Private credit has become one of Wall Street’s top worries this year, as several funds, managed by the likes of Apollo Global Management, Blackstone and Ares, have faced billions of dollars in redemptions amid questions about their exposure to software companies at risk of losing out to AI.

“These outflows have reached a tipping point, whereby everybody on a rational basis has to ask for their money back,” said John Aylward, the founder of Sona Asset Management. “You have a large amount of distress, and you have forced selling, and it’s going to provide great opportunities that we’re already seeing.”

The proportion of US leveraged loans with an interest coverage ratio — a measure of ability to service debt — weaker than a level that is considered stressed has more than doubled to 20 per cent since 2019, according to recent research by credit investor Davidson Kempner.

At the same time, more borrowers are opting to defer repayments and increase their loan balance. This combination has led some investors and analysts to argue the true corporate default rate is far higher than reported.

Distressed investors bristle at being called “vulture funds” — a sobriquet that gained traction in the 1980s for hedge funds that swoop in when companies run into trouble — arguing they are far more sophisticated than the industry’s early bottom feeders. Many are no longer purely distressed investors, having diversified into buying higher-quality credit and equity.

Milgram at Marblegate said he and his team go on “regional swings” — short trips to cities such as Cincinnati and Charlotte to get a sense of the mood there. He takes local bankers out to steak dinners to learn what keeps them up at night.

For the past several months, the picture has grown increasingly grim. Milgram said the head of restructuring at a large regional bank recently told him private equity firms were abandoning deals and handing over portfolio companies at an “alarming rate”.

SVP specialises in hard assets such as Texas toll roads and Europe’s largest car park business. But recently Khosla directed a small team to study software. “We can’t for the life of us figure out who the losers and winners will be yet,” he said. “It’s too early.”

Since the start of 2025, the firm has invested $3.8bn but sold assets worth more than double that amount, far more than usual, suggesting it wants to have cash reserves as distressed opportunities emerge.

“Our view is that the [distressed] situations will overwhelm the amount of dry powder that’s out there,” said David Walch, partner and co-portfolio manager at asset manager King Street, referring to available capital.

Even the biggest players in private credit are preparing for the worst. Apollo’s chief executive, Marc Rowan, told investors in December that he needed to position the firm to make money “when something bad happens”.

This is not the first time in recent memory distressed investors have predicted a downturn. When Silicon Valley Bank imploded in 2023, there was a similar buzz that many more companies might buckle under the fatal combination of high interest rates and heavy debt loads.

Yet that wave of failures never came about. One executive of a top private capital firm said distressed investors were trying to drum up excitement.

“Hedge fund managers . . . have to create a sense of like, ‘the house is on fire, the house is on fire’. What they’re trying to do is to get banks to pull [credit] lines. They’re trying to create a frenzy because otherwise it’s going to be like watching paint dry,” they added.

(ZeroHedge) Subprime Crisis 2.0: Will Private Credit Be The Trigger?

Subprime Crisis 2.0: Will Private Credit Be The Trigger?

After 30 years of watching credit cycles expand, distort, and collapse, I’ve learned one reliable rule:
“When enough people start drawing comparisons to 2008, it’s worth stopping to check whether the analogy holds up — or whether fear is doing the analytical work for them.”
Right now, judging by the amount of commentary on social media, the stress in the private credit market has everyone’s attention. Most of the commentary being generated makes the immediate jump from private credit firms “gating” exits to the onset of the next subprime crisis in the financial system. Those claims are certainly alarming and generate many clicks and views, but the question is whether those claims are based on facts rather than opinions.
Just recently, Goldman Sachs CEO David Solomon flagged the risk of private credit in his annual shareholder letter. Lloyd Blankfein, who piloted Goldman through the Global Financial Crisis, warned publicly that the financial system appears to be “inching toward another potential catastrophe.” Meanwhile, Goldman’s own research arm published a note concluding that private credit stress is “unlikely to generate large macroeconomic spillovers on its own.”
So which is it? A repeat of the subprime crisis of 2008, or a painful but contained credit cycle? The honest answer most likely sits somewhere in between, and understanding exactly where private credit differs from subprime tells you a great deal about how worried you should actually be.
Let’s revisit 2008.
What Made The Subprime Crisis So Catastrophic
It is hard to believe that we are rapidly approaching the 20-year anniversary of the “Great Financial Crisis” that nearly destroyed the financial system as we knew it. There are many investors and commentators in the markets today who only know about the event from reading history books. Having lived through it, it is a different reality.
Crucially, the 2008 subprime crisis wasn’t simply a mortgage problem. It was a leverage-and-derivatives problem that started in mortgages. That distinction matters enormously when you’re sizing up today’s private credit stress.
At the heart of the crisis was a product called the collateralized debt obligation, or CDO. Banks packaged pools of subprime mortgages into tranches, which were rated by agencies using flawed models. Those CDOs were then re-sliced into “CDO squared” structures, layering additional complexity and opacity on top of already opaque assets. The real acceleration came when synthetic CDOs entered the picture. Unlike cash CDOs, which required actual mortgages, synthetic CDOs referenced mortgages through credit default swaps. Journalist Gregory Zuckerman found that while roughly $1.2 trillion in subprime loans existed in 2006, synthetic structures created more than $5 trillion in exposure referencing those same loans. The CDS market alone reached a peak notional value of $62.2 trillion by year-end 2007. That is not a typo.
But the derivatives machine required raw material to function, and Wall Street’s insatiable hunger for collateral triggered what historians of the crisis now call the “race to the bottom” in mortgage underwriting. To keep the CDO assembly line running, originators needed volume. That demand for volume led to a collapse in underwriting standards. By 2006, no-money-down mortgages were commonplace.
  • NINJA loans, “No Income, No Job, No Assets,” were extended to borrowers who could not remotely service the debt once introductory teaser rates reset.
  • Stated-income loans, in which borrowers self-reported earnings with no verification, became the industry norm rather than the exception.
  • Adjustable-rate mortgages were sold to buyers who qualified only at the teaser rate and had no capacity to absorb resets of 3 to 4 percentage points two years later.
The Mortgage Bankers Association later estimated that subprime originations reached $600 billion in 2006 alone, up from roughly $160 billion in 2001. Most importantly, the loans were designed to be sold, not held. In other words, the originator of the loan bore no long-term risk and had every incentive to close as many transactions as possible, regardless of quality.
That single misalignment of incentives was the original sin of the entire subprime crisis.
What compounded the damage beyond even that was systematic, institutionalized fraud at the origination and securitization level. The Financial Crisis Inquiry Commission documented widespread “robo-signing,” where bank employees executed thousands of mortgage documents per day without reviewing them. They affixed signatures and notarizations to paperwork they had never read. Countrywide Financial, Washington Mutual, and others were found to have misrepresented loan quality in the representations and warranties they made to investors purchasing MBS tranches, fraudulently inflating the apparent collateral quality of the pools they sold.
Appraisers faced pressure, and in many cases direct financial incentive, to hit predetermined valuations that supported loan amounts the underlying properties could never justify. The FBI reported that mortgage fraud suspicious activity reports increased by more than 1,400% between 2000 and 2007. When losses eventually surfaced, investors discovered they had purchased securities backed not just by bad loans, but by fraudulently documented ones. That distinction made recovery values nearly impossible to model and turned settlement litigation into an industry unto itself for a decade afterward. JPMorgan alone paid $13 billion in 2013 to resolve government claims over mortgage securities, and that figure represented only a fraction of industry-wide settlements.
When housing prices began falling, that entire structure detonated in both directions simultaneously. Banks that held CDO tranches faced mark-to-market losses. Banks that sold CDS protection, AIG being the most famous, faced collateral calls they couldn’t meet. Here is the most crucial point. These instruments traded freely in liquid markets, so price discovery occurred in real time, compressing the panic into a matter of weeks. The interconnection was total. Twelve of the thirteen largest U.S. financial institutions were at risk of failure, according to then-Fed Chair Ben Bernanke.
That’s what systemic risk actually looks like.
Private Credit Stress Is A Different Animal
The private credit market now stands at roughly $1.7 to $2 trillion in deployed capital, a figure that has grown rapidly since banks retreated from middle-market lending after the Global Financial Crisis. That growth is precisely what generated the current stress. Redemption requests have surged across major platforms. Blackstone’s BCRED fund saw record redemptions of $3.8 billion in Q1 2026, exceeding its 5% quarterly buyback limit. Apollo, Blue Owl, and Morgan Stanley’s North Haven fund have all imposed withdrawal restrictions. That gating of withdrawals led to an obvious decline in inflows across retail private credit funds. Those inflows fell to roughly half their 2025 pace, according to Goldman Sachs estimates.
So far, the catalyst is concentrated in software companies, which represent an estimated 15% to 25% of many private credit portfolios. They are under pressure as AI disruption fears potentially erode their earnings power and their ability to service debt. The headline default rate sits around 2% as of 2025, but Goldman Sachs Asset Management’s own research acknowledges that figure understates the true level of stress. When you include liability management exercises and distressed exchanges, the real rate approaches 4% to 5%. That’s meaningful deterioration. It’s not catastrophic, but it’s real.
J.P. Morgan’s analysis showed that for senior direct lending to produce negative total returns, default rates would need to exceed 6% while recovery rates would collapse below 40% simultaneously. Those numbers have historically appeared only during COVID and the Global Financial Crisis itself. That’s a high bar — but it’s not an impossible one. However, that would require a deterioration in macroeconomic conditions, a continuation of the Iran conflict oil shocks, and a contraction of consumer spending, which could certainly amplify risks. As shown below, the current structural comparison between the subprime crisis and the private credit sector today is markedly different.
The Importance of the Gating System
The most structurally significant difference between 2008 and today is also the one that generates the most debate. Unlike the subprime crisis, private credit funds can gate their exits. When Blackstone caps BCRED redemptions at 5% per quarter, it’s not a failure of the fund; it’s the mechanism working as designed. In 2008, there was no such circuit breaker. MBS and CDOs traded continuously in secondary markets, meaning every forced seller found a bid at a lower price, triggering more mark-to-market losses, which in turn triggered more forced selling. The feedback loop was instantaneous and brutal.
Gating slows that process considerably. LPL Research noted that while gating makes for terrible headlines, it prevents the forced liquidation that accelerated subprime losses. Goldman Sachs estimates that retail private credit inflows will remain in net outflow throughout 2026 and likely into 2027, a slow bleed, not a cliff. That’s a very different contagion profile.
That said, gating is not a cure. It transfers the problem in time, not away from investors. Those sitting in redemption queues face a multi-year wait to exit positions that may continue to deteriorate. The opacity of private credit portfolios and manager-reported valuations means stress can accumulate invisibly until it can’t.
“The key risk in private credit is not what is visible, but what remains hidden.” – The Daily Economy
Goldman Sachs economist Manuel Abecasis concluded that, even in an adverse scenario, private credit stress would only drag on GDP by 0.2% to 0.5%. His reasoning is straightforward: the private credit sector holds about $1.7 trillion in levered loans, or roughly 4% of all credit to the private non-financial sector. That’s is not nothing, but it’s not the $62 trillion CDS market either. Goldman also notes that bank lending to businesses has actually accelerated recently, providing a partial offset if private credit tightens.
Blankfein’s view carries different weight precisely because he’s been through the real thing. He warned that private credit assets “can be hard to analyze, may feature hidden leverage, and can become tough to sell.” He’s right that opacity and illiquidity create conditions where problems compound before they surface. The question is whether those conditions, combined with a still-manageable scale, produce systemic contagion or simply painful losses for a subset of investors.
“Private credit stress is unlikely to generate large macroeconomic spillovers on its own.” — Goldman Sachs Economist Manuel Abecasis, March 2026
I’m inclined to side with Goldman’s macro conclusion. However, with a caveat that matters. The base case holds only so long as private credit problems don’t compound with a broader recession, a sustained oil shock from the Iran conflict, and a sharper-than-expected deterioration in software company cash flows. Any two of those three conditions occurring simultaneously change the calculus. Goldman’s own research acknowledges this. The bigger risk isn’tprivate credit alone. It’s private credit stress coinciding with the wider tightening of financial conditions.
What Investors Should Pay Attention To
The structural differences between today and the subprime crisis are real and important. There’s no synthetic subprime CDO chain multiplying private credit losses to a $5 trillion notional exposure. Most critically, the investor base is primarily institutional, not retail money market funds holding fraudulently rated paper. Fund-level leverage is modest, and the gating mechanism, whatever its imperfections, prevents the instantaneous price cascade that made the subprime crisis so destructive.
What this most closely resembles is a normal credit cycle playing out in an untested asset class. Not a systemic collapse, but not a benign correction either. Goldman Sachs Asset Management’s own European research found that “stress events are likely to remain elevated relative to the last decade,” concentrated in smaller companies and cyclical sectors. That pattern will probably hold in the U.S. as well.
Three things would change my view and warrant genuine alarm.
  • First, if default rates push past 8% in tech-heavy private credit portfolios as AI disruption accelerates.
  • Second, if bank credit facilities to private credit managers get pulled at scale, triggering forced asset sales.
  • Third, retail penetration of private credit grows, as institutional investors sell, leaving less-sophisticated money to hold the bag.
None of those conditions is inevitable. All of them are possible.
The subprime crisis analogy fails on the specifics. But the lesson from the subprime crisis isn’t about CDOs. It’s about what happens when credit markets expand rapidly, underwriting discipline erodes under competitive pressure, and opacity masks deteriorating loan quality. On those broader conditions, the warning is more relevant than the Goldman bulls would like to admit.
That is why we continue to underweight risk for now until we have better clarity about the future.
Key Catalysts Next Week
This is the most structurally loaded week of the quarter. The calendar stacks a Q1 close, a Q2 open, and a full employment gauntlet into five sessions, with markets still metabolizing whatever the Fed just delivered..
Tuesday is the pivot. Consumer Confidence is the marquee release, and it’s the first full-month reading that captures the Iran conflict, the tariff widening, and February’s payroll shock in a single survey. The prior print of 91.2 was already soft. The Expectations component, which the Conference Board flags as a recession signal below 80, is the number to watch. A sharp drop would validate the stagflation fears the Fed just tried to navigate around. Chicago PMI and Case-Shiller Home Prices round out the morning, and then Q1 closes at the bell. Expect elevated volume as pension funds and mutual funds finalize window dressing and mark final positions, totaling roughly $62 billion on the buy side.
Wednesday flips the calendar to Q2 and immediately delivers a triple shot: ADP private payrolls, ISM Manufacturing, and JOLTS. After February’s -92,000 NFP shock, the ADP print will either stabilize the labor narrative or accelerate the deterioration thesis. ISM Manufacturing is the tariff passthrough read, the Prices Paid subindex will tell us whether producers are eating costs or passing them through, while New Orders reveal whether demand is contracting under policy uncertainty. JOLTS completes the picture with the openings-to-unemployed ratio that the Fed uses to assess labor market slack.
Friday is the week’s anchor: March Nonfarm Payrolls. February was distorted by a Kaiser Permanente strike and severe weather, giving bulls a one-month excuse. If March payrolls bounce back above 100,000, the “transitory weakness” camp wins. If they print flat or negative again, the labor market deterioration becomes undeniable, and the pressure on the Fed to act, despite sticky inflation, becomes immense. ISM Services PMI that morning adds the services-sector inflation read alongside Wednesday’s manufacturing data.

WSJ : Why the Cost of Your Coffee Has Soared—and Isn’t Going Down Soon

Why the Cost of Your Coffee Has Soared—and Isn’t Going Down Soon
One roaster’s ride on the roller coaster of coffee pricing helps explain the many reasons consumers keep paying more for a cup of joe

Fans of Reverie Roasters’ Boneshaker Espresso blend are familiar with the sticker shock that has hit coffee prices nationwide.

For years, a 12-ounce bag of the bestseller cost $15 at its Kansas roastery and two coffee shops. As of last spring, it is $17. And Reverie’s owner, Andrew Gough, is raising the price again next month, to $18.


That reflects just a share of the often-hidden factors driving up the cost of doing business for artisanal purveyors like Reverie. Tariffs inflated his expenses, but so have crop failures, higher rents and rising labor costs.

Now, after commodity coffee prices dipped in recent months, concerns about the Iran war and stepped-up trading in coffee futures markets are driving them up again.

To drill down into why retail coffee prices have risen—and keep going up—The Wall Street Journal took a deep dive into Reverie Roasters’ expenses. Gough says he has absorbed roughly half the cost surge because passing it to consumers is risky, too.

“You always worry that if you raise your prices you are going to lose a customer,” Gough said.

Gough founded Reverie Roasters in Wichita, Kan., in 2013. In addition to its shop and online sales, it sells coffee to about 160 wholesale clients like churches, offices, schools and grocery stores. For more than decade, he didn’t have to think much about the price of the coffee.

But when it came time to order beans early last year, he got a shock: The unroasted green coffee Reverie buys had jumped to $4.30 a pound from $2.41 just a few months before. That cost includes the price of the beans themselves, plus freight, insurance and any duties.


That alone would have meant a nearly $200,000 cost increase for the roughly 95,000 pounds of coffee Reverie roasts annually. “It kind of freaked us out,” Gough said.

Behind the jump: Extreme weather, including droughts in Brazil and Vietnam, had hit coffee crops. And even before any tariff increases, hedge-fund bets anticipating the levies were pushing commodity prices higher.

Then in July, President Trump slapped an additional 40% tariff on goods from Brazil, which produces more than a third of the world’s coffee, citing legal action against its former President Jair Bolsonaro and U.S. tech firms as justification. Altogether, Gough says, he paid a little over $14,000 in tariffs in 2025.

Across the country, retail coffee prices soared as many coffee sellers passed on costs to customers. Coffee outpaced other grocery items in the inflation-tracking consumer-price index last year.


In hindsight, Gough said, he didn’t raise prices as much or as fast as he should have. He opened an account with the aim of stashing away $50,000 to eventually buy futures contracts and lock in cheaper, longer-term prices, as bigger roasters do. But the account now has $5.02.

As for his cash flow? “It’s been spiraling downwards every week,” Gough said.

Coffee commodity prices began sliding again in November after Trump pulled back the 40% tariff on Brazilian food items, including coffee, to address affordability issues. But Reverie’s green-coffee costs have risen even higher than they were in the early fall.

Small roasters don’t have the pricing power to buy coffee in massive volumes like the Starbucks of the world. Gough said he has to buy coffee at the price available to him in the moment, and Reverie isn’t big enough to hire a dedicated buyer who can scour for deals.

Another reason commodity prices are surging again? The same speculative trading that contributed to price surges early last year is ramping up again.

Back then, traders piled into coffee contracts—betting the price would go up—after an Agriculture Department report warned of a coffee-supply shortage. By February 2025, hedge funds controlled about a third of all coffee contracts—a $10.4 billion bet that helped fuel a surge in the C-price.

As a result, commodity prices have “just been roller coaster stupid,” said Emory University professor Peter Roberts, who developed the Specialty Coffee Retail Price Index. The index tracks prices for roasted coffee sold by about 60 companies in North America, including Starbucks, Stumptown and Peet’s. “Hedge funds are just liking the gamed uncertainty.”


Hedge funds sold off a big chunk of their positions early in 2026 after Brazilian authorities said they expected a record year in coffee production. Now, surging fuel and freight costs caused by the closure of the Strait of Hormuz have triggered a new round of coffee-contract purchases and rise in commodity coffee prices.

Gough says those fluctuations are too much to keep on top of when he’s running every aspect of Reverie’s operations.

“There’s too many variables for us to follow,” he said. “I’ve got to fix the toilet. I’ve got to do the broken doorknob.”