TheChek News Flow

The National Transportation Safety Board (NTSB) has opened an investigation into Waymo after its robotaxis have been spotted illegally passing stopped school buses numerous times in at least two states.

The NTSB is specifically focusing on the more than 20 incidents that have occurred in Austin, Texas, it said in a post on X on Friday.

“Investigators will travel to Austin to gather information on a series of incidents in which the automated vehicles failed to stop for loading or unloading students,” the NTSB said in a statement to TechCrunch. A preliminary report is expected within 30 days, and the safety board will publish a more detailed final report in 12 to 24 months.

It’s the first time Waymo has been investigated by the NTSB, but it’s the second investigation launched into Waymo over its school bus problem. The National Highway Traffic Safety Administration’s (NHTSA) Office of Defects Investigation opened a similar probe in October.

Waymo also issued a software recall last year to address the problem. But previous software updates have not been enough to stamp it out, and in Austin, Texas — where the bulk of incidents have been captured on camera — the school district has asked the company to suspend operations during pickup and drop-off times.

The new investigation comes as Waymo is in the middle of a rapid expansion around the United States. Just this week, the company started offering a robotaxi service in Miami, adding to its operations in Atlanta, Austin, Los Angeles, Phoenix, and the San Francisco Bay Area.

“We safely navigate thousands of school bus encounters weekly across the United States, and the Waymo Driver is continuously improving. There have been no collisions in the events in question, and we are confident that our safety performance around school buses is superior to human drivers,” Mauricio Peña, Waymo’s chief safety officer, said in a statement to TechCrunch. “We see this as an opportunity to provide the NTSB with transparent insights into our safety-first approach.”

The NTSB is different from the NHTSA in that it is not a federal regulatory agency. It cannot issue fines or penalties. Rather, the safety board usually performs deep investigations to identify root causes of problems in the transportation world. When an investigation is complete, the board often holds hearings and issues non-binding recommendations.

The first notable incident where a Waymo vehicle passed a stopped school bus happened last September in Atlanta. The Waymo pulled out of a driveway and crossed perpendicularly in front of the school bus from the bus’ right side. The robotaxi then turned left and proceeded down the street while kids were getting off the bus.

Waymo said at the time that the vehicle was unable to see the stop sign or flashing lights, and has since said that it addressed this particular scenario with a software update.

But as Waymo patched the specific scenario it encountered in Atlanta, some of the company’s vehicles were caught passing stopped school buses in Austin, Texas. Local news outlet KXAN published videos it received from cameras mounted on school buses that showed Waymo vehicles making illegal maneuvers on multiple occasions.

“We continue to engage productively with the Austin Independent School District and applaud their reported success in reducing human-driven violations around school buses from 10,000+ a year,” Peña said.

Authentic Brands Group has finalized the purchase of a 51 percent stake in the $6 billion brand.

Guess is poised to get a lot bigger.

On Friday, Authentic Brands Group finalized the deal to acquire a 51 percent stake in a newly formed entity that will own and license substantially all of Guess’ intellectual property and assume the majority of its product licensing agreements globally. The remaining 49 percent will be owned by Guess cofounders Maurice and Paul Marciano; Nicolai Marciano; Guess chief executive officer Carlos Alberini, and certain other trusts, foundations and affiliates.

The all-cash deal, first announced in August, is for $16.75 a share, a premium of 73 percent to the company’s closing price on March 14 when WHP Global offered $13 a share. The deal values Guess at $1.4 billion and takes the company private after 30 years in the public market.

Authentic pulled off an 11-hour deal since Guess had nearly inked a deal with its smaller rival, WHP Global, earlier last year.

Guess, which has sales of around $6 billion, will become the second largest company in Authentic’s portfolio, behind Reebok, and bring Authentic’s annual retail sales to $38 billion.

Current Guess management will continue to operate the company and its headquarters in Switzerland will remain as the Guess design, creative and distribution hub.

“Although Guess is doing $6 billion, the question is, why isn’t it doing $10 billion,” Jamie Salter, chief executive officer of Authentic, told WWD. “The reason is, they’re still working off a bunch of old practices — the way they did business for 45 years. I think we can change that.”

Salter added that while Guess is “a ginormous brand, it could use a little bit of help on the marketing side, the social side, to bring it into the 21st century.”

So it won’t be long until Authentic brings Guess into the entertainment, live events and hospitality space, Salter said. “You’ll see some increases there.”

He said Paul Marciano, who founded Guess with his brothers in Los Angeles in 1981 as a denim brand, “is very open to working closely with us on the marketing and social side of the business. As you know, content drives commerce.”

Paul Marciano concurred.

Until fairly recently, he said he didn’t know much about the brand management companies like Authentic and WHP Global and their ability to split the operations and intellectual property of businesses to help them grow.

So even though Guess continues to perform “extremely well,” Marciano said, it was time to escape the scrutiny of Wall Street.

“We have been public for 30 years, but to be public today in the retail and fashion world is a completely different experience.”

He continued: “We have always been in control of the brand, my brothers and I, and after my brothers retired, I’m still working every day. I love my job, and I have so many ideas and future projects that I want to do, but as a public company, it is very difficult, because everything is driven by quarterly reports and quarterly results, and it’s not viable for fashion company, especially a women’s fashion company, to predict the business every single week, every single quarter.”

He said being private will allow him to explore avenues of growth he was unable to as a publicly traded company.

“To have the freedom to be able to do projects that we need to build [the company], like going into a new territory like Brazil or Argentina, takes investment, but the Street will not give you that.”

He said once he met Salter, he could envision a bright future for Guess.

“He is really an incredible brand builder,” Marciano said. Although when Authentic started, it focused more on distressed companies, that strategy has shifted and it is now adding strong brands. “That caught my attention,” he said.

He stressed that he and the Guess team will continue to have 100 percent control over product, but he’s eager to tap into Authentic’s expertise in other areas.

“I don’t know entertainment — they do. I don’t know sports — they do. They have great IP for Marilyn Monroe, David Beckham, Kevin Hart. The crossover could be tremendous.”

And in terms of product, Marciano can see expansion into luxury underwear, loungewear, home as long as they are “relevant to the brand. We’ve done nothing but protect the brand for 45 years, and that’s very different than a lot of companies.”

Salter said he also sees opportunity to expand the existing Guess portfolio, with “big increases in sportswear, denim and kids.”

Marciano said he just returned from Italy and sees potential for adding to the Guess-Authentic portfolio.

“We’re looking at companies that have lost their founders, especially in Italy, in the denim business, in the fashion business. Since the founders are not there anymore, these companies have lacked discipline, direction and strategy, and they could use our help.”

But more immediately, Salter believes Guess’ rich archives are another plus.

“Going back 45 years, there’s just so much good stuff in the archives,” Salter said. “When we look to buy brands, we look at what is in the archive files, and how we bring certain styles back. That will make Guess even more valuable in the future.”

Salter said Authentic will work quickly to grow the Guess business.

“We come out of the gate strong and hard,” he said. “And then we go to a normal growth rate. So for the next 24 months, Guess is going to have a huge increase in revenue globally, just because it’s now going on our platform. We’re incredibly diversified in 150 countries all over the world.”

Salter said the deal means Guess will be “neck and neck” with Reebok as Authentic’s largest brand.

“We’re super excited. Paul’s absolutely brilliant. He’s very passionate about the brand, and I truly believe with the Authentic AI platform, we can speed up so many things for him. Our platform is completely digitized and we’re way ahead of most of the industry on the AI side of the business. AI is magic and you need great magicians to make magic. But I truly believe we’re going to get Guess moving so much quicker on the marketing, creative and the sourcing side.”

Harvey, the high-flying legal AI startup, has acquired Hexus — a two-year-old startup that builds tools for creating product demos, videos, and guides — as the company continues its aggressive expansion amid fierce competition in the legal tech market.

Hexus founder and CEO Sakshi Pratap, who previously held engineering roles at Walmart, Oracle, and Google, tells TechCrunch that her San Francisco-based team has already joined Harvey, while the startup’s India-based engineers will come onboard once Harvey establishes a Bangalore office. Pratap adds that she will lead an engineering team focused on accelerating Harvey’s offerings for in-house legal departments.

“What we’re bringing to Harvey is deep experience building enterprise AI tools in adjacent problem spaces,” Pratap said. “This expertise helps Harvey move faster in a market that’s becoming increasingly competitive.”

Hexus had raised $1.6 million from Pear VC, Liquid 2 Ventures, and angel investors before the acquisition. While Pratap declined to share deal terms, she said the structure was aligned around “long-term team incentives.”

The acquisition comes as Harvey looks to cement its position as one of AI’s hottest startups. The company confirmed last fall that it’s now valued at $8 billion after raising $160 million, bringing its funding across 2025 to $760 million. Andreessen Horowitz led that newest round, joined by new investors T. Rowe Price and WndrCo, alongside existing backers Sequoia Capital, Kleiner Perkins, Conviction, and angel investor Elad Gil. (It started the year with a $3 billion valuation after Sequoia led a $300 million Series D round in the company.)

Harvey now claims more than 1,000 clients across 60 countries, including a majority of the top 10 U.S. law firms.

When TechCrunch spoke with co-founder and CEO Winston Weinberg in November, he traced Harvey’s origin story back to a cold email sent to OpenAI CEO Sam Altman. Weinberg, then a first-year associate at O’Melveny & Myers, and co-founder Gabe Pereyra, a researcher who worked at Google DeepMind and Meta and was Weinberg’s roommate at the time, tested GPT-3 on landlord-tenant law questions from Reddit. When they showed the AI-generated answers to attorneys, two out of three said they’d send 86 of 100 responses with zero edits.

“That was the moment when we were like, wow, this entire industry can be transformed by this technology,” Weinberg said.

They emailed Altman on July 4, 2022, got on a call that same morning, and landed their first check from the OpenAI Startup Fund shortly after. According to Weinberg, the OpenAI Startup Fund remains Harvey’s second-largest investor.

Industry has faced pressure since pandemic, when many universities made exams optional

ETS, a nonprofit that owns the GRE and Toefl tests, has held talks to sell the exams or find one or more strategic investors, according to people familiar with the matter.

While the GRE has been a pillar of admissions to graduate programs for nearly a century, the number of people taking the exam has fallen in recent years as universities made it optional. The Toefl, used by U.S. universities to assess international applicants’ English-language proficiency, has faced rising competition.

President Trump’s campaign to shake up higher education is placing additional pressure on the exams. He has moved to limit international enrollment at U.S. universities, which affects foreign speakers and international-heavy graduate programs.

ETS Chief Executive Amit Sevak declined through a spokesperson to comment.

The nonprofit, based in New Jersey, has narrowed the talks to several firms that could potentially buy the exams or act as a strategic investor to expand ETS’s reach into the Middle East and India, said people familiar with the matter.

Interested parties include the Singapore-based investment firm Hillhouse, the private-equity firms Nexus Capital and Veritas Capital and the education entrepreneur Martin Basiri, some of the people said. Elysium Management, the family office of Apollo Global Management co-founder Leon Black, had been in talks with ETS but is no longer engaged, said a person familiar with the matter.

ETS has been seeking bids of around $500 million, some of the people said.

Hillhouse, which got its start managing money for Yale University’s endowment, was long a favored way for U.S. endowments and foundations to get investment exposure to fast-growing technology companies in China. Geopolitical friction between the U.S. and China has led its founder, the onetime Yale endowment intern Lei Zhang, to emphasize the firm’s global investment reach and cultivate a broader client base outside the U.S. in recent years.

Hillhouse’s discussions with ETS have been fluid. The talks have also included the possibility of Hillhouse’s joining with a U.S. investor or investors, said people familiar with the matter.

More recently, the talks have focused on whether Hillhouse, potentially with one or more partners, could invest in the commercial arm of ETS to help it expand the markets in which it offers the GRE and Toefl. Under this scenario, Hillhouse wouldn’t buy the tests or have access to test takers’ data, one of the people said.

The testing industry has faced significant pressure since the pandemic, when many universities made the GRE exam optional. In 2024-25, about 200,000 people took the GRE, down from more than 350,000 in 2020-21, according to ETS.

The Toefl, which long dominated testing for English-language proficiency, has faced competition in recent years from Duolingo, which offered a test that students could take from home during the pandemic. Toefl exams are generally taken in testing centers.

Private-equity firms in recent years have made moves into the education sector. Nexus has investments in the ACT, a college-entrance exam taken by millions of students, and Ruffalo Noel Levitz, an enrollment-management firm.

FT : A difficult year for Burgundy

A first taste of the ‘nightmare’ 2024 growing season

“Nightmare”, “misery”, “exhausting”, “unremittingly grim”, “desperate” — capped by “gruesome” final yields — those are some words that have been used to describe the 2024 burgundy growing season. For the UK merchants currently trying to sell the resulting wines, a very different narrative prevails: it’s a “classic”, “precise”, “elegant” vintage, “defined by freshness, clarity and poise”.

Burgundian vignerons have been worrying about the dangerously drier climate that seems to be heading their way for years, but in 2024 it was wet weather that proved the real hazard. They had almost enough rain for a decade’s worth of growing seasons. During the crucial flowering period in late spring, heavy rainfall in both frost-hit Chablis and the main Côte d’Or region positively pummelled the little vine flowers that should have developed into grapes, with later-flowering Pinot Noirs particularly hard hit.

The rain, accompanied by unusually low temperatures, continued throughout June and July, encouraging rampant mildew that, unusually, had begun as early as the flowering season. Almost as soon as one fungicide was administered, it was washed off the vines, which then required another round of spraying. According to Mark Haisma, who professed himself “delighted, at least, that we managed to get to harvest and did actually harvest something”, average annual rainfall records were smashed only halfway through the year.

Vines at the soggy foot of the Côte d’Or, the “golden slope” that’s home to Burgundy’s most famous vineyards, tended to suffer more than those higher up, where some of the rain drained away downhill.

One after another, vignerons who had previously committed themselves to organic viticulture decided to abandon that path in favour of more effective systemic sprays, simply to save their crop. Thibaud Clerget of Yvon Clerget told me he managed to stick to his organic principles only by going through the vineyard with permitted treatments 16 times — far more than usual. Elsewhere in the Côte de Beaune, Matthew Hayes, Burgundy specialist for JancisRobinson.com, reported that the much-admired Pierre-Yves Colin-Morey of Chassagne-Montrachet managed “respectable” yields of as much as 35 to 40 hl/ha (average in most years, but unusually high for 2024), by abandoning his usual organic preparations early on.

Admittedly, the Côte de Beaune, with its greater reliance on Chardonnay, was slightly better off than the Côte de Nuits in the northern half of the Côte d’Or. There were widespread reports of potential red wine crop losses of up to 80 per cent, with the white wine crop down by up to 30 per cent.

Late August brought a little respite. The grapes that had escaped mildew and rot benefited from slightly warmer weather, which raised their sugar levels. But the alcohol content in the resulting wines is unusually low, often less than 13 per cent. Uncharacteristically for this warmer century, many vignerons felt compelled to add a little sugar before or during fermentation to increase the final alcohol level.

Thanks to the cool summer, and the grapes’ struggle to ripen, acid levels in the resulting wines, reds as well as whites, are high.

The most successful reds could be described as pretty, but never as concentrated; and in some examples this seemed to accentuate the acidity and made the wines taste a bit raw. They are relatively pale, with tannins barely perceptible, and some wines already seemed dangerously evolved. A vintage to drink relatively early, then — if you can afford to buy it.

If 2024 had been a generous crop, there might have been some hope of a reduction in the sky-high prices of burgundy we have seen in recent years but the fact that 2025 is also a relatively small vintage has kept prices more or less stable.

The paucity of grapes made life difficult in many a cellar. Tanks that would usually be full for the fermentation were only half-full or less. This increases the risk of oxidation, microbial infection or, the vigneron’s nightmare, a stuck fermentation (where the yeast stops converting sugars into alcohol prematurely). In recent years, many producers have been cutting back on the proportion of new oak they use because oaky wines are so unfashionable. And anyway, the 2024 fruit was hardly robust enough to handle the impact of new oak. But for wines, such as grands and some premiers crus, for which a higher-than-average proportion of new oak barrels are customarily employed, there were instances of having to use as much as 50 per cent, simply because there was only enough wine to fill two barrels.

One of the major operations in red winemaking is extraction, persuading dark-skinned grapes to release colour and flavour from the skins (see below). Grégory Patriat, Jean-Claude Boisset’s winemaker, who seems to have managed to make a fist of what he called his “garage wines” because quantities were so small, observed that in the cellar the danger with the 2024s was “to try to extract what wasn’t there”.

In Burgundy, and with Pinot Noir vinifications in general, there has been a fixation on the proportion of whole bunches, or whole clusters, of grapes, that go into the fermentation vat rather than being destemmed. The stems and bunches can usefully aerate the fermenting must and can add freshness to the resulting wine, especially if the stems haven’t turned from green to brown (which Patriat claims hardly ever happens in Burgundy).

Because Chardonnay grapes were less affected by 2024’s vicissitudes, white burgundies — ever popular according to UK wine merchants — are more widely available than reds, even if they also tend to taste brisk — sorry, “classic” and “precise” — rather than generous.

Despite all this doom and gloom about the 2024s, and their paucity, London’s customarily hectic week of burgundy tastings took place as usual this month with what felt like as many different samples as ever — presumably spurred by the UK merchants’ desire for cash — even if the available quantities of each wine were presumably much smaller than usual. Quite a few of the tastings included numerous older vintages as well, indicating a certain resistance to burgundy price levels among the merchants’ customers.

One important thing to note, however, is that the vineyards south of the Côte d’Or — in the Côte Chalonnaise, the Mâconnais and Beaujolais — suffered far less in 2024 and quantities produced there were much more generous. It’s also true that the wines from these regions, less widely shown in London than the Côte d’Or classics, are not just generally much cheaper than those from the Côte d’Or, they have also become significantly more sophisticated in recent years. For 2024s, head south for value.

The most successful producers I encountered in the nine tastings I attended are listed here but my colleagues tasted at 10.

The insurer completed a major turnaround after the financial crisis and is now an inexpensive play.

AIG shares dropped 7% in early January due to CEO Peter Zaffino’s announced departure, leading to a 14% year-to-date decline.

AIG trades below its book value of $75 per share and at nine times estimated 2026 earnings, a discount to its peers.

AIG aims to boost earnings per share by 20% annually from 2025 to 2027 and has increased its dividend by 12.5% in 2025.

American International Group has completed a major turnaround since the property-and-casualty insurer nearly collapsed during the 2008-09 financial crisis.

A streamlined and increasingly profitable AIG offers investors an inexpensive insurance play. That’s doubly true right now. The insurer’s shares dropped 7% in early January on surprise news that CEO Peter Zaffino would step down on June 1 while remaining chairman.

That move disappointed Wall Street because Zaffino, 59, has led AIG for the past five years and overseen the company’s revival, highlighted by an improvement in its formerly poor insurance underwriting results.

AIG said Zaffino’s decision was his own and that its board “remains very confident in Peter’s leadership and integrity.”

AIG shares, at about $72, are down 14% so far this year after hitting a new 52-week low this past week.

“On any metric, the stock is cheap now,” says Colin Hudson, a co-manager of the Oakmark Equity and Income fund, which owns shares. “Zaffino has done a fantastic job turning around AIG. I understand the disappointment that Zaffino will step down as CEO, but he’s leaving the company on a strong footing.”

AIG stock trades below its book value of $75 a share—a discount to peers like Travelers, Chubb, and Hartford Insurance Group, which trade for 1.7 to two times book. AIG also trades for about nine times estimated 2026 earnings of nearly $8 a share, also a discount to P&C peers.

The entire P&C sector has been weak this year, with major stocks, excluding AIG, down an average of about 5%. This reflects concern that P&C insurance pricing is weakening after several years of strength.

AIG has underperformed the sector due to the Zaffino departure and waning speculation about a potential takeover bid by industry leader Chubb for AIG. That talk was prompted by a December report in an industry publication, Insurance Insider US, that Chubb had made an “informal approach to AIG.” Chubb denied that any offer was made, and AIG said that it isn’t for sale.

When AIG earlier this month named Eric Andersen, an executive at Aon, a large insurance broker and consulting firm, to succeed Zaffino as CEO, it further dampened takeover talk because it signaled no change in AIG’s strategic direction, according to a note from UBS analyst Brian Meredith.

While the takeover talk has subsided, a Chubb/AIG combination isn’t out of the question. There is strategic merit to creating a larger P&C insurer. Such a merger would probably “involve material expense synergies due to duplicative functions,” BofA Securities analyst Joshua Shanker wrote in December.

Another intriguing aspect of a possible deal is that it would allow a member of the Greenberg family to once again head AIG.

The father of Chubb’s CEO Evan Greenberg, 70, is insurance legend Maurice “Hank” Greenberg, 100, who built AIG into the world’s dominant P&C insurer over decades before leaving the company in 2005 amid board concerns about several regulatory inquiries at the time, according to coverage in The Wall Street Journal.

Chubb has about $56 billion of annual premium revenue, more than double the $24 billion at AIG. Its market capitalization is $118 billion, three times that of AIG, which also has a large investment portfolio totaling $89 billion.

AIG operates globally and offers a range of P&C insurance policies, including directors and officers coverage, workers’ compensation, general liability, marine, and airline. It gets about a quarter of its premiums from policies for individuals, including a high-end homeowners program called Private Client Select that competes against Chubb’s Masterpiece.

AIG required a government bailout during the financial crisis after large losses at its now-shuttered financial derivatives unit. That federal support resulted in huge dilution to shareholders. There followed a decade of poor underwriting results as AIG underpriced insurance policies in what analysts called a bid to maintain market share.

Results have improved considerably under Zaffino. The company has had substantial underwriting profits in each of the past five years—helped by a strong market for P&C pricing.

Zaffino has also shrunk AIG to its core P&C business. The company sold its reinsurance business, Validus, in 2023, and took public its life and retirement business as Corebridge Financial in 2022.

AIG has sharply reduced its Corebridge stake to just under 10%, and used the sizable proceeds from several Corebridge stock sales to help fund a large stock-repurchase program. AIG bought back 17% of its stock—some $8.6 billion—in the five quarters ending in the third quarter.

Buybacks however are expected to be more modest at $1 billion in 2026, the company said on its third-quarter conference call.

It’s possible that AIG may boost the buyback when it reports fourth-quarter results on Feb. 10 due to the weakness in the stock. “With the stock in the low 70s, AIG could get more aggressive,” Hudson says.

AIG announced a few growth initiatives in conjunction with the third-quarter earnings including a deal to take over some P&C underwriting business from Everest Group with about $2 billion of annual premiums.

KBW analyst Meyer Shields says that AIG is “scaling intelligently” in a softer market. He has an Outperform rating and a price target of $96.

And the “soft market” for insurance pricing may be somewhat overstated. While prices of property coverage are down double digits this year following a light hurricane season in 2025, many types of coverage like general liability are still rising at close to 10%. AIG’s North American pricing, excluding property, was up 5% in the third quarter, down from a 6% increase in the second quarter—hardly a major change.

P&C insurers earn profits from underwriting and by investing the premiums. A soft market doesn’t mean underwriting profits will evaporate, but industry profit growth could slow.

CFRA analyst Cathy Seifert says the “kerfuffle” around AIG should settle down as the new CEO settles in. ‘You’ve seen a deflation of the takeover balloon, and the management transition was less than stellar,” she says. Analysts and investors had expected Zaffino to stay on for a few more years—his contract is set to expire in 2027. Seifert has a Buy rating and $90 price target on the stock.

AIG isn’t in a class with Chubb in returns, underwriting results, and culture, but it’s narrowing the gap and trades at a discount to Chubb.

AIG’s financial reports also are a tough slog. The company’s preferred earnings measure that it calls “adjusted after-tax income attributable to AIG common shareholders” includes about 20 adjustments to operating earnings.

AIG laid out some ambitious financial goals at its investor day last year. It aims to boost earnings per share by 20% annually over the 2025 to 2027 period, boost its dividend by 10% a year, cut its expense ratio to under 30% of premium revenue, and generate a core return of equity of 10% to 13%. So far, it’s on track for those goals.

AIG’s adjusted earnings are expected to be up 40% in 2025 and 12% in 2026. The dividend was raised 12.5% in 2025 and now offers a yield of 2.5%.

AIG offers a low valuation, an improving earnings outlook, and the possibility of a takeover. That seems like a risk worth taking.

Nearly half of investors plan to increase exposure to hedge funds, Goldman Sachs survey finds

Hedge funds achieved their best performance in over a decade in 2025, with assets under management reaching a record $3.5 trillion.

Investor interest surged, with 45% of fund investors planning to increase hedge fund exposure in 2026, a Goldman survey found

Top-performing strategies included healthcare, technology, and macro funds, with some firms gaining over 30% and management fees increasing.

Hedge funds are back on top of Wall Street’s pecking order.

Performance in 2025 across hedge funds was the best it has been in over a decade, with stock pickers and bond and currency traders reporting some of the best results. Managers of funds with elevated returns last year include Rob Citrone, part owner of the Pittsburgh Steelers; tech specialist Gavin Baker; and British billionaire Chris Hohn.

Last year, pension plans, endowments and other clients put more money into hedge funds than in any year since 2007, according to research firm HFR. The recent stretch of strong returns is making hedge funds the most favored asset class for institutions that allocate money to outside investment firms, according to a Goldman Sachs GS -3.75%decrease; red down pointing triangle survey of 317 fund investors viewed by The Wall Street Journal.

About 45% of them plan to increase their exposure to hedge funds in 2026, net of those looking to decrease it, an all-time high in survey records going back to 2017 and an 18-percentage-point increase from last year. That well outpaces the share that are planning increases to private-equity, private-credit, venture-capital and other types of fund managers, nearly all of which have seen declining interest since last year.

“Hedge funds have retaken that pole position,” said Freddie Parker, co-head of prime insight and analytics at Goldman. “All the numbers point to a very optimistic story.”

Assets under management in hedge-fund vehicles jumped to a record of roughly $3.5 trillion in 2025, according to Goldman estimates. That is up from $3 trillion in 2024.

It has been a long climb back to the summit. For most of the 2010s, hedge funds struggled to generate profits with short-term interest rates at zero, volatility at low levels and high correlations between different stocks. Hundreds of hedge funds closed shop, and the number of fund liquidations started exceeding the number of new launches. A unit of JPMorgan that invests client money in hedge funds called the period the “alpha winter.”

Now, with rates above 2%, divergence in monetary policy across the world and bouts of volatility and dispersion in stock markets, conditions are ripe for hedge funds to perform well. The shift has reverberated across Wall Street, with banks reporting record profits from their business of executing trades for and lending money to hedge funds.

“We’re in the heat of the summer right now,” said Paul Zummo, chief investment officer of JPMorgan Alternative Asset Management. “It’s been an excellent environment and people are taking advantage of it.”

A composite index from research firm PivotalPath that tracks the performance of nearly 1,300 hedge funds with $2 trillion in assets under management gained 11.9% in 2025. That was its best year since 2013. The top decile of funds in that PivotalPath index returned nearly 30%.

The best performing hedge-fund strategies last year included firms that bet on and against healthcare stocks, those that trade tech stocks and macro fund managers that predict big-picture market moves using economic and geopolitical information.

Biotech hedge funds rallied in the second half of the year thanks to an acceleration in merger activity and promising results from clinical drug trials. Roderick Wong’s $9 billion firm, RTW Investments, gained about 55% in its main biotech hedge fund, a person familiar with its performance said.

For tech hedge funds, surges in stocks involved in the artificial-intelligence supply chain helped turbocharge returns. Baker’s Atreides Management gained about 40% in its hedge fund’s portfolio of public stocks, according to an investor update. Positions in Ciena and Lumentum, makers of optical networking equipment and components, were winners. An Atreides portfolio that also holds stakes in private companies, including Elon Musk’s SpaceX and xAI, did even better, gaining about 47%.

Stock picker Chris Hohn made $18.9 billion for clients of his TCI Fund Management last year, a record one-year haul, according to estimates from Swiss investment firm Edmond de Rothschild. TCI finished 2025 with a 27.8% gain in its master fund, the Journal previously reported.

Turbulence in markets as well as geopolitics presented plenty of opportunities for others. Castle Hook Partners, a roughly $10 billion macro hedge-fund firm run by David Rogers, gained about 50% last year, helped by bets on AI and other stocks, a person familiar with its performance said.

Citrone’s Discovery Capital generated a 37% return, people familiar with his performance said. Gains came from big bets it had placed on Nigeria, especially on stocks with major Nigerian operations, and against stocks of companies poised to be disrupted by AI, according to an investor outlook viewed by the Journal.

The run of recent returns has made clients of hedge funds more comfortable paying higher fees. The average management fee that respondents to the Goldman survey reported paying ticked up to 1.64% of assets under management, and the average performance fee ticked up to 17.8% of investment profits. Each was the highest since at least 2016.

Rokos Capital Management, the macro hedge-fund firm run by British billionaire Chris Rokos, told clients last year he planned to raise performance fees to 25% from 20% over the next three years, according to people familiar with the matter. Rokos gained about 20% last year, one of the people said.

Yann LeCun’s new venture, AMI Labs, has drawn intense attention since the AI scientist left Meta to found it. This week, the startup finally confirmed what it’s building — and several key details have been hiding in plain sight.

On its newly launched website, the startup disclosed its plans to develop “world models” in order to “build intelligent systems that understand the real world.” The focus on world models was already hinted at by AMI’s name, which stands for Advanced Machine Intelligence, but it has now officially joined the ranks of the hottest AI research startups.

Building foundational models that bridge AI and the real world has become one of the field’s most exciting pursuits, attracting top scientists and deep-pocketed investors alike — product or no product.

World Labs, a direct rival founded by AI pioneer Fei-Fei Li, became a unicorn shortly after coming out of stealth. After launching its first product, Marble, which generates physically sound 3D worlds, World Labs is now reportedly in talks to raise fresh funding at a valuation of $5 billion.

There’s little doubt that VCs would be equally eager to invest in LeCun, adding credibility to rumors that AMI Labs might be raising funding at a $3.5 billion valuation. According to Bloomberg, VCs in talks with the startup include Cathay Innovation, Greycroft, and Hiro Capital, to which LeCun is an advisor. Other potential investors reportedly include 20VC, Bpifrance, Daphni, and HV Capital.

Regardless of who writes the checks, investors may want to note an important detail: As LeCun has made clear, he is AMI’s executive chairman, not its CEO. Instead, that role belongs to Alex LeBrun, previously co-founder and CEO at Nabla, a health AI startup with offices in Paris and New York.

LeBrun’s transition from Nabla to AMI is part of a partnership announced last December by Nabla, which develops AI assistants for clinical care and to which LeCun has been an advisor. In exchange for “privileged access” to AMI’s world models, Nabla’s board supported LeBrun’s shift from CEO to chief AI scientist and chairman, clearing the way for his new role.

As AMI Labs’ CEO, LeBrun will be surrounded by familiar faces. After Facebook acquired his previous startup, Wit.ai, the serial entrepreneur and AI engineer worked under LeCun’s leadership at Meta’s AI research laboratory, FAIR. According to reports, the duo will also be joined by Laurent Solly, who stepped down as Meta’s vice president for Europe last December.

The talent overlap between AMI and Meta likely won’t stop there. LeCun told the MIT Technology Review that his former employer could well be AMI’s first client. But he has also been publicly critical of some of Meta’s strategic choices made under Mark Zuckerberg’s direction. More broadly, the Review interprets AMI Labs as a contrarian bet against large language models (LLMs).

The limitations of LLMs that LeCun has pointed out include hallucinations, which are a serious concern in contexts like medicine, as LeBrun also knows firsthand. AMI Labs’ CEO told Forbes that a big reason he took the role was the prospect of applying its world models to healthcare. But the startup will also target other high-stakes applied fields.

“AMI Labs will advance AI research and develop applications where reliability, controllability, and safety really matter, especially for industrial process control, automation, wearable devices, robotics, healthcare, and beyond,” it wrote in its mission statement. “We share one belief: real intelligence does not start in language. It starts in the world.”

Unlike generative approaches, which LeCun and his team see as poorly suited for unpredictable data such as sensor input, the startup promises that its AI systems will not only understand the real world, but also have persistent memory, the ability to reason and plan, and be controllable and safe.

The startup plans to license its technology to industry partners for real-life applications, but says it also plans to contribute to building the future of AI “with the global academic research community via open publications and open source.” LeCun said he plans to keep his professor position at NYU, where he teaches one class per year and supervises PhD and postdoctoral students.

This means that the French-born researcher will remain based in New York, but he told the MIT Technology Review that AMI Labs “is going to be a global company [that’s] headquartered in Paris.” The news was welcomed by French President Emmanuel Macron, who expressed his pride that LeCun, who is also a Turing Prize winner, chose Paris. “We will do everything we can to ensure his success from France,” he said.

The startup will also have offices in Montreal, New York, and Singapore, but its decision to pick Paris for its headquarters will help consolidate Paris’ reputation as an AI hub, where it will join the ranks of H, Mistral AI and several international labs, including FAIR. It’s fitting, perhaps, that AMI is pronounced a-mee — like “ami” in French, which means “friend,” LeCun has pointed out.

The OpenAI co-founder and his wife had avoided the limelight until deciding to seize a presence for themselves in national politics.

In November, Greg Brockman, OpenAI’s president and co-founder, and his wife, Anna, enjoyed a glitzy night out in Washington: a White House state dinner for Saudi Crown Prince Mohammed bin Salman.

The Brockmans were among a broad contingent of tech elite in attendance at the fete, hosted by President Donald Trump. Soon after the event, the Brockmans posted pictures of themselves posing with Elon Musk, Nvidia’s Jensen Huang and his wife, Lori, and David Sacks, the venture capitalist who’s become Trump’s AI and crypto czar.

It wasn’t the only time in the last few months that the Silicon Valley couple has made splashy appearances in Trump’s orbit. In the fall, they attended two other White House suppers and went to Trump’s revamped Kennedy Center Honors, dining later that evening with one of the honorees, actor Sylvester Stallone. But nothing has more loudly announced their arrival as Trump-friendly figures than their decision to become megadonors: giving $50 million to Leading the Future, a bipartisan super PAC focused on combating state-level AI regulation, and $25 million to MAGA Inc., a Trump super PAC.

On one level, the donations reflect a widely recognized political shift in Silicon Valley, where tech companies and their leaders are clamoring to get close to Trump, who has made it apparent he can be swayed by their praise and attention.

Nonetheless, the Brockmans’ political giving caught many people who know Greg off guard. The executive had never seemed particularly concerned with politics, according to four people who’ve worked with him. If Brockman’s political stance had to be defined, he’d “probably come down a little bit more on the libertarian side,” mainly because of his interest in minimal government regulation, said one person who knows Brockman well.

Still, some within Brockman’s circle say such a public-facing position is something that would appeal to his desire for broader recognition. “When I see him doing all these political things, I imagine it feels good to him to be hobnobbing with heads of state and going to fancy parties and getting photographed with Jensen,” said a person who has worked closely with him.

True enough, the spotlight has never fallen on Brockman as it has recently. In the 11 years since OpenAI’s start, he has worked largely in the shadow of CEO Sam Altman—even as Altman turned himself into a household name and nearly all of the other 11 co-founders of OpenAI departed. Praised as a workhorse and as OpenAI’s top engineer, Brockman has nonetheless struggled to define his role because of clashes with other staffers over his abrasive working style. Most recently, he’s helped lead OpenAI’s infrastructure buildout, which he has described as the company’s biggest challenge.

One constant in Brockman’s life has been his wife, Anna, whom he married six years ago. Even though she has never had a formal role at OpenAI, she is a familiar figure to OpenAI staff, in part because of her tendency to turn up unexpectedly at internal meetings and events. Throughout that time, she has avoided the kind of visibility she and her husband are poised to receive as outsize political donors, working hard to keep the personal details of her life private.

For Greg, his recent major donations seem to be an anomaly. Until 2025, he had made only three political donations in his entire life: $2,700 in 2018 to Rep. Patrick McHenry, a North Carolina Republican—and $5,400 to support Hillary Clinton’s 2016 presidential campaign. Meanwhile, no public records exist of Anna ever having given any money before, though it is possible she donated under her maiden name (which couldn’t be learned thanks to her efforts to safeguard her privacy).

While the Brockmans are emerging as the latest Silicon Valley power couple, a longstanding corporate drama involving OpenAI has pushed Greg further into the limelight just in the last couple of weeks: the high-profile lawsuit between Musk and OpenAI over the startup’s ownership structure.

The litigation has unearthed some of Brockman’s personal diary entries, which Musk seems intent on using as ammunition against OpenAI. Already, he has shared excerpts of them on X. Last week, the two men sparred over the context and meaning behind the text.

Musk’s co-opting of Brockman’s diary adds a new dimension to their relationship—and makes that photo session at the Saudi state dinner even more curious in retrospect. When they were posing together for the photo in November, Brockman and Musk were on relatively good terms, according to the person who knows Brockman well.

While Brockman’s status as a political megadonor is new, OpenAI has been placing him in front of government officials since 2016, when he spoke on behalf of the company during the first ever congressional hearing on AI.

During his Senate testimony, a fresh-faced, then-29-year-old Brockman offered an enthusiastic assessment of the then-nascent AI industry within the U.S., speaking about competition from countries like China and South Korea and arguing for increased government funding and AI competitions to better measure the field’s progress.

A year later, he testified again, telling the House of Representatives about recent AI advancements in image recognition and translation as well as the quest for humanlike AI—artificial general intelligence, OpenAI’s primary goal.

The person who knows Brockman well says his personal interest in politics has grown through a relationship with Trump’s secretary of the interior, Doug Burgum. Burgum is a former governor of North Dakota, Brockman’s home state. The two initially met at a tech event when Burgum came to visit Silicon Valley while he was governor, according to a person close to Brockman: The pair bonded over AI and their shared origins, and Burgum gave Brockman a North Dakota pin. (Before politics, Burgum had a lengthy career in tech that included years at Microsoft and serving as an Atlassian board member.)

In the run-up to the 2024 presidential election, Burgum set up a meeting between Trump, Brockman and OpenAI Chief Operating Officer Brad Lightcap in Las Vegas to give Trump a briefing on the current state of AI, according to the person who knows Brockman well. (Altman was originally supposed to attend but fell ill at the last minute and was replaced by Lightcap.) The OpenAI executives gave Trump a ChatGPT demo and also discussed with him its infrastructure needs and how AI technology was becoming a matter of national security, the person said.

Around the same time, OpenAI staffers had also reached out to the Biden campaign and ended up giving a similar AI-focused briefing to senior campaign staff, though Brockman didn’t attend that meeting, according to the person who knows him well.

When Trump was elected in November, Brockman congratulated Trump on X, praising the president’s “tech forwardness.”

Brockman’s newfound interest in politics may also reflect the influence of Chris Lehane, a Clinton White House veteran who joined OpenAI in 2024 as its president of global policy.

Two former OpenAI insiders see the fingerprints of Lehane, who previously helped a pro-crypto super PAC stack Congress with dozens of crypto-friendly candidates, on the Brockmans’ evolution as players in Washington. “This has Chris written all over it,” said a former OpenAI employee. Separately, the person who knows Brockman well said Lehane provides informal advice to staffers at Leading the Future, the PAC that received $50 million from the Brockmans.

(Neither Lehane nor Brockman would comment on the exact nature of their relationship. And in a statement, an OpenAI spokesperson said the Brockmans are making the donations in a personal capacity.)

It’s not uncommon, however, for a tech company to coordinate with its executives on their political donations as part of the business’s broader lobbying strategy, said Nu Wexler, a former policy communications executive at Google and Meta Platforms, who now runs Four Corners Public Affairs, a Washington-based communications firm. When a company wants to enlist one of its executives to woo a politician, it will consider that executive’s past party affiliations—and whether they have the personal wealth to make an impact.

As a vessel for courting Trump, Brockman, having only given a handful of tiny bipartisan donations, might have made more sense than Altman, who has a lengthy and well-publicized history of donating to Democrats. And as an OpenAI co-founder and former Stripe executive, Brockman is certainly rich enough to afford the donations.

Opportunistic political giving—rather than donating because of ideology—has become much more common during the second Trump administration, argued Wexler. “The transactional nature of the Trump administration has convinced a lot of tech executives that political giving is worthwhile,” he said.

Broadly, OpenAI has been ramping up its political efforts. It mounted a concerted opposition to California’s SB 1047 AI safety bill, which Gov. Gavin Newsom later vetoed, while supporting other AI-focused legislation supporting child safety and protection against deepfakes. In all, OpenAI’s lobbying expenditures grew sevenfold from $260,000 in 2023 to $2.1 million last year, part of a greater industrywide effort to sway America’s politicians. Anthropic, for instance, spent $2.3 million during the same period, a nearly identical increase.

Brockman’s ambitions have already propelled him far. Born in Thompson, N.D. (population roughly 1,100), he showed an early aptitude for math and science that helped earn him a place at Harvard University. He initially planned to spend his time in Cambridge studying math and computer science, but he quickly dropped out after deciding his classmates couldn’t challenge him and transferred to the Massachusetts Institute of Technology. He dropped out of MIT, too, after lining up a job as an early Stripe employee.

Brockman spent five years at Stripe, eventually becoming chief technology officer before leaving to start OpenAI in 2015.

Around 2018, he started dating Anna. Little is known about her past before she met Greg. Today, she serves on the board of the Schwarzman Animal Medical Center, a nonprofit in New York, alongside current OpenAI board member Nicole Seligman, a former Sony Entertainment president.

Anna has been diagnosed with autism, a condition that has given her “some true superpowers,” Greg wrote in an X post last year, as well as a deep connection with animals. Her affinity with the natural world seems to extend further, and one of Anna’s chief interests in seeing OpenAI create humanlike AI is for the technology to benefit people and animals alike, said the person close to Brockman.

In a Federal Election Commission filing tied to her MAGA Inc. donation, she listed her profession simply as “retired.”

Anna and Greg married in 2019 in a civil ceremony at OpenAI’s San Francisco office. Fellow co-founder Ilya Sutskever officiated the ceremony, and for a ringbearer, they relied on a robotic hand.

In recent years, she has become a mainstay at the office, attending meetings and company events with and at least one without him despite the fact that the company doesn’t employ her, said a person with knowledge of OpenAI. In summer 2022, for example, she was the only nonemployee present at a technical offsite near Yosemite National Park for OpenAI leaders around the launch of GPT-4, according to a former OpenAI employee who saw her there.

Tumult has marked the last couple of years for Brockman at OpenAI. When OpenAI’s board fired Altman in November 2023, Brockman quit in protest before returning with Altman five days later. The next summer, Brockman took a sabbatical from the company following complaints about his corrosive working style from researchers and engineers, who said he would make changes to projects without consulting others and redo other people’s work out of a belief that his approach was better.

Brockman has also been known to clash with other executives at OpenAI—including Mira Murati, the company’s former chief technology officer—over who would get credit for various products like the GPT-3 and GPT-4 application programming interfaces, according to a person with knowledge of the disagreements.

Brockman eventually returned from his sabbatical in November 2024 in the wake of Murati’s high-profile departure to begin a rival AI startup, Thinking Machines Lab.

When he came back from the break, some OpenAI executives wondered if he would still have a place at the company. Altman had since effectively taken over the company’s technical teams, with whom Brockman had historically spent the majority of his time, jumping between projects like rewriting code for model training and working on reasoning models. Altman had said he wanted to be closer to AI development. That’s when Brockman pivoted to focus on OpenAI’s infrastructure and managing the company’s massive clusters of Nvidia chips, along with the hundreds of staffers working on them.

As if he’s not busy enough, Brockman now has to contend with an escalating courtroom battle with the world’s richest man.

That contretemps is related to the 2024 lawsuit Musk filed against OpenAI, Altman and Brockman. The suit claims they deceived and defrauded him by shifting away from OpenAI’s original mission to be a nonprofit.

Musk, one of OpenAI’s co-founders, gave the company around $40 million to get it off the ground, the majority of its seed capital, he claimed in a recent court filing. He now claims he was stiffed out of up to $134 billion when it created a for-profit subsidiary in 2019.

A trial is set to begin this spring, and dozens of new documents were made public this month, a trove that included Brockman’s diary entries. In those pages, Brockman outlined his support for creating a for-profit arm of OpenAI and mulled the fortune he could make : “We truly have the chance to make this happen. Financially what will take me to $1B?” he wrote in one entry.

At the same time, he lamented that the moves he and other executives were discussing—turning OpenAI into a for-profit and booting out Musk—were “morally bankrupt.” “It’d be wrong to steal the nonprofit from him,” Brockman wrote in the same entry.

Musk seized upon these diary entries, declaring in an X post on January 16 that they revealed a “conspiracy” among OpenAI’s leaders “to commit fraud and steal the charity.” Brockman hit back by describing Musk’s excerpts from Brockman’s diary as “cherry-picking.” OpenAI published a blog post that included additional snippets of a conversation between Musk, Brockman and Sutskever that showed Musk was also involved in discussion around creating a for-profit OpenAI entity. Musk wanted to gain control of OpenAI, the blog post claimed, and expressed a desire to “accumulate $80B for a self-sustaining city on Mars.”

What Brockman’s diary does show explicitly is that he has been thinking about his potential legacy for a long time. In fact, he coined a term for himself and the group of people around him at work on OpenAI and the technology that overturned Silicon Valley: As Brockman put it, they are the “kings of AI.”