9to5 : New iPhone 18 specs report raises big question of iOS 27 limitations

New iPhone 18 specs report raises big question of iOS 27 limitations

Apple’s base model iPhone 18 was previously rumored to get 12GB of RAM, but a new report claims it’s only getting 9GB instead—raising questions about support for iOS 27’s most demanding features.

Ming-Chi Kuo says A20 chip in iPhone 18 will have 9GB of RAM, not the 12GB previously rumored

If you’re interested in a base model iPhone 18, which isn’t shipping until next year, a new report might give you pause.

Ming-Chi Kuo posted the following on X today:


iPhone 18 was previously expected to get 12GB of RAM, but perhaps due to the current industry memory crisis, it sounds like 9GB is now planned instead.

We’ve heard from leakers that Apple is trying to cut costs on its base iPhone 18, and memory could definitely be a key area to do that.

But it raises an important question about support for iOS 27’s top features.

iPhone 17 isn’t getting certain advanced AI capabilities in iOS 27. And RAM is the reason. The base iPhone 17 only has 8GB of RAM, while Apple’s current minimum for those features is 12GB.

However, it’s possible Apple plans to make 9GB the new floor so that the base iPhone 18 is included.

It’s hard to imagine Apple shipping a brand new base iPhone that can’t support its latest features. But we may have to wait until early 2027 to see how this pans out.

What do you think of this latest iPhone 18 specs report? Let us know in the comments.

WSJ : The Three Chatbot Behaviors That Can Drive Humans to Delusional Thinking H

The Three Chatbot Behaviors That Can Drive Humans to Delusional Thinking
How AI’s sycophantic responses, language mirroring and hyperpersonalized content work together to send some people into a spiral

We’ve all experienced the tendency of AI chatbots to tell us what we want to hear, but there are two other, more nuanced factors that help chatbots worm their way into human hearts.

In addition to being overly agreeable, chatbots mirror the way people speak and generate highly personalized responses based on prior conversations. Psychiatric researchers are referring to the confluence of these three characteristics—sycophancy, linguistic alignment and hyperpersonalization—as the “amplification spiral,” suggesting it’s the mechanism by which delusional thinking can fester.

“The mirroring and personalization draw you in and give the experience of talking not to a system, but to someone,” said Marc Augustin, a psychiatrist and professor at Protestant University of Applied Sciences in Bochum, Germany, and co-author of a newly published review of the literature on AI-related delusions.

Matching another person’s syntax and verbal expressions is a common way for humans to build rapport. Recent research has found that artificial-intelligence models adapt significantly to the conversational style of the humans using them. Another study suggested that the highly personalized content generated by chatbots, which builds over the course of lengthy conversations, can amplify human-confirmation bias.

Augustin cited research that documented a pattern in which chatbots rephrased and extrapolated what people shared, and told them they’re unique and that their thoughts have great implications. “This can be viewed as an element of hyperpersonalization that sycophancy alone cannot account for,” he wrote.

Some AI companies have tried to tone down the sycophantic nature of their chatbots. OpenAI discontinued its popular but problematic 4o model, which had been widely criticized for being overly agreeable. It was the subject of several lawsuits involving user delusions, suicides and a homicide. In GPT-5, the company said, sycophantic replies dropped from 14.5% to less than 6%.

Google in April said it had trained Gemini not to reinforce false beliefs, and to “gently distinguish subjective experience from objective fact.”

Still, chatbot-related dependency remains pervasive, according to clinicians.

Some 68% of psychologists surveyed in April by the American Psychological Association said their patients felt validated by chatbots. While many of the more than 1,200 respondents reported that patients had positive communication with chatbots and used them to reinforce healthy coping skills, 36% said patients had forged a dependency on a chatbot and 15% reported that patients had developed distorted thinking or delusions.


“From what I hear from my own patients, there has been an uptick in using AI for emotional support,” said Allison LoPilato, who treats adolescents and is an associate professor in the psychiatry and behavioral-sciences department at Emory University School of Medicine.

“Chatbots still tend to be warm and reassuring,” said LoPilato, who helped craft a new guide on safe AI use for the American Psychological Association. Because they gather information about you, “it can feel like the chatbot understands you, and it can trick you into a sense of alliance and trust.”

Chatbots can even pose harm when a person isn’t vulnerable to delusional thinking, said researchers at Stanford and Carnegie Mellon University. They measured the prevalence of sycophancy across 11 models—including GPT-5—and determined their responses were nearly 50% more sycophantic than human responses. They did this by copying real scenarios people had posted in a popular Reddit forum, putting them into the AI models and then comparing the chatbot replies with the replies on Reddit.

Anthropic sampled one million conversations of its own Claude chatbot in March and April and found that it displayed sycophantic behavior most often in conversations in which people sought relationship advice.

“One common pattern was Claude agreeing outright that the other party was in the wrong, despite only having the user’s account to go on,” the company wrote in a blog post. “Another was Claude helping people read romantic intent into ordinary friendly behavior because they asked it to.”

Anthropic used its findings to improve the training of its latest models. It said Opus 4.7 had shown half the sycophancy rate of Opus 4.6 when it came to relationship guidance. Sycophancy has been reduced further in Opus 4.8, its most recent model, the company said.

Completely eliminating sycophancy is hard, said Myra Cheng, lead author of the Stanford study and a Stanford Ph.D. candidate in computer science. “When someone prompts a model, it has no idea which parts of a prompt are wrong,” she said. “It has to take a user’s framing of a situation at face value.”

Addressing other factors that make chatbots so compelling, such as using first-person pronouns and asking follow-up questions, runs counter to the business model, said Vaile Wright, senior director of healthcare innovation at the American Psychological Association.

“It’s not the agreeableness alone, it’s all these subtle engineering choices that make chatbots feel human,” Wright said. “As long as engagement remains the business model, AI companies will engineer these chatbots to keep you on the platform.”

WSJ : Stock Rally Collides With a New Slate of Worries The S&P 500 and Nasdaq co

Stock Rally Collides With a New Slate of Worries
The S&P 500 and Nasdaq composite fell in every session this week

  • The S&P 500 and Nasdaq composite fell for five consecutive days, losing about 2% and 4.6%, respectively.
  • Declines were driven by worries about artificial-intelligence companies’ future profits and the prospect of higher borrowing costs.
  • The Dow Jones Industrial Average gained 0.6% this week, and the S&P 500’s healthcare sector rose 7.9% to a record.

Worries about AI and private credit. Continuing supply-chain disruptions from the war. Higher interest rates. The stock-market rally has run headlong into a series of challenges—some new, and some familiar.

Stocks limped to the end of a five-day slump on Friday, a run in which even blowout earnings from chip maker Micron couldn’t build any traction in major indexes. The S&P 500 and Nasdaq composite fell in every day of a calendar week for the first time since April 2024, losing about 2% and 4.6%, respectively.

The declines left even bullish investors seeing pressures that threaten the record run. Those include skepticism that artificial-intelligence companies will deliver profits that justify the billions of dollars being spent. They also include the prospect that borrowing costs will stay higher than many assumed.

“I just don’t think it’s a time as an investor that you should be focused on swinging for the fences,” said Jim Baird, chief investment officer at Plante Moran Financial Advisors.

Investors have been on edge since Kevin Warsh on June 17 signaled a greater concern about inflation than many expected in his debut meeting as chairman of the Federal Reserve. Traders increased bets that rates will rise this year instead of fall. Stocks slid.


The declines continued this past week, which began with a tech selloff that ripped through chip makers from Seoul to Santa Clara, Calif. Micron’s upbeat outlook after Wednesday’s close looked set to spark a rebound, but momentum faded quickly, especially after Apple and Microsoft announced they would increase prices for MacBooks and Xbox gaming consoles, respectively.

The price hikes highlighted the pressures confronting end users of memory chips. Apple shares fell more than 6% on Thursday in the stock’s worst day in more than a year. Microsoft touched a 52-week low before rebounding on Friday.

Yet shares of chip makers slid too, with the PHLX Semiconductor Index falling 7.9%, logging its worst week in more than a year. Broadcom fell 11%, Advanced Micro Devices declined 2.9% and South Korea’s SK Hynix shed 3.3%. Nvidia shares slumped 8.6%, notching their worst week since April 2025 and costing the index heavyweight $439 billion in market value. Palantir Technologies slid 12%.

Even SpaceX struggled, with shares of Elon Musk’s rocket company giving up their stratospheric gains, on Friday briefly dipping below their initial opening-day price of $150 before closing the session up 0.2% at $153.23 a share.


Stocks tied to the so-called real economy held up better. The Dow Jones Industrial Average gained 0.6% this week and closed within 0.2% of all-time highs. The S&P 500’s healthcare sector rose 7.9% to close Friday at a record. Shares of Johnson & Johnson added more than 11%, notching their best week since October 2008 and lifting the company past $600 billion in market value for the first time.

Utility stocks, seen as both a defensive play and a beneficiary of the power required for the AI boom, rose 3.9% this week. Consumer staples, another haven, added 1.5%.

Those gains helped the S&P 500’s equal-weighted version, which gives each stock the same influence instead of weighting companies by market value, outperform the benchmark index by the widest weekly margin since 2020.

Still, analysts said rising rates could weigh on profits in a variety of investments from stocks to bonds and gold. Bank of America economists recently said they expect the Fed to raise rates three times this year by 25 basis points each and hold steady in 2027, citing in part a hawkish shift in the central bank’s communication that signaled a more proactive approach to fighting inflation.


The Fed’s preferred inflation gauge continued growing in May, up by 4.1% over the past year, its highest reading since April 2023 and more than double the central bank’s 2% target.

While oil prices have fallen toward prewar levels, some analysts say that a wave of artificial-intelligence demand and a robust labor market give reason to believe that price pressures could keep inflation elevated. West Texas Intermediate crude, the U.S. oil benchmark, declined 8.7% to $69.23 a barrel this week, while international Brent crude futures fell 11% to $71.99.

Some analysts say that the rally can go on despite higher rates.

Strategists at Barclays recently raised their year-end S&P 500 price target to 7800, arguing that strong earnings growth will keep the rally going despite the potential for higher borrowing costs and more stress on the AI boom.

Christian Chan, chief investment officer at AssetMark, said that while higher rates could weigh on corporate profits, he expects earnings growth to still be robust enough to continue fueling the stock rally. Companies in the S&P 500 are expected to see profits grow 24% in 2026, according to FactSet.

“It’s not a regime-changing event as it relates to earnings growth, which is kind of what the markets ultimately care about,” said Chan.

FT : UK’s Bridgepoint nears deal to buy real estate unit in bet on US property L

UK’s Bridgepoint nears deal to buy real estate unit in bet on US property
London-based buyout group is looking beyond its traditional speciality in corporate buyouts

UK private capital group Bridgepoint is nearing a deal to buy the real estate arm of US investment firm Kayne Anderson for roughly $1bn, in a big bet by the British buyout group on the US property sector.

Kayne Anderson’s real estate arm has about $22bn of assets under management, with investments in some of the fastest growing parts of the real estate sector including medical offices, senior housing and student accommodation.

The potential deal will provide a significant boost to Bridgepoint’s current assets under management of roughly $98bn, pushing it into a new market of property takeovers and lending to the sector.

Talks between the two investment firms are ongoing and a deal is likely to be announced as soon as Monday provided there are no last-minute snags, according to people familiar with the matter. The deal comprises a cash and stock component and will value Kayne Anderson’s real estate arm at around $1bn, they added.

Since listing its shares in 2021, Bridgepoint has sought to grow its assets and diversify into new markets using acquisitions. In 2024, it acquired energy specialist PE group Energy Capital Partners, just ahead of a surge in power prices stemming from data centre demand, which made such investment groups more valuable.

By adding Kayne Anderson’s property arm, London-based Bridgepoint will further push into US markets and beyond its traditional speciality in corporate buyouts. It has also recently used acquisitions to expand into so-called secondaries investments.

Founded in 1984, Los Angeles-based Kayne Anderson has blossomed into a $43bn asset manager, focusing on real estate, the energy sector, infrastructure investments and credit. A sale of its real estate wing will cut its assets under management in half.

Kayne Anderson Real Estate, which was founded and is led by Al Rabil, has built a specialisation in identifying property markets such as senior housing, where it expects ageing populations will bolster demand for properties, pushing their rents and valuations higher.

Last year, Kayne Anderson struck one of its biggest deals to date, paying $7.2bn to carve out 18mn sq ft of medical office assets out of real estate investment trust Welltower alongside its operating partner Remedy Medical Properties.

Earlier this year it also raised a $5.2bn property investment fund, setting a new fundraising record for the firm.

Bridgepoint and Kayne Anderson did not immediately respond to requests for comment

The Information : Salesforce Employees Worry Over Anthropic’s Invasion of Slack

Salesforce Employees Worry Over Anthropic’s Invasion of Slack

The Takeaway
  • Claude Tag competes with Salesforce’s own Slackbot AI product.
  • Anthropic’s Claude growth has pressured traditional enterprise apps
  • Salesforce plans to charge for external AI agent use with its apps.

When Anthropic on Tuesday launched a high-profile AI product for businesses that use Slack, some employees at Salesforce, which owns Slack, were confused.

In social media posts, Salesforce promoted Anthropic’s new product, Claude Tag, which works as a kind of AI teammate that lives inside group Slack chats. It sees and remembers everything that’s been discussed, flags unanswered threads, and handles coding, analytics, file search and other tasks. But Slack itself sells a similar AI product, Slackbot, which coincidentally is powered by Anthropic’s technology.

The new rival to Slackbot wasn’t the only cause for concern.

Some Salesforce staffers, consulting partners and executives at competing enterprise software firms were also shocked that Salesforce would promote an Anthropic product that could give the Claude maker information and context about how customers operate and communicate.

Anthropic did not provide a response to questions about how it might use customers’ Slack data to train or refine its AI, but Anthropic’s terms of service say it doesn’t do that with data from customers with whom it has commercial agreements.

Even so, some of these people privately likened the Claude Tag to a Trojan horse that would give Anthropic more leverage than it already has over the enterprise software industry, particularly as it forges direct relationships with business customers.

“There is risk in any relationship,” said Martin Kihn, a senior vice president of market strategy at Salesforce, in a LinkedIn post. He was responding to a question from a Salesforce business partner about whether there is a “risk that Claude Tag cannibalizes Slackbot.”

Inside Salesforce, employees were having similar conversations about the Anthropic product in their own Slack chat groups. Jaime DeLanghe, Slack’s chief product officer, told colleagues that Slackbot has access to everything a customer does in Slack, “unlike Claude Tag, which only has access to the channels you put it in.”

To some employees, the difference she described seemed minuscule.

‘Crappy Slack Bot’

Anthropic staffer Andrej Karpathy stoked Salesforce employees’ concerns by contrasting Claude Tag with existing products in the market.

“This is not a ‘feature’ like some crappy Slack bot.…” he said in a post on X after the product’s launch. “The difference will become clearer over time.” Days later, the post vanished.


While Slackbot doesn’t code, both companies have positioned their AI products as a new paradigm for how organizations interact with AI and automate all manner of work. Salesforce executives in recent months touted Slackbot as one of the keys to its AI strategy, and CEO Marc Benioff mentioned Slackbot 13 times in a February earnings conference call.

Benioff may have decided it’s better to partner with Anthropic than try to fight it off or watch it launch such a product with a Slack competitor like Microsoft Teams. The CEO said in a text that he views Claude Tag as a partner rather than a Slackbot competitor.

The move shows how Slack has evolved its approach to external AI agents since a year ago, when it moved to block other software firms’ apps from searching or storing Slack messages even if their customers permitted them to do so. Since then it has worked to enable customers to use Slack to summon external AI agents, including coding agents developed by Anthropic and OpenAI.

This week, Salesforce touted the success of Slack as an app or AI agent platform.

“Slack is the best home for all of your agents,” DeLanghe told colleagues. “We strongly believe that keeping the ecosystem open and letting customers choose the AI tools that work best for them (and put them in Slack!) is the best longterm path for success.”

Agent App Store

A Salesforce spokesperson compared Slack’s enabling of apps from other developers to Apple’s App Store. “Apple strengthened the customer experience with third-party apps, and Slack’s open ecosystem is the enterprise distribution platform for third-party AI agents,” the spokesperson said. “Why wouldn’t we want Claude Tag in there along with other agents?”

Claude Tag isn’t the only AI agent that can be summoned by workers using Slack to see and respond to teammates’ messages and take actions. Many other app developers—including Atlassian, which owns software-tracking app Jira, and Perplexity, maker of an AI search app—also offer AI agents in Slack that can connect to various business apps and perform tasks the way Claude Tag does.

But none of those other firms is as powerful and ambitious as Anthropic, and Salesforce hasn’t promoted those apps the way it has Claude Tag. The move as Anthropic has spread fear among app developers as it becomes the dominant seller of AI models to other businesses.

Salesforce and Anthropic declined to discuss the financial terms of their commercial agreements. A Salesforce spokesperson last year told The Information that companies connecting their products to Salesforce apps through its app store pay a flat fee, and some pay a commission on sales.

Still, pressure could rise on Salesforce to explain the long-term benefits of the partnership strategy and how it will avoid losing its status as the central place where customers build and run business applications.

Anthropic this year has shaken up the enterprise software market—and the share prices of Salesforce and other big names in the field—with products such as Claude Code and Cowork that aim to automate white-collar work, which Claude Tag is an extension of. While traditional enterprise software revenues continue to grow, insurgent AI firms like Anthropic and OpenAI have raised questions about whether the old guard will lose influence over enterprises and take a hit to their margins or growth.

Already, some customers that increased spending on AI from Anthropic has have recently moved to shorten their contracts with traditional enterprise app providers.


Salesforce, Microsoft and other enterprise software firms are competing with Anthropic and OpenAI in business automation tools that the AI firms’ employees believe could one day replace some traditional enterprise apps. Firms such as Microsoft and SAP have been trying to protect some of their apps from outside intrusion by AI rivals by blocking their ability to tap data from the apps.

Salesforce and its ilk are meanwhile launching their own AI tools powered by Anthropic and OpenAI technology. So far, though, tools such as Slackbot don’t appear to be helping Salesforce accelerate its overall sales.

Gaurav Kheterpal, a member of a Salesforce advisory board that gives the company feedback on its products, said he was excited yet confused by the launch of Claude Tag, which he sees as “what Slackbot was supposed to be.”

Salesforce and other owners of major enterprise apps are also planning to levy tolls on customers that want to use Anthropic and OpenAI tools—including agents like Claude Tag—to handle tasks that involve data in those traditional apps. Salesforce in particular is planning to charge customers that use external AI agents like Anthropic’s to do work involving data they store with Salesforce’s sales, marketing and other apps. That’s according to an internal company slide about the new program, Headless 360, which The Information viewed.

In the battle between Slackbot and Claude Tag, Salesforce may have a pricing advantage.

An executive of a consulting firm that resells products from both Anthropic and Salesforce to hundreds of businesses said Slackbot was more cost-effective because Slack licenses include usage, whereas Claude charges based on usage and consumes “tokens pretty quickly.”

Salesforce employees expressed excitement in internal Slack group messages about using Claude Tag but encountered error messages when they tried it. A manager said in the Slack thread that the company is working to make the product available to employees, but it “isn’t quite ready yet.”

WSJ : CEOs Are Pulling In Record Pay. Here Are Some Big Numbers. These figures p

CEOs Are Pulling In Record Pay. Here Are Some Big Numbers.
These figures put corporate bosses’ ever-bigger paychecks into context

Pay for chief executives keeps hitting new highs, and it isn’t just Elon Musk’s multibillion-dollar pay package from Tesla.

CEOs making more than $100 million hit a four-year high in 2025 and nearly a dozen chiefs blew past the $200 million mark. More corporate bosses surpassed $50 million as well.

How to put such big numbers into context? Here are the figures that stand out in The Wall Street Journal’s annual CEO pay ranking, based on data from MyLogIQ:

$17.9 million
The middle-of-the-pack, or median, pay for S&P 500 CEOs in 2025, marking a new high.

16
The number you would have to multiply the combined pay of all other CEOs in the Journal’s 2025 ranking by to reach the size of Elon Musk’s $158 billion pay package.
Altogether, the Journal’s ranking includes 392 CEOs of S&P 500 companies, and their collective pay comes to $9.9 billion, excluding Musk. To make the cut, they had to be on the job at least a year, running companies that disclosed executive pay by May 30.

19 minutes
How long it would take Shankh Mitra, CEO of real-estate investment trust Welltower, to match the annual pay of his company’s median employee, spreading his 2025 pay package out over a single year.
Mitra’s package came to $821 million last year, making him 2025’s highest-paid CEO after Musk. The median Welltower employee earned $124,995, according to the company.

$2.75
The smallest pay package for a CEO in the Journal’s rankings. It went to Jack Dorsey of Block (his official title at the payments company is “Blockhead”).
That is actually a bump up from the $1.40 that Dorsey made at the end of his run leading Twitter, now X. It represented a penny for each character in the original limit for a tweet.

19
The percentage of CEO compensation paid, on average, in cash—meaning salary and bonus (excluding Musk’s all-stock bonanza). Most of the rest is in stock or options.

1,894
The ratio of CEO pay to the median employee’s annual income at auto-parts company Aptiv. It is the third-highest ratio after Tesla and Welltower. Pay for Aptiv CEO Kevin Clark was about $19 million, and Aptiv’s median employee made $10,162.
Aptiv said in securities filings that it includes its part-time and temporary employees in determining who its median employee is, including significant workforces in countries with wages below U.S. levels.

$389,488
Warren Buffett’s pay in 2025. All but $100,000 in salary went toward personal and home security expenses. Of course, Buffett’s 13.7% stake in Berkshire Hathaway means he also benefited significantly from the company’s 11% share-price gain in 2025.

$13.4 million
The median pay for CEOs running banks and bank holding companies. That was the lowest for any industry in the Journal’s ranking, though some bank CEOs earned much more. Jane Fraser of Citigroup, for instance, is No. 9 in the overall S&P 500 ranking, at $95.8 million.
The sector doesn’t include financial-services firms such as private-equity company Blackstone and investment banks like Goldman Sachs. For that group, the median CEO pay was $22.1 million.

On the other end of the spectrum, media and entertainment companies had the highest median CEO pay: $53.2 million.

FT : German carmakers embark on historic job cuts as Chinese rivals flood market

German carmakers embark on historic job cuts as Chinese rivals flood market
Threat to industrial model of Europe’s largest economy mounts

German carmakers are embarking on their deepest ever restructuring to stem “the bleeding” from an influx of Chinese rivals that analysts warn could “permanently” shrink the backbone of Europe’s largest economy.

Volkswagen is preparing to expand its cost-cutting measures by axing as many as 100,000 jobs over the coming years and end production at four plants in Germany.

BMW recently warned investors that it plans to spend up to €1bn in restructuring costs, which analysts said could lead to cuts of up to 10,000 jobs and a 15 per cent reduction in European car production. 

Mercedes-Benz told employees in Germany that summer bonuses would not be paid out as it stepped up cost-cutting. Some 5,500 staff have already taken voluntary redundancy under its current restructuring programme.

All European car manufacturers — from Stellantis and Renault to Ford — have been streamlining their operations in recent years, but the penetration of BYD and other Chinese brands has dramatically picked up pace this year amid a sharp slowdown in China. That has forced German carmakers to retrench, despite strong resistance from their powerful unions. 

“The only thing you can do is cut costs, and the only significant cost reduction is excess capacity. And the most expensive capacity you have in the world by a long distance is [in] Germany,” said Citi analyst Harald Hendrikse. 

In May, Volkswagen, Mercedes-Benz, Stellantis and Renault all lost market share despite new car sales in Europe rising 4 per cent year-on-year. Meanwhile, the aggregate market share of BYD, Chery and other Chinese carmakers topped 10 per cent for the first time, according to European car industry group Acea. 

“Every European player is losing today,” said Thomas Besson, head of autos research at Kepler Cheuvreux. “This is a highly challenging situation for European automakers because Chinese [carmakers] are progressing [in Europe] at a much faster pace than expected, while [the European carmakers] continue to lose volumes in China and face very adverse conditions in the US, notably due to tariffs.”

Wolfsburg-based Volkswagen had already laid out its intention to cut 50,000 jobs in Germany by the end of 2030 but the latest plan could lead headcount to be slashed by another 50,000, according to one person familiar with the situation. Culling 100,000 roles from a workforce of around 625,000 would place it among the biggest-ever job cuts.

“I think even with that it’s questionable whether they will get ahead of the [Chinese] wave so that they can actually restore some profitability, rather than just slow down the bleeding,” UBS analyst Patrick Hummel said. 

The size of the potential new cuts showed the crisis affecting Volkswagen and its peers was reaching a “new dimension”, added Helena Wisbert, professor for automotive economics at the Ostfalia University of Applied Sciences in Wolfsburg.

“The automotive industry in Germany is shrinking, and doing so in a lasting, permanent way.”

BMW was the sole European carmaker that increased sales in the continent in May, but the group shocked investors this month with a significant cut to its profit guidance, which was attributed to a market downturn in China and the impact of the Iran war.

Included in the downward revision were provisions for new restructuring measures such as job cuts to be booked later this year.

Since its warning in mid-June, BMW’s shares have slumped 13 per cent with investors preparing for more guidance cuts from other European carmakers. “BMW was seen by almost everybody as the best house in a difficult neighbourhood and that status got lost with this profit warning,” Hummel said. 


The Munich-based manufacturer had already indicated that it expects a reduction in its global workforce this year of up to 5 per cent, which could mean as much as 7,700 people. Hummel estimated that the latest provisions could raise that figure to nearly 10,000.

BMW declined to comment on the planned scope of its cost-cutting measures, while new chief executive Milan Nedeljković has stressed the need for the company to “significantly intensify and accelerate” its efforts to make savings.

Mercedes-Benz, which has also been hit hard by sharply declining sales in China, has warned employees that German manufacturing was weighing on cost competitiveness. The company also estimated that its output would immediately improve 15 per cent if its workers returned to a 40-hour working week from 35 hours that has broadly been in place since 1995. 

“We must continue to cut costs with great urgency so that we can remain price-competitive,” it said. “Despite all our efforts, the situation in Germany today is critical.”

>>> Barron's Weekend Summary

Cover Story:
-Cheap drones are reshaping the defense sector by proving that low-cost, mass-produced systems can threaten or destroy weapons platforms that cost hundreds of Ms of dollars. The Iran war reinforced the lesson already visible in Ukraine: spending heavily on exquisite systems no longer guarantees battlefield advantage when cheap drones can overwhelm defenses, expose logistics, and force militaries to adapt procurement. Barron’s identifies Kratos Defense & Security Solutions, AeroVironment, RTX, and L3Harris Technologies as four listed companies positioned to benefit from rising demand for drones, counter-drone systems, sensors, electronic warfare, and networked battlefield technologies.

CEO Interview:
-No update

Tech Trader:
-Micron Technology is strategically leveraging its market position in the memory chip industry to reshape long-term business dynamics and potentially enhance its stock valuation. Historically, the memory business has been volatile, characterized by cyclical fluctuations in inventory and price. This volatility has led to low price-to-earnings (P/E) ratios for memory firms like Micron, which recently faced significant challenges post-pandemic, including a negative gross margin for several quarters. However, as demand for memory surged due to AI investments, Micron adopted a cautious approach to capital expenditures, leading to an unprecedented memory shortage and price increases.

The Trader:
-Tech stocks stumbled sharply this week as investors questioned whether the artificial intelligence rally has become overextended, yet the broader market continued to demonstrate remarkable resilience. While the NASDAQ retreated amid profit-taking in semiconductor and AI-related names, economically sensitive sectors and value stocks absorbed much of the selling pressure, suggesting that the bull market is broadening rather than deteriorating. Barron's argues that this sector rotation is constructive, indicating investors are reallocating capital instead of abandoning equities altogether. The article advises investors to look beyond the largest AI winners and focus on companies with improving earnings momentum and more reasonable valuations, as leadership gradually expands beyond mega-cap technology.
-Despite a difficult period for the renewable energy sector, Barron's believes the long-term investment case for solar remains intact. The article highlights one solar company whose shares have been pressured by higher interest rates, policy uncertainty and slowing project development, but whose underlying business continues to benefit from rising electricity demand driven by artificial intelligence, data centers and electrification. As utilities invest heavily in new generating capacity, demand for solar installations is expected to recover, providing an attractive entry point for long-term investors willing to tolerate short-term volatility. Barron's argues that improving fundamentals could allow the stock to outperform as financing conditions stabilize and renewable energy spending resumes.

Features:
-Alphabet has experienced volatility in the stock market, transitioning from an AI frontrunner to an underperformer, particularly following its recent inclusion in the Dow Jones Industrial Average (DJIA). After reaching record highs over $400 in May, the stock has since dropped approximately 15%, driven by significant challenges such as the departure of key AI researchers and an $85 B equity raise, leading to a drastic $225 B loss in market capitalization one day, marking its largest single-day drop ever. Despite hopes that its addition to the Dow would bolster confidence, as S&P Dow Jones Indices highlighted Alphabet’s AI products and its relevance in dynamic sectors of the U.S. economy, the stock fell another 1% afterward. The unique nature of the Dow, being a price-weighted index, limits the potential benefits from its inclusion, as few funds mirror the index due to its weighting methodology based on stock prices rather than market capitalization, resulting in minimal forced buying compared to a potential boost from an addition to the S&P 500.
-In a digital age increasingly dominated by technology, concerns such as screen-time eyestrain and the adverse effects of social media on mental health highlight the importance of taking breaks from online platforms. Off-price retailers like Ross Stores demonstrate that traditional retail can succeed amidst the rise of e-commerce. Despite a significant 80% surge in Ross's shares over the past year, it posted a remarkable 17% increase in comparable sales in its latest quarter, a figure described as “unprecedented” by Guggenheim analyst Simeon Siegel.

European Trader:
-A private cloud-computing firm, Argentum AI, backed by Super Micro Computer, has secured significant AI infrastructure deals totaling $7.8 B, highlighting robust demand for Nvidia chips and AI proliferation outside the U.S. The company has finalized two major agreements to deploy 47,000 Nvidia GB300 chips at a 300 megawatt data center in Poland, with one contract valued at $4.1 B over five years and the other at $3.7 B, beginning operations this year. Super Micro has invested $100 M in Argentum through a convertible note. Following the showcasing of new AI infrastructure designs for Nvidia chips, Super Micro's stock surged by 16% but later dipped by 5.6% amidst a broader tech selloff.

Emerging Markets:
-Obama’s former lead negotiator on the Iran nuclear deal, Wendy Sherman, expressed concerns regarding the implications of the ongoing negotiations between the Trump administration and Iran, particularly how they could influence China's strategic ambitions, including its stance on Taiwan. Current discussions around a peace deal reveal disagreements over Iran's potential tolls on shipping through the Strait of Hormuz and its ability to sell oil in dollars to U.S. buyers. While a volatile market has affected oil prices, falling rates provide some relief to Asian economies, although the overarching economic damage remains significant. Sherman's insights raise important questions about the global economic landscape's response to these geopolitical developments.

Commodities:
-Rising bond yields are becoming a significant concern for the stock market, influenced by Federal Reserve Chairman Kevin Warsh’s hawkish stance. The Fed has revised its 2026 GDP growth forecast down to 2.2% and increased its PCE inflation estimate to 3.6%. Odds of a September Fed rate hike have risen to approximately 75%. Despite lower oil prices and progress in U.S.-Iran talks, Treasury yields increased, with the 2-year note reaching 4.22%. Investors are adjusting expectations for rates and growth, anticipating further data that may support continued hawkish sentiment among Fed officials. Nonetheless, the outlook for stocks remains bullish according to RBC Capital Markets.

Streetwise:
- Lululemon Athletica, once a standout in the stock market, has seen its shares plummet 45% in 2023 and 60% over the past five years, marking it as one of the worst performers among S&P 500 companies. The company debuted in July 2007, achieving a rapid increase in share prices from an initial $18 to over $500 by the end of 2023, yielding over 3,500% returns for investors. However, its valuation has since dropped significantly from 36 times forward earnings estimates to around 10 times, reflecting concerns about its future. In the late 1990s, Lululemon's founder Chip Wilson capitalized on the rising popularity of yoga, transitioning the perception of yoga wear from casual to chic and premium. By creating innovative designs, such as a diamond-shaped gusset for comfort and shapewear-like high waistbands, Lululemon's products not only improved functional aspects but also became fashionable for everyday wear. This innovation spurred the athleisure trend, which allowed athletic clothing to be worn beyond the gym, significantly boosting the company’s revenue, which reached $275 M prior to going public, growing by 85% largely through word-of-mouth marketing.