Author: Eldin Yovlz

McKinsey Fears AI’s Existential Threat to Consulting
For nearly a century, McKinsey & Company has occupied a rarified perch as the elite advisor to CEOs, governments, and boardrooms across the world. Its consultants—often MBAs from the likes of Harvard, Wharton, and Stanford—once commanded deference bordering on awe. Their PowerPoint slides were gospel, their recommendations the blueprint for billion-dollar strategies. But today, that aura faces a threat no competitor has ever posed: Artificial Intelligence.
In an era where software can generate business plans, synthesize reports, and even produce McKinsey’s trademark crisp prose in seconds, the entire business model of Big Consulting is under siege. And no firm feels the heat more acutely than McKinsey itself. As The Wall Street Journal reported, the firm’s own executives admit the scale of the challenge is nothing less than “existential.”
“AI can increasingly do the work done by the firm’s highly paid consultants, often within minutes,” WSJ observed in its feature.
The Business Model That Built McKinsey
To understand why AI is such a disruptive force, you have to appreciate how McKinsey and its peers like Bain and BCG have made their money for decades.
Leverage is the foundation: partners sit atop pyramids of junior consultants, who do most of the heavy lifting—research, analysis, modeling, and slide production. A single engagement manager might lead a team of 10 or more. For each month of work, McKinsey charges six- or seven-figure fees, often billing clients three times the cash compensation of each junior staffer.
This approach has always depended on a perception that armies of bright, credentialed young analysts were irreplaceable. That perception is now crumbling. As Bob Sternfels, McKinsey’s global managing partner, told the Journal:
“AI is now a topic of conversation at every meeting of McKinsey’s board.”
It’s no wonder. AI systems can perform in seconds what once took weeks: consolidating reports, summarizing interviews, even generating presentations in the firm’s own voice. McKinsey’s most-used AI bot today isn’t even an analytics tool—it’s a writing assistant that drafts documents in the “McKinsey tone.”
A Smaller Pyramid—and Shrinking Profits
Traditionally, a strategy project required an engagement manager plus 14 consultants. Today, it might need just two or three consultants—and a suite of AI agents that never sleep or take vacations.
“Clients will expect much smaller and more senior-heavy teams,” noted a former McKinsey partner in The Financial Times. “That throws a wrecking ball into how McKinsey generates enough revenue to produce the lofty pay to which its top brass has become accustomed.”
If AI can automate the legwork, clients inevitably ask: Why pay seven figures for a deck we can largely assemble ourselves?
The strain is already visible. McKinsey has shrunk its global headcount from 45,000 in 2023 to 40,000 in 2025, according to the The NY Budgets—a staggering reduction for a firm that grew rapidly over the past two decades.
AI Advising: Lifeline or Illusion?
McKinsey’s strategy, for now, is to pivot: from being a firm that helps clients “do strategy,” to being one that helps them implement AI. As Wired reports, the firm claims advising on AI now accounts for 40% of its revenue.
Skeptics question whether that figure is inflated. “It’s defined as broadly as possible to sound like they’re at the bleeding edge,” one consultant told Bloomberg. Even if true, that shift doesn’t solve the core problem: AI is commoditizing much of what made McKinsey indispensable.
In the dot-com era, a rush of new demand offset the commoditization of some advisory work. But in 2025, that surge hasn’t materialized. Corporate budgets remain tight, and procurement departments are under pressure to justify every dollar spent on consultants—especially when a chatbot can produce first drafts of the same slides.
A Recruiting Challenge Years in the Making
McKinsey has long marketed itself as the ultimate destination for elite graduates—a place where brilliant young people can earn big paychecks and a ticket to the C-suite. But if AI deflates profit margins, the economics supporting that model begin to unravel.
As a New York Times retrospective on McKinsey’s past noted, the firm once responded to similar pressures in the 1980s by raising partner pay to keep up with Wall Street and private equity. To fund the increases, it expanded the size of junior teams. That strategy simply won’t work in the age of AI, when clients want leaner teams and more automation.
Even McKinsey’s unique brand of “eliteness”—the intimidation factor of sending a dozen Ivy League MBAs to a meeting—has eroded. As one ex-partner told The NY Budgets:
“Today, the executive suite is stocked with people who went to the same schools, worked at the same banks, and read the same case studies. They no longer feel like they need McKinsey to translate the world for them.”
Inside the AI Transformation
The firm has responded with an all-out push to integrate AI internally. Thousands of AI agents now help consultants assemble PowerPoint decks, take notes, and check logical consistency.
McKinsey is piloting models to auto-generate detailed strategy recommendations. The company is exploring how AI can facilitate “co-creation” with clients, moving away from static reports to real-time workshops and simulations.
Yet even these changes come with risks. A growing chorus of observers worries that AI-generated analysis will create an illusion of certainty and precision—while propagating errors. A recent FT op-ed warned:
“There’s an unspoken fear that clients will treat AI outputs as gospel, when in fact many are based on data that is incomplete, outdated, or wrong.”
A Future of Smaller, Nimbler Firms?
Ironically, the same forces threatening McKinsey’s leverage model may benefit smaller consulting shops. With AI leveling the playing field, boutique firms can compete more effectively—especially those with specialized knowledge or access to proprietary networks.
As Nick Studer, CEO of Oliver Wyman, told the Journal:
“Companies don’t want a suit with PowerPoint. They want someone who is willing to get in the trenches and help them align their team.”
McKinsey, for its part, has begun offering more outcomes-based contracts—tying compensation to measurable results rather than billable hours. Roughly a quarter of the firm’s revenue now comes from these agreements.
None of this means consulting is dying. But the industry is in the midst of an epochal shift—and no firm embodies the stakes more than McKinsey.
If AI keeps advancing, if clients keep demanding leaner teams, and if junior staff no longer generate leverage at historic margins, the world’s most famous consultancy may look very different a decade from now.
As one ex-partner told Wired:
“McKinsey used to be the firm you hired because no one ever got fired for hiring McKinsey. But when your board asks, ‘Why didn’t you just use AI?’—what will you say then?”
August 1, 2025
I Turned Down Mark Zuckerberg’s $1 Billion Job Offer
I Turned Down Mark Zuckerberg’s $1 Billion Job Offer. Illustration: © NYBudgets/Getty Images
Stock Widget
In an era where billion-dollar valuations and nine-figure compensation packages have become the new arms race in artificial intelligence, one story out of Silicon Valley has taken even hardened tech veterans by surprise: Meta META -1.50% ▼ CEO Mark Zuckerberg reportedly offered more than $1 billion to a single AI expert — and was turned down.

According to Wired, which broke the story, Meta has been relentlessly courting AI scientists from Thinking Machines Lab (TML), an elite AI startup co-founded by former OpenAI CTO Mira Murati. In its most brazen effort yet, Meta allegedly floated an eye-watering $1 billion compensation package to a single individual — a sum to be paid over several years. The response? A polite but firm “No, thanks.”
Zuckerberg, it seems, is pulling every lever possible to accelerate the buildout of his new AI division — Meta’s Superintelligence Labs — in a frantic attempt to catch up with rivals like OpenAI, Google DeepMind, and Anthropic. But the refusal by TML’s team — not one of whom accepted any offer — paints a telling portrait of the current AI climate: money isn’t everything, even when the checks have nine zeros.
Meta’s effort to poach elite AI talent stems from its broader push into developing what Zuckerberg has dubbed “personal superintelligence.” The company is racing to become a leader in AGI (Artificial General Intelligence), launching its Superintelligence Lab earlier this year with splashy announcements and big names.
According to Wired and confirmed by Meta spokesperson Andy Stone, the company made several offers to Thinking Machines Lab employees. Stone disputed the $1 billion figure but did not deny the aggressive recruitment campaign. In addition to hefty salaries, insiders say Meta dangled enormous equity grants and bonuses — some of the most lavish offers ever made in the tech world.
The company is “putting its chips on the table,” according to The New York Budgets, pumping billions of dollars into computing infrastructure, AI models, and research teams in hopes of catching up to more advanced competitors.
In one leaked message obtained by Wired, Zuckerberg personally wrote to a recruit:
“We’ve been following your work on advancing technology and the benefits of AI for everyone over the years. We’re making some important investments across research, products and our infrastructure in order to build the most valuable AI products and services for people.”
But the recipients of those messages — and the money — have largely said no.
Why They Turned It Down
Despite Meta’s immense offers, Thinking Machines Lab employees have remained loyal — a rare move in an industry where “exit packages” and stock grants often win over even the most mission-driven minds.
Multiple factors explain the mass refusal.
1. Leadership Concerns
Sources told Wired that many TML scientists are skeptical of Meta’s newly appointed AI leaders — most notably Alexandr Wang, the 28-year-old founder of Scale AI, who was brought in to help lead Superintelligence Labs.
Wang’s leadership style was described as “questionably experienced” for a project of such scale, and insiders were worried that his “move fast” startup mentality could clash with the rigorous demands of cutting-edge AI research.
2. Meta’s PR & Trust Problem
Meta’s recent stumbles — including the botched rollout of its LLaMA 4 language model — have damaged its reputation among AI researchers.
According to The Verge, Meta allegedly manipulated benchmark scores for its AI model to appear more competitive than it actually was. That undermined confidence in the company’s transparency and scientific rigor — values deeply held within AI research communities.
3. TML’s Own Success
Thinking Machines Lab is no underdog. In just 12 months, the company has raised the largest private AI funding round in history, reaching a $12 billion valuation. With ample capital, it can afford to pay its top talent generously and stay competitive without losing its soul to a corporate behemoth.
A Broader Bubble?
Zuckerberg’s billion-dollar offer is more than a headline — it’s a signal of what some analysts fear is becoming a bubble in AI.
In interviews with The Wall Street Journal, venture capitalists and economists warn that the current investment frenzy feels eerily reminiscent of the dot-com bubble of the late ’90s. Companies are making enormous bets on models and infrastructure that are still far from producing significant revenue — or, in many cases, even working reliably.
Meta’s desperation to hire top-tier talent reflects a fear of missing out on what may be the defining technological revolution of the century. But this gold rush mentality also suggests some companies — Meta chief among them — may be sacrificing long-term stability for short-term wins.
Is Zuckerberg Losing His Touch?
The optics of Zuckerberg — one of the richest and most powerful men in tech — offering a billion dollars to a startup employee only to be turned down may raise questions about his current influence and Meta’s direction.
Once a juggernaut that could acquire or hire anyone it wanted, Meta is now struggling to retain trust, especially in areas as ethically and philosophically charged as artificial intelligence. Some in the AI world still associate Meta with surveillance capitalism, political controversy, and internal leaks that have revealed a troubling culture of “build first, ask later.”
The Power Has Shifted
The incident with Thinking Machines Lab reflects a larger truth: the power dynamic has flipped. Top AI researchers are now the prize — and they are choosing where to go, who to trust, and how they want to contribute to the future of intelligence.
One former OpenAI engineer, speaking anonymously to TechCrunch, said:
“We don’t want to just build smarter systems. We want to build responsibly, with values. And that means saying no — even to a billion dollars.”
Zuckerberg may yet succeed in staffing up his Superintelligence Lab. But this high-profile rejection shows that even unlimited funds can’t buy loyalty — or vision.
And for once, Silicon Valley is learning that a bigger check doesn’t always get the yes.
August 1, 2025

Intel to Lay Off 15% of Workforce, Cancel Billions in Projects as Part of Strategic Rebound

Intel Corporation has unveiled a sweeping restructuring plan aimed at reversing its declining market position amid rapid competition in artificial intelligence (AI) and semiconductor manufacturing. The company announced it will lay off 15% of its global workforce—approximately 24,000 jobs—and cancel multi‑billion‑dollar factory projects in Europe and the U.S.

In its Q2 2025 results, Intel posted $12.86–$12.9 billion in revenue, slightly ahead of consensus expectations, yet absorbed a sharp net loss of $441 million to $2.9 billion, largely due to $1.9 billion in restructuring charges and impairments. This marked a steep fall from the small profit it reported a year earlier.

CEO Lip‑Bu Tan, who took charge in March 2025, emphasized that the firm will no longer support expansion without confirmed customer demand—declaring an end to the era of “blank check” spending on chip fabs and platform development (notably 14A and 18A process technologies) until demand is locked in.

The layoff targets include middle management across all divisions, shrinking Intel’s headcount from ~109,000 in 2024 down to ~75,000 by end‑2025, reflecting a ~15% reduction globally. A specific focus is on Intel Foundry, with an estimated 10,000 to 20,000 roles impacted globally (roughly 15–20% of that division), spanning technicians, engineers, and factory support roles. No severance or voluntary buyouts will be offered, with decisions based on performance and strategic alignment. Intel has officially canceled planned fab projects in Germany and Poland, delayed construction in Ohio, and shut down packaging operations in Costa Rica—moving some functions to Vietnam and Malaysia.

Intel is doubling down on core pivot areas:

Strategic Area	Details
AI and Foundry Business	Foundry revenue grew 3% in Q2 to $4.4B—but Intel remains far behind TSMC in client traction.
Technology Development	Delay or pause in the rollout of next-gen nodes—like 14A and 18A—if anchor customers aren’t secured.
Organizational Streamline	Reduction of management by ~50%, with a full return-to-office mandate starting September to.

Analysts warn the AI boom has largely benefited Nvidia, AMD, and TSMC—not Intel, which has yet to meaningfully capitalize on generative AI demand.

Intel shares fell nearly 4% after hours following the earnings call—closing the day up ~13% year-to-date. Despite outperforming on revenue expectations, the net loss and aggressive restructuring plan rattled investors.

Analysts at Benchmark Research maintain a “Hold” rating, citing uncertainty over market share recovery and external foundry reliance if internal execution continues to falter.

“These decisions reflect painful but necessary discipline,” said an analyst at Moor Insights. “Intel must align its operations to real demand—or risk bleeding further cash chasing future tech without customer backing.”

Others emphasize the broader context:

The company’s earlier $1.6B Q2 loss under former CEO Pat Gelsinger had already prompted a 15,000‑job cut in 2024—part of a longer pattern of retreat from leading-edge ambitions without roadmap delivery.

Faced with mounting losses and faltering leadership in AI chip making, Intel’s dramatic cost-cutting and strategic reset represent a high-stakes gamble. By halting unchecked expansion and refocusing on demand-driven investments, CEO Lip‑Bu Tan aims to forge a leaner, more disciplined Intel. Success will hinge on execution—and whether the company can reclaim relevance in a silicon boom it largely missed.

July 27, 2025

HPE is owed nearly $1 billion by the estate of U.K. tycoon Mike Lynch, who died after his luxury yacht sank
Hewlett Packard is owed nearly $1 billion by the estate of the late Mike Lynch and his former business partner over HP’s acquisition of their British software firm Autonomy, London’s High Court ruled on Tuesday.
HP was seeking to recoup its losses from Lynch – who died last year when his luxury yacht sank off Sicily– and Autonomy’s former chief financial officer, Sushovan Hussain.
The technology giant welcomed Tuesday’s ruling, which said HP was owed $944 million in relation to the difference between the price it paid and the price it would have paid for Autonomy had it known its “true financial position.”
In addition, HP is entitled to another $47.5 million for losses suffered by Autonomy group companies in relation to hardware sales and other transactions.
A spokesperson for Lynch’s family released a statement, which the spokesperson said Lynch prepared last year before his death, having seen a draft of Tuesday’s long-delayed ruling.
Lynch said in the statement that the judgment showed HP’s initial claim for up to $5 billion was a “wild overstatement.”
A further hearing will take place in November, to determine any applications for permission to appeal and how damages to be paid will be divided between Lynch’s estate and Hussain, with whom HP settled earlier this year.
HP sued Lynch and Hussain in 2015, accusing them of masterminding an elaborate fraud to inflate the value of Autonomy, which HP bought for $11.1 billion in 2011.
The deal spectacularly unraveled in less than a year and HP wrote down Autonomy’s value by $8.8 billion within a year before bringing a $5 billion lawsuit against Lynch and Hussain in London.
‘Considerably less’
The High Court ruled in HP’s favor in 2022, though a judge said the company would receive “considerably less” than $5 billion.
Lynch, once hailed as Britain’s answer to Bill Gates, had always maintained his innocence and blamed HP for failing to integrate Autonomy into the company.
He was acquitted of criminal charges over the deal in the US and had intended to appeal the High Court’s 2022 ruling, a process which was on hold pending Tuesday’s decision on damages.
Judge Robert Hildyard ruled that HP would have paid 23 pounds a share, rather than the 25.50 pounds it actually paid, when it bought Autonomy.
HP had been seeking up to $4 billion, its lawyers said at a hearing last year.
July 22, 2025
TSMC Stock Rises. Outlook Is Bright as the AI Chips Boom Outweighs Tariff Fears
Shares of Taiwan Semiconductor Manufacturing Company (NYSE: TSM) climbed Thursday after the world’s largest contract chipmaker reported record-breaking second-quarter profits, driven by booming demand for artificial intelligence (AI) chips. Despite global currency headwinds and rising concerns over U.S. tariff policy, investors appear confident that AI tailwinds will continue to drive TSMC’s growth for the foreseeable future.
TSMC reported net income of $9.4 billion for Q2 2025, a new quarterly record and up 22% from a year earlier. Revenue came in at $20.2 billion, also beating analysts’ expectations, as the company continued to benefit from its dominant position as the manufacturer of choice for advanced chips powering everything from AI data centers to smartphones and autonomous vehicles.
The earnings beat was largely attributed to explosive demand for high-performance chips used in AI training and inference—particularly from major clients like Nvidia, AMD, and Apple.
“AI is no longer just a future growth theme—it’s here, and it’s driving volume at the cutting edge,” said CEO C.C. Wei during TSMC’s earnings call. “Our 3nm and 5nm technologies are in high demand, and we expect this momentum to accelerate into 2026.”
TSMC’s advanced technology nodes (5nm and below) now make up nearly 59% of total wafer revenue, a significant increase from 48% a year ago.
Following the earnings release, TSMC’s ADRs rose 3.6% to close at $168.42, marking their highest level since February. The company also issued a bullish outlook for Q3, projecting revenue between $21.0 billion and $21.8 billion, and a gross margin between 52.5% and 54%—stronger than Wall Street estimates.
Analysts hailed the results as another signal that TSMC remains central to the global semiconductor supply chain, especially as AI workloads expand across cloud, edge, and enterprise infrastructure.
“TSMC continues to deliver operational excellence while capitalizing on the AI supercycle,” said Chris Danvers, semiconductor analyst at EverBright Research. “Even with external risks, their pricing power and technological leadership remain unmatched.”
One shadow over the otherwise sunny outlook is the growing uncertainty surrounding U.S. trade policy. Washington has been evaluating new tariffs on high-end chip imports as part of broader efforts to bolster domestic manufacturing and reduce dependency on Asia. While Taiwan has historically enjoyed favorable treatment, policy shifts could still impact TSMC’s U.S. customer base and logistics.
Still, company executives downplayed the immediate risk of trade restrictions, stating that long-term supply agreements and geographically diversified facilities—including TSMC’s new Arizona fab—provide a cushion against potential policy shocks.
“We’re monitoring the policy environment closely,” said CFO Wendell Huang, “but our global footprint positions us well for resilience and flexibility.”
TSMC acknowledged that a stronger Taiwan dollar and volatile foreign exchange rates trimmed its revenue slightly in USD terms, but not enough to derail its earnings beat. Operational efficiency and high-margin AI-related products helped protect its bottom line.
The company’s gross margin for Q2 was 53.9%, up from 51.5% last quarter, reinforcing investor confidence in its ability to maintain profitability even amid macroeconomic uncertainty.
TSMC reiterated its 2025 capital expenditure forecast of $32–$36 billion, underscoring its aggressive push to expand capacity at the leading edge. Much of this investment is tied to facilities in Taiwan, Japan, and the United States.
Notably, the company’s U.S.-based Arizona plant, expected to begin partial operations in late 2025, is seen as a strategic hedge against geopolitical risk and U.S. localization pressures.
TSMC’s stock has gained more than 47% year-to-date, outperforming the broader semiconductor index (SOX) and peer rivals such as Intel and Samsung. The strong Q2 print and guidance are expected to drive bullish revisions to analyst targets.
Currently, 29 of 33 analysts tracking the stock rate it a “Buy” or “Strong Buy,” according to Bloomberg data.
TSMC’s record-breaking second quarter confirms its unmatched position at the heart of the AI chip boom. While global economic pressures and geopolitical tensions continue to loom, the company’s cutting-edge technology, diversified client base, and bold capital investments are positioning it for long-term dominance.
As artificial intelligence continues to expand across industries and continents, TSMC stands not just as a beneficiary—but as the backbone of the next era of computing.
July 18, 2025
Crypto Industry Scores a Win as Congress Passes Stablecoin Bill
The crypto industry notched a major victory on Thursday, securing legislation that could lead to digital assets becoming a significant part of Americans’ everyday lives. But delays in enacting the bill shows the industry’s power still has limits.
On Thursday afternoon, the House of Representatives in a 308-122 vote passed a bill that would set rules for so-called stablecoins, a type of cryptocurrency whose value is most often pegged to the dollar and backed by reserves. The Senate already passed the bill in June, and the White House on Thursday said President Donald Trump will sign it into law as soon as Friday.
The bill, called the Genius Act, has been a longstanding target for stablecoin company Circle Internet Group and crypto trading platform Coinbase Global. Its passage is the culmination of a multiyear effort to lobby lawmakers over to crypto’s cause—and to finance the campaigns of others who promised to support the industry.
Crypto supporters largely agree that the bills considered by Congress could transform the sector and allow for more investments, especially from institutions.
The crypto industry still has much it wants to accomplish even after Thursday’s expected victory. High on crypto boosters’ wish list is legislation to establish rulesfor crypto exchanges, brokers and tokens. But it will be more difficult for the industry to build the coalition it needs to push through that larger agenda.
On Thursday, the House also passed a bill on a 294-134 vote to establish those other rules. Seventy-eight Democrats voted in favor of the bill, more than the 71 Democrats who voted in favor of a separate bill to create crypto rules last year.
Unlike the stablecoin bill, the Senate has yet to vote on market-structure legislation.
The Senate seems a long ways away from building the bipartisan support needed to avoid a filibuster and move it into law, wrote TD Cowen analyst Jaret Seiberg in a research note this week.
“Passing this bill is symbolically important, but what will matter is the language that the Senate can pass,” Seiberg said.
Seiberg said he doesn’t expect to get full details on what the Senate plans until late this year or early next year.
Part of the hangup is that while several senators for years have pushed digital-assets tied legislation, the Senate and its committees as a whole haven’t done nearly as much work building a consensus as their counterparts in the House. Former Sen. Sherrod Brown (D., Ohio), who chaired the Senate Banking Committee, was a crypto skeptic, and he and other progressive Democrats largely froze bills moving forward even as the Republican-led House forged ahead.
The second hang-up is substantive. Some Democratic senators, including Sen. Elizabeth Warren (D., Mass.), have argued that the crypto bills would let the industry run roughshod over
investor protection laws and leave enforcement to undermanned agencies like the Commodity Futures Trading Commission not used to policing a large industry. They say that current investor laws that pertain to securities are already good enough.
Not helping matters is Trump’s own crypto ventures. The stablecoin bill itself almost faltered in the Senate after Democrats expressed concerns that it didn’t prohibit Trump or other government officials from profiting from the coins.
The Trump family owns an interest in crypto firm World Liberty Financial. Its token sales generated more than $57 million for the president, according to anethics disclosure, and the firm this year launched its own stablecoin. Some Democrats will no doubt push for restrictions on Trump profiting off crypto in a new bill.
July 18, 2025
How the Mayor of San Jose Is Working to Build an AI Capital
Nearly 150 stoplights in San Jose, California, are equipped with an artificial intelligence tool aimed at optimizing bus trips. The tech has allowed the buses to run at higher speeds and reduced commute times for riders by 20%, in part by making it more likely buses will reach a green traffic light.
“We know that what meaningfully drives up ridership levels is the frequency of the service, how often the buses come, and the speed of getting to your destination,” said San Jose Mayor Matt Mahan, noting the program will now be scaled citywide. “A more than 20% improvement in commute times is a big deal.”
Mahan sees the signal priority technology as one of San Jose’s most successful AI implementation efforts to date. The Silicon Valley city is trying to vastly expand its use of the technology in city hall and across government, Mahan said. It’s part of his administration’s broader push to make the city a hub for using AI tools in government, and a destination for AI companies and talent.
The city is one of a number of places exploring such initiatives. But San Jose is the founding member of the GovAI Coalition, which includes hundreds of government entities across the US that share information about AI-related projects, safeguards and rules, and procurements. The city has offered up to $50,000 in incentives for early-stage AI startups that relocated there. It’s also rolling out an AI upskilling course for city workers in collaboration with San Jose State University to teach them how to better use the technology.
The interview has been edited for length and clarity.
Could you talk to me a bit about the city’s AI ambitions?
We want to be the most AI-enabled city hall in the country, and we started on this journey before we even really knew what AI was. If you go back about eight, nine years ago, we were starting to look at public safety technologies and realized that there are privacy issues that we need to work on. (The city has used license plate reader technology for nearly two decades, and developed a use policy for the technology in 2017.) And so we really started focusing on data privacy and data security related to applications that had nothing to do with AI, but it gave us the muscle.
So that was the beginning. About two years ago, we invited some other cities to join us on a very informal monthly Zoom call to just talk about artificial intelligence because it was starting to emerge as essentially a trend. And that blossomed into this platform called the GovAI Coalition, which we host on our city website. And we now convene with partners from over 700 public agencies around the country.
What are some of the projects you’re working on?
We’ve driven some of the leading pilots on object detection. So on roadways, we put sensors on city vehicles. We can very accurately identify emerging potholes, graffiti, illegal dumping, street lights that are out, lived-in vehicles, and a number of streetscape issues where we can automate reporting.
Language translation — we actually have the largest Vietnamese population of any city outside of Vietnam in San Jose. So we have a lot of training data that’s native to our city based on our population. We took our training data in Vietnamese, plugged it into Google’s base model for Google Translate, and actually have improved by 8% translation of government websites and government documents into Vietnamese.
How are you thinking about privacy, especially around law enforcement technologies?
There’s been growing skepticism, if not cynicism, that the technology sector wants to own our data to monetize it, even if it’s not in our interest. And so I guess we just culturally felt that if we’re going to bring best-in-class technologies and tools into city hall, we can’t fall into that trap.
Government can’t afford that kind of violation of trust. A startup can fail. Government really can’t. So we felt that getting that piece right was really important. And I shouldn’t say it that way. There is no getting it right. It’s about ongoing dialogue and transparency.
We’re rolling out — and this is not a novel technology — red light and speed safety cameras, which there have been some limitations on in California. We’re a pilot city in the state to start rolling out more of what some might consider aggressive surveillance and enforcement tools around traffic safety. We have a Vision Zero plan. We’re trying to make our roads safer, but we have taken the time to get it right. We’re going around to different neighborhoods and having community meetings and, for example, assuring people that we’re not doing facial recognition, assuring people that we’re going to equitably distribute these.
If you don’t answer thoughtfully and engage people and create a forum for them to ask those hard questions, I think you can really risk creating cynicism and making it harder in the long run to get things done.
When it comes to the economy, San Jose is a tech and AI hub already. A new Brookings Institution report dubbed San Jose and San Francisco AI “superstars” among US jurisdictions, finding that the two metro areas around these cities have excelled in adoption, talent concentration and innovation.
City government is piloting all of these initiatives and at the same time creating new incentives for startups and other new entrants. Is there a point where it could be too much of one industry?
I do worry a little bit about diversification. At the same time, we’re one of the few big cities in the country that still has a meaningful manufacturing base. So while San Francisco has gotten a lot of headlines around AI related to software applications, we’re really still a hardware town in a lot of ways. So we do robotics, we do vertical takeoff and landing vehicles, batteries. Our history was really in semiconductors.
I’m trying to think about how we leverage the fact that we’re still a city that makes things — almost one in five workers in our city is in manufacturing— and continue to be a city of economic mobility. And so our challenge is to make sure that we continue to say yes to manufacturing.
We have this incredible asset in San Jose State University. It’s like the pipeline for young diverse talent in the South Bay. We’ve never had an intentional strategy for helping those graduates start companies and stay local. And so one of my major initiatives is to create a startup cluster in downtown San Jose adjacent to the university so that we can keep more of that young talent there actually build their own economic future. That’s the thing we’re excited about.
July 18, 2025
Xiaomi’s stock price has reached new highs following a strong reception for its new electric vehicle, which is priced to compete aggressively with Tesla
Hong Kong-listed shares of China’s Xiaomi surged over 5% to hit a record high on Friday, a day after its electric car drew a strong response from customers.
The consumer electronics company, a relatively newer player in the EV market, took aim straight at rival Tesla with its new electric luxury vehicle, YU7. The SUV’s pricing starts at 253,500 yuan ($35,322), CEO Lei Jun said Thursday, pointing out that the vehicle was 10,000 yuan cheaper than Tesla’s Model Y, which starts at 263,500 yuan in China.
The YU7 received more than 200,000 orders within just three minutes of its launch, Xiaomi said.
Prior to the official price announcement, a Citi report had listed expectations that the YU7 SUV would be priced around 250,000 yuan to 320,000 yuan ($34,800 to $44,590), forecasting monthly sales of about 30,000 units. Once the pace picks up, Citi predicts annual sales of 300,000 to 360,000 units.
Xiaomi’s company’s SU7 sedan launched last year was also priced below Tesla’s Model 3.
Lei on Thursday claimed the YU7 beat Tesla’s Model Y on a range of metrics, but still came short on driver assist. The YU7 comes with driver-assist software, the most advanced version of which is powered by Nvidia’s Thor chip. Pre-sales start at 10 p.m. on Thursday, with deliveries expected within one to five weeks.
Xiaomi had initially said it would launch its YU7 in July. The earlier event takes place amid an intensifying electric car price war.
Xiaomi revealed its YU7 SUV in late May, less than a year after launching its first electric car, and claimed the vehicle would have a driving range of at least 760 kilometers (472 miles) on a single charge.
That’s well above the 719 kilometers advertised for Tesla’s extended-range Model Y. Driving range has been a selling point for consumers worried about frequent battery charging.
While Xiaomi has not promoted its artificial intelligence as much as other consumer brands, Thursday’s launch event showcased several AI car features, such as allowing drivers to change a song using hand motions, or ask a phone app to describe where the car is parked.
The YU7 also supports Apple Car Play and Apple Music, Lei said.
The Chinese smartphone and home appliance company launched several other products on Thursday, including highly-anticipated artificial intelligence-connected glasses.
The AI glasses, which rival Meta’s Ray Bans smart offering, can change the tint of the lenses and scan a QR code to make payments, mimicking China’s mobile smartphone apps. Xiaomi also announced similar features to those of the Meta glasses, such as being able to take photos and videos, as well as use interactive AI to identify a flower or translate text.
Xiaomi’s AI glasses start at 1,999 yuan ($279). A Xiaomi spokesperson said there were currently no plans to sell the glasses overseas. Meta’s version isn’t officially sold in China.
June 26, 2025
GE Appliances is shifting washing machine production from China to Kentucky, and the reasons might surprise you
Some of President Donald Trump’s steepest tariffs are on products like washing machines, and on Thursday, GE Appliances said it would spend a half a billion dollars to make even more of them in the United States.
Tariffs, however, weren’t the driving factor behind the decision, the company’s CEO says, but they did serve as an accelerant.
GE Appliances announced it would spend $490 million to move some washing machine production from China and build a high-tech clothes care operation at its massive industrial park and headquarters in Louisville, Kentucky, where it already churns out washers and dryers for the US market.
The move of more than a dozen front-load washer models comes as US trade policy uncertainty has reached a high-stakes fever pitch as Trump’s July 9 tariffs deadline approaches.
GE Appliances move that is expected to be complete in 2027 and add 800 jobs, has been in the works for six years — shortly after a new line of front-load washers launched in 2019 — and follows a several-year stretch of high-dollar investments made to bolster the company’s US manufacturing footprint, CEO Kevin Nolan told CNN.
“We’ve had a strategy that making appliances in America makes sense; it’s an economic thing, and it’s also how we can serve our customers in a better, more efficient way,” Nolan said, adding that GE appliances is “not a company that was saying, ‘Hey, we’re going to outsource everything, and oh my God, now we’ve got to bring it back.’”
“[The trade policy] makes the payback for these things much, much greater,” Nolan added. “And with that, of course it’s going to accelerate (plans), because the quicker we can do these things, the quicker we can realize those benefits.”
The emergence and threat of tariffs also are influencing future decisions, Nolan said, noting that plans to reshore other components and parts are moving up the pecking order.
Earlier this week, an expansion of Trump’s 50% tariffs on steel extended to “derivative products,” including consumer appliances such as dryers, washing machines, refrigerators, ovens and garbage disposals. The US currently has a minimum 30% tariff on Chinese exports; however, it has soared as high as 145% in recent months.
“The current trade policy is the most dynamic thing anyone’s ever seen in their business career; I mean, it can change in a day, it can change in a week, it can change in a month,” Nolan said. “These investments are strategic, and you’ve got to look at the long term and what makes sense. You can’t do these just for trade policies.”
The move follows similar reshoring efforts made by GE Appliances in recent years, Nolan said. The company, which has been a subsidiary of China-based Haier Group since 2016, has invested $3.5 billion in its US manufacturing facilities during the past decade, he said.
Still, in recent months, the dramatic shifts in US trade policy and the Trump administration’s tumultuous tariff rates have loomed large over manufacturers like GE Appliances and its parent company’s appliance-making affiliate Haier Smart Home.
Earlier this month, Haier Smart Home executives told investors that the company’s localized supply chain in North America could help reduce its tariff exposure, according to translated company filings from June 4. Haier also flagged opportunities in “tariff-driven competitor weakness,” according to the filing.
Tariffs accelerated decision
In 2019, GE Appliances wanted to move quickly to market after developing what it believed was an innovation in the clothes care space: Making a less stinky front-load washer.
“To get this washer out fast, we said we’re going to tackle building the dryer plant (in Louisville) to make those and we’ll share the load and have the washer made in China,” he said. “We always had a forward-looking view that this thing was going to come back to America, but to get the things in the market quick, we did it that way in China first.”
The high-tariff environment “definitely made the numbers on this look very good; so, we said, ‘OK, let’s pull this thing in; let’s get this thing done now,’ because it just makes sense. The engineering work’s been going on; there’s a reason we can move fast on this,” he said. “But these tariffs could go away tomorrow, and once we make these decisions, we don’t back off. So that’s where you’ve got to make sure it’s the right thing to do.”
The half-a-billion-dollar investment announced Thursday will bring over more than 15 models of front-load washing machines, including a washer-dryer combo, to the 750-acre, multi-plant Appliance Park, where the company already manufactures top-load washers and clothes dryers.
The latest addition — which is expected to heavily feature automation, including robotics, automated guided vehicles and autonomous mobile robots — is expected to bring the company’s clothes care production at the site to 33 football fields in size.
Still, when it’s complete in 2027, the new production lines are expected to result in the addition of 800 full-time jobs to the 8,000-person campus.
“It’ll definitely be our flagship plant from a technology standpoint, so a lot of opportunities for upskilling employees,” Nolan said, referring to efforts for employees to learn new skills. “In order to be able to successfully manufacture in the United States, we have to be efficient.”
In recent years GE Appliances has touted a “zero distance” strategy to be closer to customers from a physical and design standpoint. To that end, the company’s 11 US plants include maker spaces and microfactories aimed at innovation and small-batch production.
The approach is a far cry from that taken in the 1980s and 1990s, when then-General Electric CEO Jack Welch leaned heavily into outsourcing and offshoring.
GE Appliances declined to share specifics as to what percentage of its manufacturing is now domestic versus overseas; however, the intent is to continue the reshoring efforts and expanding US manufacturing operations, Nolan said.
Why reviving US manufacturing isn’t easy
Trump, like presidents Obama and Biden before (and after) him, has long stated a desire to revive the US manufacturing industry and sought to wield tariffs to make that happen. However, economists and supply chain efforts have questioned the effectiveness in broad-based tariffs to that approach.
Several companies in recent months have announced plans to make investments in US manufacturing in recent months — with the White House taking credit — however, not only were these decisions already longer term in nature, any kind of large-scale rebound in domestic manufacturing will take time, said Jason Miller, a professor of supply chain management at Michigan State University.
“Right now, given all the profound uncertainty about tariffs, folks are not going to take action until some clarity emerges,” he said.
Companies that are able to announce or make moves now, he added, likely have existing capacity currently in place. To build a factory from scratch not only would take many months, if not many years, and then companies would run up against another challenge: finding enough skilled workers, he said.
About 22% of US plants have cited a lack of labor or labor skills as a key reason for their facilities running below full capacity, Miller said, citing his analysis of recent Census Bureau data.
GE Appliances has a waitlist of folks for jobs at its plants, but there’s still a huge need for a stronger pipeline of skilled workers, Nolan said.
“That’s just a national shortage,” he said. “When you look at us versus other countries, how many engineers are graduated, we’re way underrepresented. And then when you look at out of those engineers who are skilled in the art of manufacturing, it’s even worse. That’s the thing as a nation we’ve got to really grapple with.”
June 26, 2025
Meta won its AI copyright case, but the judge indicated that other lawsuits on the matter are still possible
Meta on Wednesday prevailed against a group of 13 authors in a major copyright case involving the company’s Llama artificial intelligence model, but the judge made clear his ruling was limited to this case.
U.S. District Judge Vince Chhabria sided with Meta’s argument that the company’s use of books to train its large language models, or LLMs, is protected under the fair use doctrine of U.S. copyright law.
Lawyers representing the plaintiffs, including Sarah Silverman and Ta-Nehisi Coates, alleged that Meta violated the nation’s copyright law because the company did not seek permission from the authors to use their books for the company’s AI model, among other claims.
Notably, Chhabria said that it “is generally illegal to copy protected works without permission,” but in this case, the plaintiffs failed to present a compelling argument that Meta’s use of books to train Llama caused “market harm.” Chhabria wrote that the plaintiffs had put forward two flawed arguments for their case.
“On this record Meta has defeated the plaintiffs’ half-hearted argument that its copying causes or threatens significant market harm,” Chhabria said. “That conclusion may be in significant tension with reality.”
Meta’s practice of “copying the work for a transformative purpose” is protected by the fair use doctrine, the judge wrote.
“We appreciate today’s decision from the Court,” a Meta spokesperson said in a statement. “Open-source AI models are powering transformative innovations, productivity and creativity for individuals and companies, and fair use of copyright material is a vital legal framework for building this transformative technology.”
Though there could be valid arguments that Meta’s data training practice negatively impacts the book market, the plaintiffs did not adequately make their case, the judge wrote.
Attorneys representing the plaintiffs said in a statement said that they “respectfully disagree” with the decision.
“The court ruled that AI companies that ‘feed copyright-protected works into their models without getting permission from the copyright holders or paying for them’ are generally violating the law,” the statement said. “Yet, despite the undisputed record of Meta’s historically unprecedented pirating of copyrighted works, the court ruled in Meta’s favor.”
Still, Chhabria noted several flaws in Meta’s defense, including the notion that the “public interest” would be “badly disserved” if the company and other businesses were prohibited “from using copyrighted text as training data without paying to do so.”
“Meta seems to imply that such a ruling would stop the development of LLMs and other generative AI technologies in its tracks,” Chhabria wrote. “This is nonsense.”
The judge left the door open for other authors to bring similar AI-related copyright lawsuits against Meta, saying that “in the grand scheme of things, the consequences of this ruling are limited.”
“This is not a class action, so the ruling only affects the rights of these thirteen authors — not the countless others whose works Meta used to train its models,” he wrote. “And, as should now be clear, this ruling does not stand for the proposition that Meta’s use of copyrighted materials to train its language models is lawful.”
Additionally, Chhabria noted that there is still a pending, separate claim made by the plaintiffs alleging that Meta “may have illegally distributed their works (via torrenting).”

Earlier this week, a federal judge ruled that Anthropic’s use of books to train its AI model Claude was also “transformative,” thus satisfying the fair use doctrine. Still, that judge said that Anthropic must face a trial over allegations that it downloaded millions of pirated books to train its AI systems.”
“That Anthropic later bought a copy of a book it earlier stole off the internet will not absolve it of liability for the theft, but it may affect the extent of statutory damages,” the judge wrote.
June 26, 2025
Salesforce CEO Marc Benioff states that AI is handling up to half of the company’s workload
Salesforce is accelerating its use of artificial intelligence in automating workloads, according to CEO Marc Benioff.
“All of us have to get our head around this idea that AI could do things, that before, we were doing, and we can move on to do higher-value work,” he said in an interview with Bloomberg, noting that the technology currently accounts for about 30% to 50% of the company’s work.
Technology companies are hunting for new ways to trim costs, boost efficiencies and transform their workforce with the help of AI.
The aftershocks have already hit the tech industry, with the software giant cutting more than 1,000 positions earlier this year as it restructured around AI.
Other technology companies have made similar moves, including cybersecurity giant CrowdStrike.
Klarna CEO Sebastian Siemiatkowski said the company has shrunk its headcount by 40% due in part to AI investment, while Amazon CEO Andy Jassy said the e-commerce giant will use artificial intelligence to reduce roles.
Benioff called the rise of AI in the workforce a “digital labor revolution,” estimating that the software company has reached about 93% accuracy with the technology.
“It’s pretty good,” he said, but it’s not “realistic” to hit 100%. He added that other vendors are at “much lower levels because they don’t have as much data and metadata” to build higher accuracy.
June 26, 2025
Trump Mobile continues to assert its phones are ‘Made in the USA,’ despite having removed that claim from its website
Trump Mobile, the wireless service provider and phone company launched by the Trump Organization, no longer promises on its website that its upcoming smartphone will be made in America.
The company adjusted language on its website on or around June 22 to drop the “Made in USA” claim, according to captures of the site by the Internet Archive. As of June 25, the company says the T1 8002 phone was “designed with American values in mind.” The website previously said the phone was “Made in the USA,” according to screenshots taken by NY Budgets earlier in June and archived versions of the site from June 18. The Verge first reported the change.
The revised language comes after industry analysts expressed skepticism about the phone’s American origins, noting that its specifications resembled a phone made by a Chinese manufacturer.
Despite changed language on the site, a spokesperson for Trump Mobile told Fox News that “the T1 phones are proudly being made in America.”
“Speculation to the contrary is simply inaccurate,” the statement said.
The Trump Organization’s press release from last week announcing Trump Mobile still says the $499 gold-colored phone will be “proudly designed and built in the United States.”
In the formal announcement from Trump Tower on June 16, Trump Mobile partner Pat O’Brien said, ”we are going to be doing phones that we are going to build in America.” But later, in a clip from an interview with conservative media personality Benny Johnson, Eric Trump said “eventually all the phones can be built in the United States of America.”
Ryan Reith, group vice president for the International Data Corporation’s Worldwide Device Tracker, previously told CNN that terms like “designed” and “built” are very vague. That makes it unclear precisely what parts of the phone making process would have taken place in the US. Apple, for example, designs its phones in California, but assembles them in areas like China and India with components from international suppliers.
Trump Mobile’s website says the phone is “brought to life right here in the USA.”
“There (are) no phones that are really being built in the US from start to finish,” Reith said last week when Trump Mobile was announced.
Some of the T1 8002 phone’s specifications also have changed, according to the Trump Mobile website. While the phone was originally listed as having a 6.78-inch screen, the website now says it has a 6.25-inch screen. That’s a noticeable change in size similar to the difference between an iPhone 16 and an iPhone 16 Pro Max. It’s rare for a tech company to make such a drastic change after announcing a phone.
Trump Mobile also no longer lists the phone’s memory, the part of the phone that stores app data and impacts performance when switching between apps.
Todd Weaver, CEO of Purism, one of the only known companies to actually manufacture a cell phone in the United States, and Max Weinbach, an analyst at market research firm Creative Strategies, independently told NY Budgets previously that they believe the originally announced T1 phone looks like a version of the already available Revvl 7 Pro 5G. That phone is made by China-based Wingtech, which provides manufacturing services for smartphones and other products, and retails for around $169 on Amazon.
The debut of Trump Mobile came as President Trump, who is not involved in the daily operations of the Trump Organization run by his sons, has been pressuring tech giants like Apple and Samsung to make their smartphones in the United States. The move is part of a push to bring manufacturing jobs back to America, although experts have said making phones domestically at scale is a challenging, if not impossible, task – particularly under the September timeframe originally promised.
“Unless the Trump family secretly built out a secure, onshore or nearshore (fabrication) operation over years of work without anyone noticing, it’s simply not possible to deliver what they’re promising,” Weaver previously said.
June 26, 2025
AI Warnings Have Become the Trendy New Tactic for CEOs to Keep Employees Worried About Job Security
Every few weeks, the Earth cries out for an artificial intelligence scare on a frequency heard only by tech CEOs. And lo, like a rain cloud over a parched valley, here comes Amazon boss Andy Jassy to shower us with fresh fear and dread.
In a memo sent to employees titled “Some thoughts on Generative AI,” Jassy spent 1,200 words largely rattling off examples of Amazon’s AI progress. It is making Alexa, its personal assistant software, “meaningfully smarter,” and turning its customer service chatbot into “an even better experience.” (How, and by what measure? He didn’t say. But “you get the idea,” he wrote.)
Then in a textbook example of burying the lede, he got to the point around paragraph 15: We’re almost certainly going to replace some Amazon workers with AI “agents.”
When? “In the next few years.”
How many jobs are we talking about? “It’s hard to know… we expect that this will reduce our total corporate workforce as we get efficiency gains from using AI extensively across the company.”
Where are all these so-called agents? “Many of these agents have yet to be built, but make no mistake, they’re coming, and coming fast.”
Fast! Soon! We expect! They’re coming!
To be clear: I’m not saying Jassy is lying. But he is clearly invoking AI to put a modern spin on a strategy as old as time: Keep workers working by making them afraid of losing their jobs.
The sentiment echoes a similar but more dramatic statement by Anthropic CEO Dario Amodei, who told CNN and Axios that AI could wipe out half of all entry-level white-collar jobs sometime in the next five years. (Why half? And why five years? Eh, why not… Amodei’s motive is to make his core technology appear both inevitable and scary-powerful.)
Not all tech CEOs agree, to be clear. Nvidia’s Jensen Huang and Google Deepmind’s Demis Hassabis — both major players in the AI space — have pushed back on Amodei’s apocalyptic take.
It’s important to keep a couple of things in mind when we get these semiannual bursts of AI fearmongering from the very people who stand to profit from advancing the technology.
One: Automation and machine learning have been around for decades, and yes, that has had (and continues to have) an impact on the labor market. But the idea that generative AI, in particular, is going to usher in some kind of doom-slash-utopia belongs in the realm of science fiction.
Large language models that power advanced AI chatbots can be impressive sidekicks and sounding boards, to be sure. They are also hallucinating more — not less — the larger they become. And they have just about run out of the kind of human-grade data engineers need to train the models.
Two: Notice that Jassy’s note to staff didn’t say AI was coming for his job, or his fellow executives’ jobs. Kinda seems like he might want to review what current AI is good at — producing OK-sounding memos, synthesizing information and (maybe) solving strategic puzzles. And then consider what AI is still really bad at — physically lifting things and moving them around.
Three: It’s curious to see Big Tech recycling the same language about “flexibility” and “efficiency” that came with literally every other workplace tech innovation of the last 30+ years. Email, Slack, Teams, Zoom, Plorfen, Globz. (OK I made the last two up.)
To be clear, those things aren’t inherently bad. They did give us flexibility that proved vital during the 2020 lockdowns. But they also gave us the flexibility to be online in perpetuity, all day and night, seven days a week.
Incidentally, Microsoft, a company that’s earmarked $80 billion in AI spending for the year, just released a report about how those innovations have — rather than liberate office workers from drudgery — actually trapped us in an “infinite workday.”
The report found the typical office worker using Microsoft’s Outlook, Teams, PowerPoint and other products increasingly spend their days getting interrupted every two minutes by a meeting, an email or a chat notification during a standard eight-hour shift. That’s 275 pings a day, my colleague Anna Cooban notes.
The average employee receives 117 emails a day, and sent or received 58 instant messages outside of their core working hours — a jump of 15% from last year.
Part of Microsoft’s solution for this “broken system,” it should be noted, includes re-orienting jobs around — wait for it — AI agents.
June 20, 2025
Pope Leo Urges Tech Leaders to Adopt an Ethical Framework for AI in Vatican Address
Pope Leo XIV says tech companies developing artificial intelligence should abide by an “ethical criterion” that respects human dignity.
AI must take “into account the well-being of the human person not only materially, but also intellectually and spiritually,” the pope said in a message sent Friday to a gathering on AI attended by Vatican officials and Silicon Valley executives.
“No generation has ever had such quick access to the amount of information now available through AI,” he said. But “access to data — however extensive — must not be confused with intelligence.”
He also expressed concern about AI’s impact on children’s “intellectual and neurological development,” writing that “society’s well-being depends upon their being given the ability to develop their God-given gifts and capabilities.”
That statement from the Pope came on the second of a two-day meeting for tech leaders in Rome to discuss the societal and ethical implications of artificial intelligence. The second annual Rome Conference on AI was attended by representatives from AI leaders including Google, OpenAI, Anthropic, IBM, Meta and Palantir along with academics from Harvard and Stanford and representatives of the Holy See.
The event comes at a somewhat fraught moment for AI, with the rapidly advancing technology promising to improve worker productivity, accelerate research and eradicate disease, but also threatening to take human jobs, produce misinformation, worsen the climate crisis and create even more powerful weapons and surveillance capabilities. Some tech leaders have pushed back against regulations intended to ensure that AI is used responsibly, which they say could hinder innovation and global competition.
“In some cases, AI has been used in positive and indeed noble ways to promote greater equality, but there is likewise the possibility of its misuse for selfish gain at the expense of others, or worse, to foment conflict and aggression,” Leo said in his Friday statement.
Although it doesn’t have any direct regulatory power, the Vatican has been increasingly vocal on AI policy, seeking to use its influence to push for ethical technological developments.
In 2020, the Vatican hosted an event where tech leaders, EU regulators and the late Pope Francis discussed “human-centric” AI, which resulted in the Rome Call for AI Ethics, a document outlining ethical considerations for the development of AI algorithms. IBM, Microsoft and Qualcomm were among the signatories who agreed to abide by the document’s principles.
Two years later, Francis called for an international treaty to regulate the use of AI and prevent a “technological dictatorship” from emerging. In that statement — which came months after an AI-generated image of Francis in a puffy coat went viral — he raised concerns about AI weapons and surveillance systems, as well as election interference and growing inequality. In 2024, he became the first pope to participate in the G7 summit, laying out the ethical framework for the development of AI that he hoped to get big tech companies and governments on board with.
Following Francis
When Pope Leo XIV became leader of the Catholic Church last month, he signaled that his papacy would follow in Francis’ footsteps on topics of church reform and engaging with AI as a top challenge for working people and “human dignity.”
The new pontiff chose to name himself after Pope Leo XIII who led the church during the industrial revolution and issued a landmark teaching document which supported workers’ rights to a fair wage and to form trade unions. With the development of AI posing a similar revolution to the one during the 19th century, Leo has suggested that the church’s social teaching — which offers a framework on engaging with politics and business — be used when it comes to new tech advancements.
“In our own day, the church offers to everyone the treasury of her social teaching in response to another industrial revolution and to developments in the field of artificial intelligence that pose new challenges for the defense of human dignity, justice and labor,” Leo said in that May address.
The Friday event, which took place inside the Vatican’s apostolic palace, included a roundtable discussion on AI ethics and governance. Among those present from the Vatican side were Archbishop Vincenzo Paglia, who has engaged with business leaders on AI, and Archbishop Edgar Peña Parra, who holds the position of “sostituto” (substitute) in the Vatican, a papal chief of staff equivalent.
Earlier this week, Leo referenced AI during a speech to Italian bishops, talking about “challenges” that “call into question” the respect for human dignity.
“Artificial intelligence, biotechnologies, data economy and social media are profoundly transforming our perception and our experience of life,” he told them. “In this scenario, human dignity risks becoming diminished or forgotten, substituted by functions, automatism, simulations. But the person is not a system of algorithms: he or she is a creature, relationship, mystery.”
A key issue at Friday’s event is AI governance, or how the companies building it should manage their need to generate profit and responsibilities to shareholders with the imperative not to create harm in the world. That conversation is especially pressing at a moment when the United States is on the brink of kneecapping the enforcement of much of the limited regulations on AI that exist, with a provision in President Donald Trump’s proposed agenda bill that would prohibit the enforcement of state laws on AI for 10 years.
In his statement, Leo called on tech leaders to acknowledge and respect “what is uniquely characteristic of the human person” as they seek to develop an ethical framework for AI development.
June 20, 2025
Cybercriminals Hack Aflac Amid Broader Attack on U.S. Insurance Industry
Cybercriminals have breached insurance giant Aflac, potentially stealing Social Security numbers, insurance claims and health information, the company said Friday, the latest in a spree of hacks against the insurance industry.
With billions of dollars in annual revenue and tens of millions of customers, Aflac is the biggest victim yet in the ongoing digital assault on US insurance companies that has the industry on edge and the FBI and private cyber experts scrambling to contain the fallout.
Erie Insurance and Philadelphia Insurance Companies have also reported hacks this month, which in those cases have caused widespread disruptions to IT systems used to serve customers. All three insurance-company hacks are consistent with the techniques of a young and rampant cybercrime group known as Scattered Spider, people familiar the investigation tell CNN.
“This attack, like many insurance companies are currently experiencing, was caused by a sophisticated cybercrime group,” Aflac said in a statement on Friday, without naming Scattered Spider. Aflac said it “stopped the intrusion within hours” after discovering it last week, that no ransomware was deployed, and that it continues to serve its customers.
It was too early to tell, the company said, how much customer information may have been stolen, but the potential exposure is vast. Aflac is one of the largest providers of supplemental health insurance in the US for medical expenses that aren’t covered by a primary provider.
The hackers used “social engineering” to worm their way into its network, according to Aflac. That tactic can involve duping someone into revealing security information to help gain access to a network. It’s a hallmark of Scattered Spider attackers, who are known to pose as tech support to infiltrate big corporations.
The loose group of cybercriminals is considered dangerous and unpredictable, in part because it is believed to be comprised of youths in the US and the UK known for aggressively extorting their victims. Scattered Spider shot to infamy in September 2023 when they were linked to a pair of multimillion-dollar hacks on famous Las Vegas casinos and hotels MGM Resorts and Caesars Entertainment.
The hackers’ tactics, and the way they target big swaths of American industries at a time, has cybersecurity executives pleading with companies to be wary of suspicious phone calls to their employees. Just last month, they were suspects in multiple cyberattacks on American retail companies.
“If Scattered Spider is targeting your industry, get help immediately,” said Cynthia Kaiser, who until last month was deputy assistant director of the FBI’s Cyber Division and oversaw FBI teams investigating the hackers. “They can execute their full attacks in hours. Most other ransomware groups take days.”
Scattered Spider often registers web domains that look very much like trusted help desks that companies use for IT support, the cybersecurity firm Halcyon, where Kaiser now works, says in a forthcoming report.
While concerns about Iranian cyber capabilities are in the news because of the Israel-Iran war, “the threat I lose sleep over is Scattered Spider,” said John Hultquist, chief analyst at Google’s Threat Intelligence Group. “They are already taking food off shelves and freezing businesses. The Iranian hackers may not even have Internet access, but these kids are in play right now.”
June 20, 2025