Hello, and welcome to TechScape. This week, tech companies are spending amounts of money that stretch the limits of the imagination. Donald Trump’s administration is spending more money with data analytics and surveillance firm Palantir. And women on both sides of the Pacific face the extreme difficulty of keeping intimate moments private online.
Update: the trillion-dollar AI race, part II
In last week’s edition of the newsletter, my colleagues wrote about the upshot of Google’s earnings call: lots of money earned, but, more importantly lots of money spent on AI. Even more money shelled out than previously expected: Google revised its predictions for how much it would invest in building up its AI capacity upwards by billions. Investors loved it. Shares up.
In the ensuing week, three more tech giants reported their quarterly earnings – Meta, Microsoft and Amazon – and disclosed that they have collectively spent $155bn. Investors expressed elation at the colossal sums. Meta’s market capitalization shot up by more than $130bn. Microsoft’s valuation soared past $4tn, making the software giant the second publicly traded company to reach that stratospheric milestone. Amazon’s financial outlook was murkier, and its shares went down. What a bummer to miss out on the AI stock party.
The $155bn sum represents more than the US government has spent on education, training, employment and social services in the 2025 fiscal year so far. One economics research firm claims AI spending has contributed more to the US economy over the past two quarters than consumer spending, traditionally the biggest factor in economic growth.
For the coming fiscal year, big tech’s total capital expenditure is slated to balloon enormously, surpassing the already eye-popping sums of the previous year. Microsoft plans to unload about $100bn on AI in the next fiscal year, CEO Satya Nadella said. Meta plans to spend between $66bn and $72bn. Alphabet plans to spend $85bn, significantly higher than its previous estimation of $75bn. Amazon estimated that its 2025 expenditure would come to $100bn as it plows money into Amazon Web Services, which analysts now expect to amount to $118bn. In total, the four tech companies will spend more than $400bn on capex in the coming year, according to the Wall Street Journal.
Read more about the gargantuan sums of money being spent on AI.
Where does Palantir have contracts?
Last week, the US army announced a new agreement with Palantir, the Peter Thiel-founded, Alex Karp-run technology company.
The agreement combined 75 separate, existing contracts between the army and Palantir into one, and allows for the possibility to purchase goods and services up to $10bn.
It’s just one of dozens of agreements between the company and the US, a relationship that’s only been growing rapidly in the second Trump administration, though it had been on the rise before then. Palantir brought in $373m in revenue from US government contracts in just the first quarter of 2025 – $151.6m more than a year prior. The vast majority of that increase – $148.7m – came from government customers who were already working with the company, according to its earnings reports.
Here’s an overview of Palantir’s biggest US government contracts:
-
Department: Defense
Worth: The Department of Defense lists its “obligations” to Palantir as $1.66bn in the government, which can encompass current and future spending, according to the US government’s database of its own spending. Financial analysts estimate Palantir earns $400m in annual recurring revenue from the DoD.
The details: The DoD remains Palantir’s biggest and oldest customer within the US federal government. The first contract between the two dates back to 2008.
The army has made no commitment and is under no obligation to purchase anywhere close to the $10bn figure listed as the value of its new agreement with Palantir, which represents the “maximum potential value of the contract”, according to the press release the government published. The number is not exactly money in hand for Palantir, but analysts seem encouraged it could represent a major source of revenue and more business from the US government.
“It’s no obligation but we believe the army will spend billions with Palantir with this contract,” said Dan Ives, managing director at the wealth management firm Wedbush Securities.
-
Department: Homeland Security
Worth: $256.7m in obligations
Details: The company has been working with the homeland security department since 2011. The vast majority of Palantir’s contracts with DHS are to provide services to Immigration and Customs Enforcement, or Ice. These amount to $248.3m in obligations. The company’s most recent contract with Ice was for $30m to make the deportation process more efficient.
-
Department: Health and Human Services
Worth: $385m in obligations
-
Department: Treasury
Worth: $140.9m
-
Department: Justice
Worth: $204.5m
-
Department: Energy
Worth: $91m
after newsletter promotion
-
Department: State
Worth: $56.1m
-
Department: Transportation
Worth: $55.92m
In the UK, Palantir has won a £330m contract with the National Health Service.
Women the world over see their intimate messages and images exposed
On both sides of the Pacific Ocean, women are being confronted with the difficulty of keeping their most personal moments private online.
In the US, an app that pledged to make IRL dating safer has suffered a breach that may imperil its users. In China, hundreds of thousands of men are sharing explicit pictures without the consent of the women in the photos.
The app in the US, Tea, offered a forum for women who subscribed to share past experiences with men so that other women could conduct DIY background checks on their prospective dates, highlighting negative “red flags” and positive “green flags”. Tea’s owners bill the app as “the safest place to spill”, in reference to the English slang term for gossip. It has topped US download charts in recent weeks, and the company has boasted about a user base of 1.6 million women. It is only available in the US.
The app promised “dating safety tools that protect women”. but in late July, the company discovered that hackers had breached its systems and leaked users’ driver’s licenses, direct messages and selfies. Users of the noxious message board 4Chan screenshotted and spread Tea users’ personal information, according to NPR. A second breach exposed more than a million messages sent by Tea’s users, including ones about sensitive topics like abortions or cheating, per 404 Media, which first reported both breaches. The company claimed in a statement that the first breach only affected users who had signed up before February 2024, but the second one was much more recent, 404 reported. In response to the hack, the app has suspended messaging entirely, the BBC reported.
Most data breaches inspire little public uproar. The exposure of an email address here, a birthday there can feel commonplace. The breach of Tea is different. The app promised safety as a core feature. It delivered the opposite. The sine qua non of a whisper network like Tea is privacy, the ability to share damning information in secret, which the app failed to protect. Exposing users’ identities and messages is the most basic type of failure, one that can be fatal to a product’s reputation. To make matters worse, the breach offers red meat to the male-dominated 4Chan forum, a node of incel culture and men’s rage in the US.
“Our team remains fully engaged in strengthening the Tea app’s security, and we look forward to sharing more about those enhancements soon,” the company said in a statement to the BBC. “In the meantime, we are working to identify any users whose personal information was involved and will be offering free identity protection services to those individuals.”
In China, women are facing down an online legion of men dedicated to invading their most private moments with spy cameras and sharing the results on the internet.
My colleague Amy Hawkins reports:
Anger is growing on Chinese social media after news reports revealed the existence of online groups, said to involve hundreds of thousands of Chinese men, which shared photographs of women, including sexually explicit ones, taken without their consent.
The Chinese newspaper Southern Metropolis Daily published a report last week about a group on the encrypted messaging app Telegram called “MaskPark tree hole forum”. It said it had more than 100,000 members and was “comprised entirely of Chinese men”.
Men reportedly shared sexually explicit images of women either in intimate settings or with so-called “pinhole cameras” that can be hidden in everyday items such as plug sockets and shoes.
In an influential 2014 essay, “Why women aren’t welcome on the internet”, the writer Amanda Hess said that receiving countless graphic death and rape threats in response to her work did not make her exceptional: “It just makes me a woman with an internet connection.” Events of the past week indicate that Hess’s headline still holds true.
Tim Berners-Lee, credited with inventing the world wide web, told the Guardian in 2020 that the internet “is not working for women and girls”. The same year, a Unesco report found that 73% of women journalists endured online threats to their safety. Other UN reports have found that significant portions of women across the world, somewhere between 16% and 58%, face threats of gender-based violence online.
