OpenAI and Anthropic are competing for partnerships with buyout firms that would allow them to quickly roll out their AI tools to ​potentially hundreds of private, established companies owned by buyout firms. This would boost adoption of their models and encourage customer stickiness at scale.

OpenAI is ‌offering private-equity firms a guaranteed minimum return of 17.5%, significantly higher than typical preferred instruments, two people familiar said. It is also offering early access to its newest AI models as it seeks to enlist investors like TPG and Advent for its joint venture, three sources said.

The rise of artificial intelligence has become a force propelling the economy and the stock market. But it is also fueling the U.S. trade deficit, as tech companies import expensive foreign computers and chips to fill their new data centers.

Unlike cars, steel and other goods, electronics were intentionally spared by Mr. Trump. Last April, the administration issued an exemption from tariffs for smartphones, computers, semiconductors and other electronics. It was a significant break for tech companies, like Apple, Nvidia and Dell, that have lobbied the president against broad tariffs.

The tariff exemption has increased demand for imports of computers, semiconductors and other equipment. So has rapid data center construction around the United States.

The A.I. boom has helped to prop up an otherwise lackluster U.S. economy. It is also powering growth in the stock market, which Mr. Trump has long seen as a metric of his administration’s success. Over the past three years, America’s largest tech stocks — a group known as the Magnificent Seven — have been responsible for more than half of the 88 percent gain in the S&P 500.

We did the math. At $185 billion a year, in eight years, Google would be spending $1.5 trillion, slightly more than OpenAI has committed to spend over the same time period. Extend that out to 10 years, as Vahdat noted, and Google would be spending $1.9 trillion.

Vahdat is clear that this is “not a promise” that Google would spend that much over the next 10 years. But the decade-long view he takes suggests the scope of Google’s bet. “The point here is that we are, at Google, investing at the highest levels,” he says.

There’s a big difference between Google’s data center ambitions and OpenAI’s: Google is a money-making machine. In the fourth quarter, Google parent Alphabet raked in $113 billion in revenue; for the full year, sales topped $400 billion for the first time in the company’s more than 25 year history. By comparison, OpenAI is spending at similar levels and only brought in about $13 billion in revenue last year — a tiny fraction of Google’s revenue, and less than half of Google’s cash reserves.

Google’s TPUs previously were only used in house for Google’s own infrastructure — to power consumer apps like Gmail and YouTube, and eventually train self-driving cars and develop and run AI models like Gemini. Now, they’re one of the industry’s go-tos: maybe not as popular as Nvidia’s top of the line Blackwells, but still useful for pretraining and operating AI models at scale. Google first started selling access to them through a cloud service in 2018, letting other companies rent out processing power. But more recently, Google has inked high profile deals, like a big contract with Anthropic, and has reportedly been in talks with Meta to use its chips. In December, Morgan Stanley estimated that TPUs could generate $13 billion for Google by 2027. “It is fair to say that the demand for cloud TPUs has been unprecedented,” Vahdat says, particularly in the last few years.

In August, Vahdat, Google Chief Scientist Jeff Dean, and 10 other researchers and execs at the company, co-published a paper aiming to contextualize AI’s power guzzling. The paper says that the median prompt for Google’s Gemini AI model uses the same amount of energy it takes to power 9 seconds of television and consumes around five drops of water, which they write is “substantially lower than many public estimates.” (One report says large data centers can consume up to 5 million gallons per day, equivalent to the water use of a town populated by up to 50,000 people.)

Coding After Coders: Summary

The New Reality of AI-Assisted Programming

• Elite software developers now rarely write code themselves — instead, they direct AI agents in plain English
• Tools like Claude Code deploy multiple agents simultaneously: one writes, one tests, one supervises
• Tasks that once took days now take under an hour

The Strange New Workflow

• Developers spend their days describing intent to AI, reviewing the AI's "plan," then letting agents execute
• When agents misbehave, developers have resorted to scolding, pleading, ALL-CAPS commands, and emotionally charged language ("embarrassing," "national security imperative") — and it seems to work
• Prompt files have become records of hard-won rules to constrain unpredictable AI behavior

Economic Stakes

• Coding was once considered near-guaranteed, high-paying employment ($200K+)
• It may be the first expensive white-collar skill AI can fully replace — unlike AI video or legal briefs, AI-generated code that passes tests is indistinguishable in value from human-written code
• Irony noted: Silicon Valley workers, who told others to "learn to code," got automated first

Developer Sentiment: Mostly Euphoric

• Most developers interviewed were energized, not demoralized — reporting 10x to 100x productivity gains
• Key insight from tech executive Anil Dash: unlike creative fields where AI removes the soulful work and leaves drudgery, in coding AI removes the drudgery and leaves the soulful parts

Historical Context: A Long Arc of Abstraction

• Each programming era simplified the one before: Assembly → high-level languages (Python) → open-source packages → now natural language intent
• AI represents the highest abstraction layer yet: developers no longer need to manage syntax, memory, or debugging minutiae
• The open question, now being asked at Anthropic itself: what is coding, fundamentally, when the code-writing is gone?

Total effective tax rates (includes all taxes: federal+state income tax, sales tax, property tax, etc) for the richest Americans declined by 2018 to a level beneath that of the bottom 50% of earners,[18] contributing to wealth inequality. Analysis by economists Emmanuel Saez and Gabriel Zucman.

The Federal Insurance Contributions Act (FICA /ˈfaɪkə/) is a United States federal payroll (or employment) tax payable by both employees and employers to fund Social Security and Medicare[1]—federal programs that provide benefits for retirees, people with disabilities, and children of deceased workers.

Since 1990, the employee's share of the Social Security portion of the FICA tax has been 6.2% of gross compensation up to a limit that adjusts with inflation.[a][9] The taxation limit in 2020 was $137,700 of gross compensation, resulting in a maximum Social Security tax for 2020 of $8,537.40.[7] This limit, known as the Social Security Wage Base, goes up each year based on average national wages and, in general, at a faster rate than the Consumer Price Index (CPI-U). The employee's share of the Medicare portion of the tax is 1.45% of wages, with no limit on the amount of wages subject to the Medicare portion of the tax.

So personal income tax in the US is ~30% for most of us (ranging from ~10%-37%), compared to Social Security's ~6.2% Medicare is 1.45% (or 12.4% + 2.9% if you count the employer portion). AND only the first ~$137K is taxable so our maximum tax amount to Social Security and Medicare is capped, while normal income tax that politicians can direct to anything from foreign wars to immigration enforcement to redistribution to different states or interest on debt driven by tax breaks to the rich that caused deficits.

An average of 9,000 refugees were admitted monthly between January 2024 to January 2025. From February to December 2025, there were 1,226 total admissions, 1,059 of whom were from South Africa.

It's quite disappointing that these policies - especially the H1B tax, which brings the best and brightest in the world to the US - all target legal immigrants.

I love this report!

This data-driven, impartial report contains historic metrics — how you use them to advocate for the changes you want to see in the country is up to you.

Most spending was on Social Security, national defense, grants to state and local governments, Medicare, and interest on the debt. Spending and revenue were both higher than their pre-pandemic levels, and the federal government ran another deficit as spending outpaced revenue.

Why do we always lump Social Security in with other national spending? Social Security is collected separately from all other tax revenue and goes directly to the Social Security trust fund. That money cannot be put anywhere else. Politicians can't direct Social Security goes into a trust fund and politicians can't change how it's spent, unlike defense spending and other spending. In my view, Social Security should be separate. It's not the government's money to spend, it's money that is given back to the people directly. So comparing national defense, which the government can choose to change the spending levels, reallocate it to other spending priorities, Social Security cannot be because it's a trust fund.

Public schools took in and spent more funds than ever before. It also had mixed impacts on teachers and students. The number of public-school teachers has increased each year since 2020 while the number of students has decreased or stayed the same. Meanwhile, test scores have fallen.

Well we have to do something about that and be drastic about it. However, I don't see how cutting funding alone - the current Republican priority - will help.

Prominent economists, including from Morgan Stanley and JPMorgan Chase, calculate that the AI buildup was directly responsible not for 92 percent or 39 percent of gains to the U.S. economy in 2025, but as little as zero.

It’s clear that the huge spending on AI is adding to the U.S. economy, but the available economic data doesn’t neatly capture its effects. The debating economists and the slippery data suggest that if the technology does start to reshape the economy, it may be challenging to detect and clearly measure. That may leave political and corporate leaders to choose the numbers that fit their preferred narratives on how AI is changing American life and work.

That’s because the $31 trillion in yearly U.S. gross domestic product, the widest measure of the economy, tallies only the final value of products and services produced domestically. Spending on imports and foreign made components is subtracted because it boosts the economies of other countries, not that of the United States.

Roughly three-quarters of the cost of an AI data center is for the computer gear and parts such as computer chips that go inside of it, technology analysts estimate. America’s AI champions, including the computer chip pioneer Nvidia, manufacture many of their products in Asia — despite efforts by the Biden and Trump administrations to reduce U.S. dependence on essential chips made overseas.

And some forecasters say that the U.S. government’s economic data is a poor measure of the impact of AI and that alternative calculations show the current boom is an even bigger boost to economic growth.

“This is a big deal, but not the be-all and end-all,” said Joseph Politano, an economic analyst who writes the Apricitas Economics newsletter. He calculates that AI-related spending contributed about 0.2 percentage points to the 2.2 percent U.S. economic growth last year.

The AI buildup is putting real money into the pockets of some Americans and U.S. businesses. Stock market gains from AI enthusiasm are plumping up Americans’ investment portfolios.

“The two engines of today’s economy are the AI ecosystem and wealthy consumers,” Richmond Fed President Tom Barkin said in a January speech.

Prominent economists, including from Morgan Stanley and JPMorgan Chase, calculate that the AI buildup was directly responsible not for 92 percent or 39 percent of gains to the U.S. economy in 2025, but as little as zero.

Prominent economists, including from Morgan Stanley and JPMorgan Chase, calculate that the AI buildup was directly responsible not for 92 percent or 39 percent of gains to the U.S. economy in 2025, but as little as zero.

Prominent economists, including from Morgan Stanley and JPMorgan Chase, calculate that the AI buildup was directly responsible not for 92 percent or 39 percent of gains to the U.S. economy in 2025, but as little as zero.

Prominent economists, including from Morgan Stanley and JPMorgan Chase, calculate that the AI buildup was directly responsible not for 92 percent or 39 percent of gains to the U.S. economy in 2025, but as little as zero.

Prominent economists, including from Morgan Stanley and JPMorgan Chase, calculate that the AI buildup was directly responsible not for 92 percent or 39 percent of gains to the U.S. economy in 2025, but as little as zero.

Prominent economists, including from Morgan Stanley and JPMorgan Chase, calculate that the AI buildup was directly responsible not for 92 percent or 39 percent of gains to the U.S. economy in 2025, but as little as zero.

Prominent economists, including from Morgan Stanley and JPMorgan Chase, calculate that the AI buildup was directly responsible not for 92 percent or 39 percent of gains to the U.S. economy in 2025, but as little as zero.

Prominent economists, including from Morgan Stanley and JPMorgan Chase, calculate that the AI buildup was directly responsible not for 92 percent or 39 percent of gains to the U.S. economy in 2025, but as little as zero.

Prominent economists, including from Morgan Stanley and JPMorgan Chase, calculate that the AI buildup was directly responsible not for 92 percent or 39 percent of gains to the U.S. economy in 2025, but as little as zero.

Prominent economists, including from Morgan Stanley and JPMorgan Chase, calculate that the AI buildup was directly responsible not for 92 percent or 39 percent of gains to the U.S. economy in 2025, but as little as zero.

Prominent economists, including from Morgan Stanley and JPMorgan Chase, calculate that the AI buildup was directly responsible not for 92 percent or 39 percent of gains to the U.S. economy in 2025, but as little as zero.