The AI industry's two-and-a-half-year investment spree is about to enter a new phase of turmoil and doubt thanks to President Trump's massive tariffs.

Why it matters: Key industry players backed Trump or made nice with him after his election win, but the centerpiece of his economic policy makes an AI slowdown — or even a crash — much more likely.

The big picture: The AI giants need cheap data and cheap energy to keep building bigger models, and mountains of cash to pay for it all.

• Recent months have seen a tidal wave of deals in the tens of billions of dollars — from OpenAI's latest investment round and Stargate project to Elon Musk's most recent xAI fundraise — based on forecasts of beyond-the-horizon growth in demand for AI services.

Yes, but: In the trade-war era kicked off by Trump's tariffs, capital will move more cautiously and growth will slow or vanish.

• The flood of billions into AI model training and infrastructure was always understood to be a big gamble on an impressive technology that hasn't yet proved its broad business value.
• Last year, some analysts began to question the industry's "incineration" of cash.

In the new post-tariff world, expect many more investors to watch and wait rather than toss more billions on the fire.

• Meanwhile, AI companies newly challenged to drum up revenue will have to do so in a market that's hurting, or even in recession.

The best case for AI companies is that their products will thrive as customers seeking savings in a downturn deploy AI and automation to try to cut costs.

• For that to work, AI will have to deliver on that promise much more widely than it has so far in a handful of realms like software development and customer service.

Between the lines: Some in the MAGA camp see AI as a key player in their project to restore the heyday of domestic U.S. manufacturing.

• Trump's Commerce Secretary, Howard Lutnick, described this vision on CBS' "Face the Nation" Sunday.
• "The army of millions and millions of human beings screwing in little, little screws to make iPhones — that kind of thing is going to come to America. It's going to be automated. And great Americans, the tradecraft of America is going to fix them, is going to work on them. There's gonna be mechanics, there's gonna be HVAC specialists, there's gonna be electricians."
• "The tradecraft of America, our high school educated Americans, the core to our workforce, is going to have the greatest resurgence of jobs in the history of America to work on these high-tech factories."

Lutnick isn't totally clear here. But it sounds like he's saying iPhone manufacturing will move to the U.S. but will be automated.

• Americans won't work in the factories like their Chinese predecessors tightening all those tiny screws; instead, Americans will work on the factories, building and maintaining them.
• Put aside what Apple experts say about the impossibility of building iPhones in the U.S. without tripling their price. Just consider what this scenario actually means for American workers.
• You can't have "the greatest resurgence of jobs in the history of America" if you're automating all the manufacturing.

The more likely scenario is a broad economic slowdown, with the AI industry, along with the rest of the U.S. economy, mired in the doldrums of uncertainty and fear.

• A tariff-induced inflation jolt, a trade-war-fueled recession, or both together would radically limit the runway for AI startups and dry up the dollars flowing into AI research.

The bottom line: The biggest companies — Microsoft, Google, Apple and Meta — all have enough cash flow and reserves to weather such a storm. Not so for the rest of the AI industry.

Elon Musk's self-deal for his AI company xAI to acquire X, formerly Twitter, is the strongest sign yet that the AI business is devouring the social media world.

Between the lines: Musk's move is a maneuver to spruce up his overall financial picture — leveraging the AI industry's fizzy valuation arithmetic to shore up X's stagnant revenue and debt-loaded capital structure.

• High expectations for xAI and every other AI darling are buoying the rest of the tech industry for now — but all that ends the moment the AI bubble bursts.

Why it matters: The entire social media world — including X, Meta-owned Facebook and Instagram, Google-owned YouTube, even the younger TikTok — has become a legacy platform.

• While this inherited business is still huge and in many cases profitable, it's not going to grow or innovate at the pace Silicon Valley and Wall Street demand.
• Right now, X may be the most financially challenged of the large social-media platforms. But Meta's and Google's properties are also benefiting from investors' bets on the parent companies' AI growth.

Catch up quick: 2 1/2 years after he bought Twitter, Musk has done little to grow its business or actually transform the company into the "everything app" he once planned.

• A month after he acquired the company, OpenAI released ChatGPT, and soon after it became clear that any new "everything app" was likely to be an AI tool, not a social media network.
• xAI gives Musk a convenient vehicle to scoot X out of the tarnished social media category and into the AI world.

Yes, but: X is a mature business with thousands of employees, hundreds of millions of customers and about $2 billion in estimated global ad revenue.

• Musk's drastic spending and staff cuts brought its overhead down, but it doesn't seem to have found a way to return to the revenue levels it last saw as the public company known as Twitter (a little over $5 billion) — let alone spark significant growth.

xAI, on the other hand, is a growing two-year-old startup that won headlines for getting a giant new data center in Memphis, Tennessee, built fast.

• The company's AI models have become contenders in the bigger AI race, earning comparisons with better-known rivals like OpenAI and Google.
• But for most of xAI's existence the bulk of its usage has come from the X network.

By the numbers: Musk says xAI's paper valuation based on its most recent private investment round is now $80 billion.

• Its exact revenue is a tiny fraction of that — maybe $100 million — meaning investors are paying 800 times revenue for their shares. (The equivalent figure for Tesla right now is about eight times revenue.)
• xAI's modest revenue stream in turn comes chiefly from X itself.
• For the moment, these dollars are all just circling the Musk-o-sphere in a slightly different configuration.

The big picture: Everyone in tech is benefiting from AI's magic multiplication powers right now.

• That definitely includes Meta and Google. Both giants have massive social media (and search) revenue streams, but both also rely on AI sparkle to tell a story about future growth — and mask stagnation in their core businesses.

The intrigue: Dollars aren't all that's flowing between X and xAI.

• xAI's models are already using the entire corpus of X user posts as training data.
• For users who object to that, the new ownership structure makes the arrangement that much more difficult to ignore.

Meta does exactly the same thing with its users' posts, of course.

• Google and YouTube appear to be somewhat less aggressive, but also apply user data to build a variety of AI-powered features.
• Then again, all of Google's services, starting with search, are essentially one giant algorithmic supercomputer powered by user data. Generative AI is just the newest and outermost layer.

The bottom line: As AI's conquest of tech-industry mindshare advances, social media is decreasingly important as a business — and increasingly treated as an exploitable data resource to build AI.

Chipmakers, cloud providers, energy producers and AI companies are all flooring the pedal on infrastructure spending to support an AI-driven world that doesn't yet exist.

Why it matters: Investors are placing hundred-billion-dollar bets that demand for AI is about to explode, while the technology has yet to persuasively demonstrate its mass consumer appeal or its business-efficiency benefits.

Driving the news: The AI Infrastructure Partnership announced Wednesday that it was adding Nvidia and xAI as new partners in its fund, which aims to build data centers and energy facilities supporting AI, mostly in the U.S.

• Microsoft, BlackRock and UAE-based MGX were already key investors in what was announced last September as a $30 billion fund that could be leveraged up to $100 billion.
• Thursday, Nvidia CEO Jensen Huang told the Financial Times he expects his company will manufacture "several hundred billion" dollars' worth of electronics in the U.S. over the next four years.
• The Abu Dhabi-based wealth fund ADQ and U.S. heavyweight Energy Capital Partners intend to invest over $25 billion in projects to power data centers and other industrial consumers, Axios' Ben Geman reported Wednesday.

These deals and announcements all follow the high-profile launch of Stargate, a partnership — including OpenAI, Oracle, SoftBank and MGX — that's raising an initial $100 billion (toward an aspirational total of $500 billion) to build U.S. data centers for OpenAI.

• The Stargate announcement came straight from President Trump in the Oval Office the day after his inauguration.

The big picture: The new deals also further extend an AI spending spree that saw massive capital-expenditure growth from tech giants Microsoft, Alphabet/Google, Amazon and Meta in 2024 — a trend the companies say will continue.

The intrigue: The Trump administration is touting all this domestic investment as a win for its efforts to boost domestic manufacturing and growth.

• But much of the investment would likely have happened no matter who was in the White House.

Two key forces are driving the spending: AI euphoria and geopolitical risk.

• The euphoria is the tech industry's certainty that AI is destined to be the underlying technology for a new, bigger-than-ever wave of digital growth.
• The risk is that the AI industry is dependent on a handful of dominant monopolies — U.S.-based Nvidia (which designs chips), Taiwan-based TSMC (which manufactures chips) and Netherlands-based ASML (which makes the machines that make chips).
• This wildly profitable global semiconductor supply chain is vulnerable in the event of an attack on Taiwan by China, which has long claimed the island, or a major earthquake in Taiwan.

The other side: This week Nvidia announced the next generation of its advanced AI chips, called Rubin, and promised it would be twice as fast as today's top-of-the-line Blackwell.

• But The Information reports that Blackwell is only now beginning to arrive in most companies' data centers.
• While "Nvidia's biggest customers like Google, Meta and OpenAI are dying to receive clusters of Blackwells as they race to develop cutting-edge AI, no one else seems as interested," the Information's Anissa Gardizy writes.
• That's a striking contrast to the scramble for, and shortages of, Hopper chips — the generation that preceded Blackwell — in the period after ChatGPT's debut more than two years ago.

The bottom line: Tech in general, and semiconductors in particular, have always been cyclical businesses. First you overbuild, then you suffer through a glut, then there's a shortage — and then you overbuild again.

• It happened with memory chips at the start of the '90s, and it happened with internet connectivity during the dotcom boom and bust.
• Plenty of people in the AI industry believe this time is different. History suggests they're headed for a letdown.

Tech evangelists predict the arrival of "superintelligence" any year now, but others doubt AI will ever produce its own Leonardos and Einsteins.

Driving the news: In a post on X Tuesday, OpenAI CEO Sam Altman touted the company's development of "a new model that is good at creative writing" and showed off its work — a thousand-word "metafictional" composition on "AI and grief."

Why it matters: Creativity could be the final hurdle for AI to leap in proving it's humanity's peer — but until then, many see it as the last bastion of humanity's irreplaceability.

The big picture: Whether telling stories or researching scientific breakthroughs, today's generative AI isn't very good at creative leaps and novel insights.

• It's bounded by what it "knows" — the data it is trained on — and how it "thinks," by guessing the next word or pixel that best fulfills its prompt.

In science, our AI models aren't going to push the boundaries because they're too eager to please people and prove their utility, Thomas Wolf, Hugging Face's co-founder and chief science officer, wrote on X last week.

• Wolf called AI that does research "yes-men on servers."
• "To create an Einstein in a data center, we don't just need a system that knows all the answers, but rather one that can ask questions nobody else has thought of or dared to ask," Wolf argued.
• The benchmarks we're using to gauge AI's advances "consist of very difficult questions — usually written by PhDs — but with clear, closed-end answers. ... Real scientific breakthroughs will come not from answering known questions, but from asking challenging new questions and questioning common conceptions and previous ideas."

Getting AI to produce compelling art looks even more unlikely.

• Most work produced by AI is literally derivative. Of course, most artists, especially at the start of their careers, learn by imitation, and many human artworks are effectively collages, rewrites or remixes.
• But memorable artists develop distinctive voices by mixing their own experiences and obsessions with whatever they've learned from the artists they admire — and even their collages "sound like them."
• People seek out art because hearing those voices inspires them, leaving them feeling connected with the artist in a way that they cherish.

The short story Altman posted showed formal facility — but many of the responses on X found it, as I did, more exercise than expression.

Between the lines: Plenty of artists will find AI a valuable creative tool or an aid to brainstorming, just as many researchers will employ it to speed their work.

• But creation is likely to remain hard work for human beings. It takes effort to wrestle a vague idea in your head into words, images or any other material for an audience to encounter.
• This is the sort of friction that AI visionaries sometimes promise to liberate us from.

Yes, but: "Friction-free art" is inert. What sends off sparks is the struggle of a person's urge to express something against the limits of form and medium.

The bottom line: LLMs are like youngsters who have read a lot but do not have experience of the world. And right now there's not much of a way for AIs to get it.

• An LLM has never felt sunlight on its arm or raindrops on its head, known a parent or a child, given birth or faced death.
• It doesn't feel the need to share such experiences or to shape them into works of writing, or music, or any other form.

What's next: Maybe the fusion of generative AI with robotics will surprise us, and an embodied LLM will find itself moving toward something humans might recognize as art.

• But it's very possible AI will never be truly creative because it has no impulse to play around for the heck of it, to impress peers or best rivals, or to leave a little mark on the world. People give AI prompts, whereas human artists get their prompts from their own lives.

While many AI insiders are betting on some kind of all-in-one AGI breakthrough within a handful of years, on-the-ground progress has clearly split into two diverging paths: optimizing AI to write code and do math, or improving AI's "soft" skills with words and creativity.

The big picture: On the one hand, the latest wave of "reasoning models" excel at computer programming and quantitative analysis. They're AI's nerdy tech geniuses.

• On the other hand, a handful of new projects show AI getting much better at replicating the nuances of how humans communicate. These are AI's English majors and theater kids.

Why it matters: For now, at least, AI development is replicating the humanities vs. STEM divide that has always split college-student populations — or, in high school terms, freaks and geeks.

Zoom in: Three developing stories this week show the process at work.

Alibaba's Qwen surprise

Chinese tech giant Alibaba Thursday released a new model, called Qwen QwQ-32B, that matches the performance of DeepSeek's R1 but requires a fraction of the computing power to run.

• Like DeepSeek, the new Qwen model is open source, and both Chinese projects are reasoning models that excel at technical work.
• The Qwen release is one more sign that plenty of further progress is possible in making this kind of AI more efficient.
• Reasoning models are making the transition from "amazing new breakthrough" to "cheap and widely available commodity" in record time.

OpenAI's taste-maker

After a week of hands-on experience with OpenAI's latest and biggest model, GPT-4.5, AI experts remain a little puzzled by it, given that it costs a fortune to use yet doesn't break benchmark records.

• But one consensus has emerged among fans of GPT-4.5: The new model has "taste."
• "It appears so far that GPT-4.5 has advantages in places like verbal intelligence, contextual adaptation, detailed knowledge, and a kind of abstract writing skill. It has better taste and aesthetics," blogger Zvi Mowshowitz wrote in a roundup of GPT-4.5 assessments.
• Economist-blogger Tyler Cowen wrote, "I view it as a model that attempts to improve on the dimension of aesthetics only."
• Or, as Sam Altman put it in introducing the model, "There's a magic to it I haven't felt before."

These endorsements offer an inescapable paradox.

• Nuanced facility with language is just the kind of skill that human beings have always viewed as their own unique preserve.
• But there's no way to measure it. It's purely subjective. And what looks like a creative leap to one observer can resemble a failure to someone else.

Sesame's charmed voice

One recent buzz magnet in AI circles is Sesame's Conversational Speech Model, released last week.

• Many testers of the live demo say Sesame's AI achieves a new level of confident ease in imitating the conversational flow and subtle imperfections of an actual talking human.

This is another advance that's very difficult to quantify or benchmark but could profoundly impact how AI is integrated into society.

• It's also easy to imagine how it could release a tide of confusion, fraud and scams.

Zoom out: The idea that digital intelligence would evolve on multiple tracks mirroring the science/humanities divide has a long pedigree.

• Just think of how the original "Star Wars" split its picture of advanced robotics between the technician R2-D2 and the diplomat C3PO.

Yes, but: "Freaks and geeks" are names for schoolroom cliques and types of aptitude. But most successful people excel to a degree in both realms, and any AI that aims to match humanity will need to do the same.

What we're watching: If the AGI optimists are to make good on their "human-level AI in the next two years" predictions, model developers will have to figure out how to bring these two evolutionary paths back together.

• "Easy!" say the engineers and entrepreneurs — this problem will succumb to enough brainpower and effort.
• Not so fast! say the humanists and historians — C.P. Snow's "two cultures" divide has been with us for many decades, and it's arrogant to think that AI is the field that will finally heal it.

Every quarter since ChatGPT's debut, Nvidia's earnings release has tied a knot in Silicon Valley's gut as investors wait for the numbers that will prolong the AI boom — or end it.

Why it matters: It's risky when any industry hangs so much of its hopes on one company's results, and Nvidia's enviable record of beating expectations means the slightest faltering could trigger a rout.

Driving the news: Today, markets await the chipmaker's first report since the arrival of DeepSeek's latest model last month cast a brief shadow over Nvidia's glow.

• DeepSeek rivaled the industry's most advanced AI models in a cheaply trained open-source package.
• That suggested the long-term demand for Nvidia's powerful but costly chips might be lower than projected.

Yes, but: Tech giants and startups in the U.S. and around the globe continue to pour hundreds of billions into AI infrastructure, new model training and data centers.

• All those projects keep orders flowing to Nvidia, which not only continues to lead the market for high-performing chips but also controls an ecosystem of supporting software tools that help AI makers optimize their products.

The big picture: There are three broader reasons to think that Nvidia could disappoint investors, if not this week then eventually.

1. The market is overconcentrated.

• The S&P 500's incredible run the last two years has been driven by gains in the so-called Magnificent 7 stocks, led by Nvidia (and also including Microsoft, Apple, Meta, Google, Amazon and Tesla).
• The Magnificent 7 accounted for more than half of the S&P's gains in 2023 and 2024, and now makes up more than 30% of the index's total market capitalization.
• By one key metric, the market was more concentrated in late 2024 than ever before — including during the dot-com bubble.

2. Demand for AI remains elusive.

• Nvidia's astonishing market ride rests on the assumption that AI will be the tech industry's next universal platform and that demand for AI products and services will be massive.
• So far, although ChatGPT and its rivals show healthy growth in usage, businesses and consumers haven't always embraced the tools. Real-world applications beyond a few specialized fields like software programming and customer support have yet to take off.
• If that doesn't change, the AI industry — along with its leading toolmaker, Nvidia — could face a sobering correction.

3. The AI chip market is uniquely vulnerable to geopolitical risk.

• That's because Nvidia only designs its chips. They're manufactured in Taiwan by TSMC.
• Taiwan has long faced the danger of an invasion or blockade by China, which has claimed the island for decades.
• Trump may have given Beijing reason to doubt he would step in to protect Taiwan. Among other remarks, Trump lamented during the campaign that Taiwan "wants protection" but "they stole our chip business."

The other side: Nvidia could keep outperforming expectations for a long time.

• A bad quarter for the company would also be a disaster for the market in the short term, but not necessarily for Nvidia's long-term outlook.
• Even if the AI bubble pops, the chipmaker would remain an incredibly valuable repository of intellectual property, design skills and research power.
• Short-term investors might take a bath, but the company would live to thrive again.

The bottom line: Silicon Valley's 75-year history has been one long cycle of booms and busts. With AI it shows every sign of continuing that pattern.

A hot startup that grew overnight into a billion-dollar behemoth is racing with established tech giants for supremacy in a new market that everyone expects will unlock a future of abundance and profit.

Flashback: That sounds like a description of OpenAI vs Google et al., but it's actually an account of the "browser wars" at the dawn of the web 30 years ago — when Netscape vied with Microsoft to control the software people would use to access the internet.

Why it matters: In 1996 or 1997, a couple years after forward-looking tech leaders first realized that "owning" the web browser would be a prize, Google — the company that would ultimately win the race — didn't even exist.

Today, as AI giants and challengers vie to build a better chatbot and seize mindshare and market share, there is similarly a good possibility that the winning bot (assuming there is only one) has not yet been invented, and the company that will make it has yet to be founded.

That's why tech's superpowers, despite their immense wealth and influence, have been running scared.

• It's also why VCs continue to pour money into new startups like Mira Murati's Thinking Machines Lab and Ilya Sutskever's Safe Superintelligence.

It can be baffling to watch the proliferation of these companies in a market that's already led by the likes of OpenAI's ChatGPT, Google's Gemini and Anthropic's Claude — with well-funded competitors like Elon Musk's xAI, open-source offerings like Meta's Llama and new contenders like China's DeepSeek also thronging the AI space.

The space's crowding leads some observers to see today's AI splurge as a bubble, and it almost certainly is. Many or most of these companies and products will fail — indeed, culling has already begun.

• But the logic of venture capital assumes that, even as most companies will fail, a few investors will hit the jackpot by getting an early piece of a company that grows unfathomably profitable the way Google or Facebook/Meta, did.

No one knows how this will play out in AI.

• The jackpot company could be one of today's market leaders like OpenAI. It could be a dark horse that's still a back-of-a-napkin sketch on some founder's dining-room table. Or maybe one of today's established giants will end up owning the market.

The browser wars make an instructive parallel.

• In the '90s, Netscape was in the OpenAI position — it kicked off the new market with fast updates of its free Navigator browser and wowed the world with a skyrocketing initial public offering in 1995.
• Microsoft fought back with the introduction of the competing Internet Explorer, a flop at first that gradually improved and won users thanks to its integration with Microsoft's dominant operating system.
• Microsoft won the first battle — triggering a massive antitrust lawsuit by the Justice Department. But over the following decade IE lost ground to more innovative competitors like Mozilla.

Google didn't even introduce Chrome until 2008, and today it's by far the most popular browser in the world.

• But by that time, the victory didn't seem to matter as much. The smartphone revolution was underway, and apps were taking on much of the "gateway to the internet" role from browsers.

The bottom line: Tech's platform shifts may feel high-velocity, but they take a long time to unfold — and you should never feel too sure you know who is going to own the future.

Nearly every computer system draws a line between the right to look at files and the right to change them — but till now, the details barely mattered to most non-programmers.

The big picture: The early days of Trump's second administration — as Elon Musk and his Department of Government Efficiency (DOGE) crew execute a hostile takeover of the federal government's digital infrastructure — are giving Washington a crash course in the importance of system permissions.

• With "read-only" permission, you can open, review and copy data and programs.
• With "read-write" access, you can delete files, alter data and rewrite code.

Why it matters: Much of the capital and the nation is still trying to figure out whether Musk is simply scouring the federal books to flag waste and fraud — as the Trump transition originally described the DOGE effort — or if he has unilaterally, and maybe unconstitutionally, seized a sort of line-item veto power over every dollar the government spends.

• "Read" or "write" could be the difference between a budget review and a spending freeze. It could also be the difference between speeding up the government or tearing it down.

Catch up quick: Over the past few days, reports that Musk and his team had gained unprecedented access to broad swathes of federal systems came with assurances that DOGE would have "read-only" access.

• The sharpest controversy has been over the Treasury Department's payment system — a venerable corner of the bureaucracy responsible for actually writing checks from federal accounts to everyone from military contractors to Social Security recipients.

Multiple outlets reported alarm late last week among nonpartisan career Treasury employees that DOGE workers had gained access to the payment system.

• The Wall Street Journal and Politico wrote that anonymous administration sources said Musk's team would only be reading Treasury information, not changing it.

But a Monday story in Wired says DOGE has obtained much broader "read-write" capabilities.

• "A 25-year-old engineer ... who previously worked for two Elon Musk companies, has direct access to Treasury Department systems," Wired reports, citing three unnamed sources.
• The engineer's access includes "the ability not just to read but to write code on two of the most sensitive systems in the U.S. government: the Payment Automation Manager and Secure Payment System at the Bureau of the Fiscal Service," two of those sources told Wired.

The other side: A letter from an unnamed Treasury Department official to members of Congress Tuesday afternoon defended the DOGE project.

• "Currently, Treasury staff members working with Tom Krause, a Treasury employee, will have read-only access to the coded data of the Fiscal Service's payment systems," the letter read. "This is similar to the kind of access that Treasury provides to individuals reviewing Treasury systems, such as auditors."
• Krause, CEO of Cloud Software Group, is now serving as a "special government employee" at Treasury, according to the letter.

Context: No documented evidence has surfaced to date showing anyone at DOGE has meddled with Treasury data or programs.

• But concerns over what DOGE employees might do with "read-write" power may not be entirely unfounded, given Musk's repeated boasts on X that he and DOGE have "deleted" one federal organization or that another agency should "die."

There are three roads this story could take from here.

1. The DOGE project really does stick to "read-only" work.

• Providing the Trump administration and Congress with informed recommendations for saving taxpayers' money, rather than taking direct action, would defuse much of the controversy — but hardly seems in character for Musk.

2. DOGE's team keeps working in a fog of confusion and no one really knows what they're doing inside federal systems.

• That means whatever happens in the next government IT meltdown or federal shutdown crisis will be blamed on them.

3. Musk starts directly blocking Treasury payments, provoking a constitutional crisis.

• If you don't mind breaking things, you can drastically alter IT systems a lot faster than any judge can stop you.

Two important facts are worth remembering, whatever happens.

• Even plain read-only access is plenty sensitive, given the nature of the payment system's data, which includes Social Security numbers and tax information.
• While tackling this project, Musk remains not only the wealthiest person on the planet but the owner and CEO of multiple giant companies that do billions in direct business with the U.S. government.

Breakthroughs from Chinese AI startup DeepSeek have stunned Silicon Valley and could bring turbulence to Wall Street, as they were accomplished at a fraction of what the U.S. giants are spending and despite export bans on top-of-the-line chips.

Why it matters: China's rapid advances suggest America's strategy of withholding technology from China might just be speeding up the evolution of its rival's AI knowhow.

• DeepSeek's rise is alarming the likes of Meta, which announced Friday that it plans $60 billion-$65 billion in capital investment this year as it scales up its own AI projects.
• But it could potentially also be bad news for Nvidia, which designs the world's most advanced AI chips, because DeepSeek is proving that rapid advances are possible even with fewer and less sophisticated chips.
• Nvidia's stock slid on Friday and again in overnight trading last night, pulling the Nasdaq down with it.

Driving the news: DeepSeek hit No. 1 on Apple's App Store a week after the Jan. 20 release of its R1 model, which works along similar lines to OpenAI's o1.

• Presented with a complex challenge, it takes time to consider alternate approaches before picking the best solution — and it explains its chain of reasoning to users. These "reasoning" models are especially good at coding and math.
• Just last month another DeepSeek model, v3, stunned AI experts by providing performance comparable to OpenAI's and Anthropic's most advanced publicly available general models, as Axios reported.
• The kicker is that DeepSeek created and released its entirely open source project for about $6 million in training costs ("a joke of a budget," in one expert's words). OpenAI is spending hundreds of millions of dollars.

The results from China have turned eyes around the world and revved up concerns in the U.S. that its lead in the so-called AI race between the two superpowers may be shrinking.

• DeepSeek spun out of a Chinese hedge-fund firm two years ago, hired ambitious young AI scientists, and set them to figure out more efficient ways to develop models, per Wired, and they focused on basic research rather than consumer product development.

The other side: It's not as though the U.S.' efforts to keep the best chips out of China haven't had an impact.

• DeepSeek is said to have already amassed a training network of 10,000 Nvidia H100s by the time U.S. sanctions were introduced in 2022.
• But DeepSeek's CEO has said that the export controls, and resulting scarcity of Nvidia's top-of-the-line chips, have been "a problem."

No one knows where DeepSeek would stand today if it didn't face those roadblocks.

• We also can't say whether DeepSeek would be making such rapid advances on its own without having the latest work from OpenAI and its U.S. competitors to aim at.

Our thought bubble: Scarcity drives innovation, China has some great AI minds at work, and the U.S. might need to rethink its strategy.

• Silicon Valley's startup culture relentlessly pushes research in the direction of the consumer market. But sometimes, particularly when a field is young and applications aren't immediately obvious, basic research is even more important than market share — and open research tends to overwhelm secret research.
• OpenAI's pivot from "world's best AI lab" to "aspiring consumer tech giant" has wowed the financial world but could leave the U.S. in second place.

"Volatile" is the one word that describes the tech industry's new reality under the second Trump administration.

Why it matters: With AI booming, social media splintering and crypto inflating, the giant companies that drive today's economy face a huge spike in uncertainty.

The big picture: President Trump's track record of sudden policy about-faces and unpredictable lurches governed by whim and self-interest puts CEOs in a defensive crouch.

• Some, like Meta's Mark Zuckerberg, have enthusiastically moved into the MAGA tent, joining tech billionaires like Elon Musk and Marc Andreessen. Others are standing outside but paying ritual tribute.
• Silicon Valley leaders won't shed tears, for the most part, as Trump's team disassembles Washington's half-constructed regulatory regime for AI and backs off from the Biden administration's full-court press on monopoly and competition.
• But Trump's priority of shutting the U.S.'s immigration door will directly harm the tech workforce. And the new president's vow to stifle trade with tariffs, if fulfilled, could disrupt tech's global supply chain in unpredictable ways.

Here are four trends that will shape tech's free-for-all world in the new Trump era.

1. Rail-free artificial intelligence

• Trump has already revoked former President Biden's AI executive order as part of his blizzard of day-one declarations.
• The incoming president has given no sign that he believes the government has an important role to play in regulating tech's new platform beyond assuring that conservative voices aren't limited.
• Elon Musk has long held the view that advanced AI could threaten humanity's future. But that fear does not appear to be front of mind as he tackles his mission of cutting the federal budget.

2. Unchained crypto

• Cryptocurrency investors and developers see Trump's return as the start of a new golden age: Regulators will get off their backs, and thousands of coins will bloom.
• There will likely be a burst of new experiments in blockchain-driven democracy, too.
• But the inauguration-eve floating of new meme coins for both the new president and first lady Melania Trump made clear where the biggest action will be: lots of fast money new coin issues, many of which will be soak-the-suckers pump and dump schemes.
• Even some Trump-friendly crypto insiders were uncomfortable with the brazen profiteering of the Trump-coin schemes.
• Things will get progressively crazier until another big failure on the scale of FTX causes a new bust.

3. Deals dynamo

• With regulators sidelined and vast amounts of cash sloshing around the globe, expect deal-making in tech to accelerate, even if the Federal Reserve keeps trying to hit the brakes.
• The biggest companies still face ongoing antitrust investigations and lawsuits, but they're likely to be less cautious and more opportunistic.
• Any deal that holds the possibility of enriching the president, his family or his allies will have that much extra juice.

4. Social media's "big sort"

• American internet companies and users are rapidly segregating themselves into separate "red" and "blue" environments.
• Meta's move to embrace MAGA and X's harbor for extreme right-wing voices are speeding this process, driving opponents to new rival media platforms on the other side of the political fence.
• Online, at least, this process dooms the quest for common ground and assures further estrangement of America's two parties and cultures.

The bottom line: Volatility compounds across different domains.

• What happens when autonomous AI agents, steered by blockchain-based voting, start executing deals on tribalized social-media platforms?
• Cyberpunk authors dreamed such visions as dystopias, but team Trump sees them as blueprints.

Chinese AI makers have learned to build powerful AI models that perform just short of the U.S.'s most advanced competition while using far less money, chips and power.

Why it matters: American policies restricting the flow of top-end AI semiconductors and know-how to China may have helped maintain a short U.S. lead at the outer reaches of the AI performance curve — but they've also accelerated Chinese progress in building high-end AI more efficiently.

Catch up quick: In late December, Hangzhou-based DeepSeek released V3, an open-source large language model whose performance on various benchmark tests puts it in the same league as OpenAI's 4o and Anthropic's Claude 3.5 Sonnet.

• Those are the most advanced AI models these companies currently offer to the broad public, though both OpenAI and Anthropic have next-generation models in their pipeline.

Stunning stat: Training V3 cost DeepSeek roughly $5.6 million, according to the company.

• OpenAI, Google and Anthropic have reportedly spent hundreds of millions of dollars to build and train their current models, and expect to spend billions in the future.
• AI pioneer Andrej Karpathy called DeepSeek's investment "a joke of a budget" and described the result as "a highly impressive display of research and engineering under resource constraints."

Between the lines: In an interview last year, DeepSeek CEO Liang Wenfeng said, "Money has never been the problem for us; bans on shipments of advanced chips are the problem."

• The V3 model was trained on Nvidia H800 chips, a less-powerful version of a chip the U.S. banned for export to China in 2022. Export of the H800 was then prohibited when the U.S. tightened controls again the following year.

The big picture: Some U.S. officials have argued for restricting China's access to advanced AI chips even further in hopes of slowing the country's development of the technology.

• On Monday, the Biden administration announced another big round of export controls aimed at choking the supply of chips to China via third-party countries.

What's next: Advances like V3 and OpenAI's powerful new "reasoning" model, o3, have lent weight to recent claims by OpenAI CEO Sam Altman and other industry leaders who predict the industry is closing in fast on artificial general intelligence (AGI). (Plenty of other observers remain skeptical.)

• AGI — or AI that can solve problems and perform tasks at a human or beyond-human level — is a holy grail for AI researchers, and many in the industry and U.S. government believe the technology's first developer will win a massive economic, scientific and security edge.
• Biden's latest export controls have led some observers to conclude the government shares the growing sense that AGI is close.
• "This is a 'break in case of emergency' policy, and the Biden administration identified that the emergency is that AGI is just a few years away," Gregory Allen, director of the Wadhwani AI Center at the Center for Strategic and International Studies, tells Axios.

Yes, but: AGI is also not well-defined, and both optimists and pessimists have complained that it's become a moving goalpost.

They say it's hard to turn a battleship around, but Mark Zuckerberg just about-faced his globe-spanning, $1.5 trillion-value, 3 billion-user company — transforming Meta from a bastion of Silicon Valley's socially progressive neoliberalism into a full-on MAGA hive.

Why it matters: After Zuckerberg's embrace of Trump and Trumpism, Silicon Valley is holding its breath to see whether a whole row of tech dominoes is about to fall in the same direction.

Some early signs of wobble:

• Google, Apple, Microsoft and Amazon have all lined up to give $1 million each to the Trump inauguration. (Apple's donation came personally from CEO Tim Cook.)
• Amazon put $40 million behind a Melania Trump documentary.

State of play: So far, while Meta's competitors have ritualistically expressed their willingness to work with the new administration, none of them has gone as far as Zuckerberg in donning the corporate equivalent of a MAGA hat.

• This week he announced Meta was ending its fact-checking program; loosening restrictions on hate speech; and shutting down the company's diversity efforts.

Publicly traded companies with billions of customers generally try not to alienate any large bloc of the public. Becoming closely aligned with either side of the U.S.'s red/blue divide risks limiting a business's market reach.

• Until now, Elon Musk has been the striking exception to this rule. Zuckerberg makes two — and his moves open the door wider for anyone else who wants to emulate them.

Yes, but: Zuckerberg, unlike his rival CEOs, has absolute voting control of his company.

• As he said in a three-hour conversation with Joe Rogan Friday, "Because I control our company, I have the benefit of not having to convince the board not to fire me."
• None of the other members of tech's trillion-dollar club can move with the same speed or independence, even if they wanted to.

Case in point: Apple has always aimed, and often managed, to transcend mere politics and inhabit a separate dimension making "great products that people love."

• Cook worked with Trump last time around, winning "Tim Apple" as a sobriquet but also occasionally speaking out about the importance of protecting immigrant workers and LGBTQ rights.
• Google and Microsoft — both with immigrant, non-white CEOs — also offered some pushback on those issues during Trump's first administration as well.
• Tech leaders united to oppose Trump's ban on transgender troops and, in a variety of combinations, took on some of his immigration policies.

This time around, these firms are quietly signaling they want to cooperate with the new Trump team on issues — like competition with China — where they see common ground.

• All these companies were the target of a federal antitrust full-court press in the Biden era, and each is wondering whether they can get a reprieve from the new team in D.C.
• Every tech giant also wants to expand the use of skilled-worker visas, and that has already led to friction with the immigrant-hostile MAGA movement.

During the first Trump term, an activist young tech work force occasionally took to the barricades to protest government policies and pressure reforms from their employers.

• But multiple rounds of layoffs and a messy pandemic recovery have left workers feeling more insecure.
• While Zuckerberg's new policies have sparked internal debate and dissent at Meta, the CEO may welcome an exodus of dissenters as a sort of voluntary lay-off.
• In a Threads post, he described users who might quit Meta's platforms in protest as "virtue-signaling."

What we're watching: With each fresh controversy the new administration touches off, tech CEOs will have to navigate a maze involving Trump's demands for loyalty, employees' emotions and wishes, and their own strategies.

• If Zuckerberg's experiment of committing a whole megacompany to taking overt sides in America's vast culture war pays off, others could follow.

The bottom line: Trump used to say that Zuckerberg would "spend the rest of his life in prison." But the incoming president's relationships with business leaders are strictly transactional, and Meta's CEO is probably resting a lot easier now.

Distinguishing truth from falsehood is frustrating, endless, thankless work — and now Mark Zuckerberg is walking away from it.

The big picture: Facebook's latest content-moderation pivot looks like part of a plan to win over Donald Trump as he takes power again. But the field Zuckerberg is abandoning is one he never wanted to play on in the first place.

State of play: The founders of social media giants like Facebook, Instagram, Twitter and TikTok didn't expect to end up in what the industry came to call the "content moderation" business — and what many critics, and now Zuckerberg himself, denounce as "censorship."

• Policing online speech costs a fortune to do right. It's impossible to make everyone happy. You're bound to make mistakes. And users' wishes keep changing.
• The whole effort is a distraction from what's always been Facebook/Meta's top priority — boosting engagement to sell more ads.
• Meta faces huge challenges this year, particularly an April trial in the Federal Trade Commission's suit to unwind its decade-old acquisition of Instagram and WhatsApp. The new Trump-friendly approach to content moderation is one of many efforts to win over the new administration, which is open about rewarding friends and punishing enemies.

Zuckerberg staked out a free-speech position in a 2019 speech at Georgetown. A few months later, he said that social media networks shouldn't try to be "arbiters of truth" — but at the same time Facebook was ramping up its truth arbitration.

• After taking blame for spreading misinformation during the 2016 election and violating users' privacy during the Cambridge Analytica scandal, Facebook was under enormous pressure to clean up its act, and the company made big investments in expanding its moderation efforts.
• It was also in 2019 that Facebook started a program using third-party fact-checking organizations from a variety of political perspectives to help it identify and limit the spread of potentially dangerous misinformation.

The fact-checking program has drawn fire throughout its existence.

• The kinds of topics it confronted — controversies over climate science, COVID-19 and vaccines, charges of election fraud — are often both matters of fact or science and also flashpoints for partisan rage.
• Believers in fact-checking insist that there's value to society in telling the public what is — and isn't — authoritative information, grounded in vetted research and verifiable records, in fields like medicine and public affairs.
• Critics say there's always another point of view that deserves to be heard, and blocking any perspective is a form of censorship.

Between the lines: Facebook tried to solve some of its content moderation headaches by setting up the independent Oversight Board.

• The company handed the Oversight Board hundreds of millions of dollars beginning in 2019 to build a kind of Supreme Court for user complaints.
• It's been particularly effective in sorting out complex issues beyond U.S. borders.
• But it hasn't insulated Zuckerberg and Meta from criticism by American conservatives and Congressional committees.
• Notably, Meta's announcements Tuesday failed to mention the Oversight Board at all.

Zoom out: Zuckerberg calls Meta's new approach a "back-to-our-roots" embrace of free expression. But there's never been any medium where absolute free speech reigned.

• Platform owners have legal obligations to governments of countries they operate in to obey the law.
• In the U.S. that means dealing with laws governing what Zuckerberg describes as "legitimately bad stuff" like "drugs, terrorism, child exploitation."

A second category where platform owners have generally felt an obligation to intervene is speech that could cause imminent harm.

• That might include death threats, violent conspiracies, or even plans to attack the U.S. Capitol.

Then there's the category of hate speech.

• It bedevils social-media owners, because it's constantly shifting and varies across cultures. But in any given time and place there are some slurs that violate public norms, and a global public platform can't just ignore them.
• Along with the company's other new content policies, Meta has now updated its Community Guidelines on "hateful conduct" specifically to allow "allegations of mental illness or abnormality when based on gender or sexual orientation." A variety of other changes in its rules appear to significantly loosen the knot around what many users will consider hate speech.

A final category of content moderation — most relevant to fact-checking — is misinformation (widely shared but inaccurate info) and disinformation (misinformation with deliberate bad intent).

• Zuckerberg and other social media owners hate playing content cop in this realm and would rather let users sort such issues out for themselves.
• His plan is for Meta to copy X's Community Notes approach, which lets users flag other users' posts for inaccuracies.

What we're watching: When Elon Musk rewrote Twitter's old content rules for X, the platform's never-decorous conversations deteriorated further. Today you don't have to look far on X to find posts espousing racism and antisemitism or deriding LGBTQ people.

• Musk is proud of what he's done with X, but it hasn't helped turn around his business.
• We don't yet know how Zuckerberg's version of "more free speech" will play out, but if Meta's platforms get nastier and uglier, too, advertisers could be spooked — and users who aren't on the MAGA side of the fence could flee.

Our thought bubble: Decades of human experience online shows that running any kind of community platform is like gardening — if you let the weeds go wild, the flowers will choke.

• This isn't Mill's free market of ideas; it's a world where bad speech drives out good speech.
• And if you think there's no such thing as bad speech, you haven't been on the internet for the last 25 years.

Wednesday's extraordinary display of Elon Musk's social media power saw the world's richest man funnel the anger of his hundreds of millions of online followers against a bipartisan compromise funding bill.

Why it matters: Musk's arresting demonstration of government-by-tweet left Washington reeling — but the magic trick will get harder to repeat.

The big picture: Shooting down a spending bill is a lot easier than passing one. Stopping government in its tracks is much easier than making it work.

• Beginning Jan. 20, Republicans will essentially control all three branches of government, and making it work will be their job.

How it works: Musk's X has become a partisan megaphone fine-tuned to promote its owner's own messages to a user base of MAGA supporters, libertarians, crypto enthusiasts, and fans of Musk himself.

• Its influence is strongest with Republican House members because its base overlaps with theirs.
• The other half of America has been voting with its feet, with many Democrats and progressives abandoning X in favor of a variety of online alternatives.

Between the lines: Musk's Wednesday call to arms took place at his point of maximum power.

• The legislators he wanted to pressure answer to the votes of the people who listen to him.

Yes, but: Republicans' control of Congress beginning in January will be precarious.

• The GOP has a margin of only a couple of votes in the House, and the new Senate majority will often hang on a handful of Republican moderates' votes.
• When it comes time to build any kind of coalition to pass a new law or a spending measure, Musk's megaphone is likely to prove a clumsy tool.
• If the Musk-led "Department of Government Efficiency" recommends cuts to Social Security or health care, mustering the same kind of support could be a lot harder — and, historically, opponents of such measures have found effective ways to exert their own "vox populi."

Right now, Trump and Musk seem to be on the same page. That could change any time.

• Trump-watchers suggest the president-elect's jealousy reflex could already have been triggered by Musk's high-profile power flex.

The "bully pulpit" got its name over a century ago, and American leaders have always urged the people to rise up and write, call or email their representatives.

• Musk's pulpit moves strikingly faster — getting results within hours on Wednesday.

What to watch: If Musk can repeat the trick, he will have ushered us into a new era.

• Then we'll have to figure out whether what's happening is really direct democracy, or rule by unelected billionaire.

Go deeper: Musk's unprecedented power

OpenAI is putting its powerful video creation tool, Sora, in the hands of basically every one of the millions of people who pay for ChatGPT, the company said yesterday.

Why it matters: We're about to witness an at-scale experiment in what it means for people to create and consume large quantities of video content that is photorealistic but fake.

The big picture: OpenAI released Sora precisely because it wants the results of that experiment.

• "We're introducing our video generation technology now to give society time to explore its possibilities and co-develop norms and safeguards that ensure it's used responsibly as the field advances," the company wrote in a post introducing the new product.

Catch up quick: When OpenAI gave the public a sneak peek at Sora last February, the clips the company showed caused jaws to drop — but also triggered an allergic reaction in Hollywood.

• Many artists view the advent of AI-generated video as a fraud and an insult. Just as with text-based AI, many creators also believe OpenAI trained its tool using copyrighted works.

What they're saying: "We don't want the world to just be text," OpenAI CEO Sam Altman said in a live-streamed announcement yesterday. Video is "important to our culture."

The company said in a statement that the latest Turbo version of Sora, which will be offered as a standalone product to ChatGPT Plus and Pro customers, is "significantly faster" than the version the firm previewed. It lets users generate videos up to 20 seconds long.

An early review by Marques Brownlee, who got to play with Sora before the release, details some of what OpenAI admits are the "many limitations" of the tool.

• In particular, Sora's videos regularly seem to have problems handling basic physics. Objects appear and disappear, particularly when something else moves in front of them.

OpenAI isn't trying to pull the wool over anyone's eyes here. In Monday's live-streamed launch, Sora team leaders said the tool was "not about generating feature-length movies" but instead provides a "co-creative dynamic" so users can explore new ideas.

• The company imagines Sora not as a filmmaker but as a filmmaker's sounding board.
• The new version of the tool lets you drop your own images in as prompts. There's also a timeline editor that lets you add new prompts at specific moments in a video.
• In a demo at Monday's Fortune Brainstorm AI conference in San Francisco, OpenAI also showed off a feature — still in private testing — that allows users to narrate videos using AI-generated versions of their own voices, in multiple languages.

The company said it understands that putting Sora into so many hands could cause problems.

• Altman said OpenAI wants to prevent illegal use of the tool, "but we also want to balance that with creative expression."
• Users' ability to upload images of people will be "limited at launch," OpenAI's blog post said.
• Moderation of user content in Sora will be "starting a little conservative," but if it "doesn't quite get it right, just give us that feedback," Altman said.

Friction point: Sora will be available globally, except in Europe and the U.K. for now — presumably because of stringent EU privacy laws.

OpenAI's competitors are also moving fast with AI video-making offerings in the meantime.

• Google last week rolled out its Veo video generator to business partners, and Meta is developing Movie Gen — but neither is yet available to the general public.

Zoom out: Today's online world isn't exactly experiencing a shortage of brief videos.

• Logged-in Sora users can view a bottomless feed of "featured" Sora-made videos, alongside the prompts that created them.
• OpenAI suggests aspiring video producers will want to use this feed as a source of inspiration, and doubtless they will — but this viewing mode also feels a lot like TikTok or YouTube.

Some observers Monday applauded Sora's capabilities, while others predicted it would deluge us with AI "slop."

What we're watching: OpenAI aims to prevent outright illegal uses of Sora to create child sexual abuse material, impersonation and other problematic material.

• But even if the company succeeds, internet users have a way of pushing the boundaries of new tools like this until every possible awful output has been demonstrated.

The bottom line: Sora's evolution will give us all an early glimpse of how our social and political systems handle broad exposure to AI-made video.

Go deeper: Video-making AI tools are moving into general use

Competition in AI is less a single race than a triathlon: There's a face-off to develop the most advanced generative AI foundation models; a battle to win customers by making AI useful; and a struggle to build costly infrastructure that makes the first two goals possible.

Why it matters: Picking a winner in AI depends on which of these games you're watching most closely. And the competition's multi-faceted nature means there's more than one way to win.

State of play: Below, we offer an overview of where the key players in generative AI stand in each of this triathlon's events.

Between the lines: The benchmarks used in the industry to compare performance among AI models are widely criticized as unreliable or irrelevant, which makes head-to-head comparisons difficult.

• Often, developers' and users' preference for one model over another are very subjective.

OpenAI

Two years after the launch of OpenAI's ChatGPT kicked off the AI wave, the startup remains AI's flagship.

• It has raised about $22 billion and is in the process of retooling itself from a safety-oriented nonprofit to a globe-spanning for-profit tech giant.

Yes, but: OpenAI's last major foundation model release, GPT-4, is now nearly two years old. A long-awaited successor had its release pushed back into 2025 amid a swirl of reports that its advances may not be game-changing.

• Meanwhile, OpenAI has pushed the field's edge with innovations like its "reasoning" model, o1, and impressive voice capabilities.

Models: OpenAI still has a lead, but it's shrinking.

Customers: OpenAI has direct access to a vast pool of over 200 million weekly active ChatGPT users and indirect access to the huge installed base of Microsoft users, thanks to its close alliance with that giant.

Infrastructure: OpenAI is highly dependent on Microsoft for the cloud services that train and run its AI models, though it has recently begun an effort to expand its partnerships.

Anthropic

Like Avis to OpenAI's Hertz, Anthropic seems to be trying a little harder. It has also faced fewer distractions from high-profile departures and boardroom showdowns than its competitor.

Anthropic was founded by ex-OpenAI employees aiming to double down on OpenAI's commitment to caution and responsibility in deploying AI.

• But it has now raised roughly $14 billion and begun to embrace OpenAI's philosophy of putting AI in the public's hands to pressure test its dangers.

Models: Anthropic's Claude 3.5 Sonnet is widely viewed as a worthy competitor to GPT-4 that, in some cases, even surpasses it.

Customers: Claude's usage numbers are much smaller than ChatGPT's, but the company is partnering with big and medium-sized firms looking for a counterweight to Microsoft.

Infrastructure: Amazon, which recently invested $4 billion into Anthropic, is putting its massive cloud resources behind the company, and Google's parent Alphabet has also provided some investment and support.

Google

Google's long-term investments in AI research made generative AI's breakthrough possible.

• But ChatGPT's overnight success caught the search giant flat-footed.
• Google has spent much of the last two years in catch-up mode — uniting its DeepMind research team and Brain unit and injecting its Gemini AI across its product line.

Models: Google's Gemini is very much in the same league as OpenAI's and Anthropic's models, though some reports suggest that it hasn't found as much pickup among AI developers.

Customers: By pushing its own AI summaries to the top of search results and integrating its AI with its Android mobile operating system, Google has ensured that its own AI would get in front of a global user base.

Infrastructure: Google has the know-how and the resources to scale up as much AI power as it needs, but the field's competitive frenzy has left it off-balance.

Meta

Meta has embraced and promoted open source AI via its Llama models.

• The strategy is a way to avoid becoming dependent on a competitor for AI services — the way it found itself reliant on Apple and Google in the smartphone era.
• Meanwhile, Meta has been deploying its own custom version of the technology, dubbed Meta AI. The chatbot lives inside Messenger and WhatsApp, has taken over search in Instagram and powers the assistant on Meta Ray-Ban glasses. The future has Meta AI even spitting out its own posts.

Models: Meta's models haven't directly taken on OpenAI and its competitors. Instead, they've offered better performance at smaller scales and the cost savings and freedom that the open source approach allows.

Customers: Over 3 billion social media users provide Meta with an enormous pool of consumers, while some business customers will be won over by Llama's low price and adaptability.

Infrastructure: Meta doesn't run its own B2B cloud, but has plenty of experience scaling up data centers.

Other players

Microsoft has tied its AI fate to OpenAI, but it has also begun to build out its own in-house strategy.

• It's developing its own models in a project led by DeepMind co-founder and former Inflection CEO Mustafa Suleyman.

Amazon has concentrated on meeting the AI boom's enormous demand for cloud services, but it's also investing in its own series of models.

xAI, Elon Musk's venture, raised about $6 billion in the spring and another $5 billion this month — along the way building what it calls the world's largest AI data center at impressive speed.

• But it's not yet clear how xAI intends to compete beyond making vague promises around freedom of speech.

Apple has played catch-up as it works to weave Apple Intelligence into its mobile and desktop operating systems and upgrade its Siri assistant.

Other races

AI's leading competitors are also engaged in an increasingly tough scramble to find more training data for their models.

• Tech giants that have been stockpiling data for years have a natural advantage.
• But data access faces challenges from copyright law, distrust of the technology by creators and the public, and privacy commitments some companies have made.

Zoom out: Most of these companies have also explicitly committed themselves to the quest for artificial general intelligence, or AGI — a level of autonomous intelligence that matches or exceeds human capabilities.

• But since everyone has a different definition of AGI, it may be tough for anyone to lay a claim to achieving it first.

Elon Musk is a wild card in the tech industry's frantic effort to game out where a Trump-dominated Washington will come down on AI regulation.

State of play: It's a reasonably safe bet that President-elect Trump will trash President Biden's modest moves to set limits on AI development and give companies a free hand to do what they want — and beat China.

Yes, but: Musk, who has been at Trump's right hand since his election victory, has two very different personas when it comes to AI regulation, and no one knows which of them will be whispering in Trump's ear.

Musk has been obsessed with AI doomsday scenarios for at least a decade.

• He co-founded OpenAI in 2015 and provided the initial cash for the nonprofit, all in the name of protecting the world from runaway super-intelligence.
• In 2014, he told MIT students that AI needs regulatory controls: "With artificial intelligence we are summoning the demon. In all those stories where there's the guy with the pentagram and the holy water, it's like — yeah, he's sure he can control the demon. Doesn't work out," said Musk.
• In August, Musk came out in support of California legislation that would impose new obligations on developers of advanced AI. The bill passed the state legislature over industry opposition, but got vetoed by Gov. Gavin Newsom.

On the other hand, Musk has also launched a crusade against what he calls "woke" AI.

• He built his new AI startup, xAI, around a commitment to abandoning guardrails against hate speech and misinformation in the name of freedom of speech.
• Musk's feud with OpenAI CEO Sam Altman, rooted in a fight for control over the nonprofit's early development, now centers on Musk's complaint that OpenAI's models have a built-in liberal bias and stifle conservative views.
• Last month, in an interview with Tucker Carlson, Musk said, "I don't trust OpenAI and Sam Altman. I don't think we want to have the most powerful AI in the world controlled by someone who is not trustworthy." Over the weekend, Musk reposted that quotation and video.

Friction point: Advocates of AI caution like Max Tegmark, the AI researcher who leads the Future of Life Institute, are counting on Musk to champion a tight leash on AI in Trump's councils.

• But MAGA's "anti-woke" warriors consider Musk as an ally in their fight to keep AI free of what they see as burdensome guardrails.

Musk's new role as a key Trump adviser means that when Trump takes office on Jan. 20, this contradiction will become a practical dilemma overnight.

Case in point: As Axios reported last week, the Department of Energy is working with Anthropic to test the company's Claude models, making sure they don't give users "help" in building homemade nuclear weapons.

• This sort of research has proceeded under the umbrella of President Biden's AI executive order, which Trump is expected to quickly revoke.
• Nuclear bomb-making — like bioterrorism, financial meltdowns and network sabotage — is exactly the kind of existential AI risk Musk has long warned about.
• But the measures companies are most likely to take today to limit the risk of such outcomes look a lot like what Musk and others on the right denounce as "censorship."

Between the lines: Since Musk is now a participant in the race to build bigger, faster, better AI models as founder and investor in xAI, his stance on AI regulation may become less a matter of philosophy than of self-interest.

• It also means his involvement in AI decision-making would be deeply compromised by conflicts of interest.

Our thought bubble: "AI safety" is in the eye of the beholder.

• "Aligning" AI with human welfare has long been the preoccupation of those, like Musk, who fear super-powered AI could run amok.
• But for AI makers busy deploying error-prone AI tools on a mass scale today, "safety" looks more like blocking an AI image maker from generating revenge porn, preventing a large language model from espousing antisemitism or stopping an AI agent from making racist decisions.

What we're watching: Musk's sway with Trump is high right now. But the president-elect's favors are fickle, and there are already signs that Musk may be overplaying his hand.