Trump's First Week: Inauguration Recap, Executive Actions, TikTok, Stargate + Sacks is Back!
Summary
Trump’s opening-week signal was a business-first reset that places technology, energy and private capital at the center of national strategy. Chamath read the inauguration’s business-heavy guest list—leaders from Meta, Google, Apple and others alongside international business figures—as “Team America,” while Thomas Laffont highlighted Scott Bessent’s sharper framing: “We’re not in a green-energy race with China; we’re in an energy race with China.” The political mandate may be broad, but Friedberg warned that congressional incentives still favor projects and jobs over deficit reduction.
DOGE’s engineering model is clear, but its political path is not: each agency gets a four-person team while legislators remain rewarded for bringing spending home. Friedberg’s best evidence was Mitch McConnell pressing the agriculture-secretary nominee about a delayed $60 million Kentucky research lab—the opposite of voting programs away. His bleak assessment: “There’s no stop in the train,” while DOGE’s last-minute move inside government changes its institutional setup.
TikTok’s US franchise plausibly supports a roughly $100 billion valuation, with a much higher ceiling if it monetizes like Meta. Laffont estimated about 100 million US DAUs versus roughly 200 million across Facebook and Instagram, yet comparable aggregate time spent; starting from Meta’s $1.5 trillion market cap produced a theoretical $375 billion US ceiling before regulatory, algorithm and execution discounts. The product’s original insight was equally important: every uploaded item gets shown to at least one person, so “your content will be shown” even without followers.
Trump’s demand that America receive 50% of TikTok prompted a wider proposal: taxpayers should share the upside whenever government grants scarce permits, land or subsidies. Chamath invoked Bernard Arnault’s one-franc acquisition of the Boussac assets containing Christian Dior and argued that a retained French stake could now be worth $175 billion; he would likewise have accepted 5%-10% government equity alongside a company he helped start’s $150 million battery-materials grant. Laffont’s pushback was fundamental: ownership risks government “picking winners,” while taxes and open auctions preserve a level playing field.
Stargate’s announced $500 billion over four years is financeable facility by facility, but the real underwriting question is whether the resulting compute earns an adequate return. Laffont’s OpenAI bull case rests on ChatGPT’s claimed 300 million weekly actives, more than one million enterprise users and 80%+ share versus Gemini and Grok—not on whether SoftBank wires $500 billion upfront. Chamath countered with a Chinese, MIT-licensed model built for millions, runnable on a laptop and, in his account, competitive with OpenAI’s o1—evidence that “spending more money doesn’t necessarily get you further along.”
Electricity, not GPUs, may be the binding constraint on American AI: the US pays roughly 1.5x-3x China’s power price and has about half its generation capacity. Thomas’s chart put the countries near 1,000 terawatt-hours each in 2000 but roughly 1,600 versus 9,000 today; Laffont argued that even export controls become irrelevant if China eventually has 10x the power and can deploy vastly more, slightly older chips. Their categorical conclusion was that America cannot catch up without nuclear, and Friedberg called regulatory delay a larger national-security threat than the border.
Trump’s first actions exposed the tension between political reconciliation and equal justice while giving David Sacks formal mandates in crypto, AI and science policy. Friedberg, from a law-enforcement family, said pardoning roughly 1,500 January 6 participants—including violent offenders—“betrayed the blue”; Jason called January 6 “a stain” but argued that unequal prosecutions and excessive sentences made a more granular review necessary, while Friedberg questioned the pardon power itself. Sacks’s new working groups seek US crypto leadership, an AI action plan aimed at global “dominance,” less ideological model bias and a PCAST committed to “truth-telling in science.”
Deep dive
1. Trump’s inauguration put business back on the national stage
Laffont’s opening frame was institutional rather than partisan: watching a democracy reverse direction showed “democracy in action, democracy self-correcting,” whereas “dictatorships double down.” That capacity to change course, he argued, explains America’s durability.
Chamath saw unusually explicit coordination between government and the entrepreneurial sector. With leaders from Meta, Google, Apple, Amazon and Tesla joined by figures such as Bernard Arnault and Mukesh Ambani, the message was that America would stop ostracizing business and start “firing on all cylinders.”
The absence of favoritism mattered to Chamath almost as much as the guest list. Elon Musk’s early support did not exclude Mark Zuckerberg or other technology leaders: “This is Team America,” a display of economic power to the rest of the world.
Laffont’s sharpest policy takeaway came from Scott Bessent: “We’re not in a green-energy race with China; we’re in an energy race with China.” China was adding roughly 100 coal plants, nuclear capacity and, according to the discussion, a 137-gigawatt hydro project.
2. DOGE runs directly against Congress’s survival incentives
DOGE was established inside government rather than as the expected outside body. The plan assigns a lead to every federal agency, with each lead assembling a four-person team including an engineer, an HR specialist and an attorney.
Friedberg found little appetite for genuine austerity among officials he met. Politicians win reelection by delivering money, facilities and jobs to constituents; cutting programs can therefore resemble “voting themselves out of a job.”
His concrete specimen was Mitch McConnell asking agriculture-secretary nominee Brooke Rollins what she would do about a $60 million University of Kentucky research lab that had not begun construction after four years. That is the political machinery DOGE must reverse.
Chamath agreed that domestic resource allocators face much harder confirmations because every senator wants a share. He nevertheless flagged DOGE’s last-minute placement inside government as a consequential change to its institutional setup.
3. Executive-order velocity is colliding with performative oversight
Trump signed 26 first-day executive orders, versus the prior day-one record of nine attributed to Biden. Measures addressed DOGE, a 75-day TikTok reprieve, January 6 pardons, birthright citizenship, energy, return-to-office rules, hiring and regulatory freezes.
Jason called confirmation hearings “hugely performative.” When OMB nominee Russell Vought tried to describe waste, fraud and abuse in Medicare and Medicaid, Sanders appeared more interested in a sound bite than the operational answer.
Jason’s governing principle was straightforward: a president who openly campaigned on an agenda should generally get the team needed to execute it, absent compromise risk or something serious in a nominee’s background. The administration then owns the results.
Davos supplied the losing side’s diagnosis. Jason recalled a panel, including Graham Allison, conceding in effect, “We lost, they won”: an elite view of the future had been rejected. Laffont separately judged the conference “a counter-indicator.”
4. TikTok’s algorithm created distribution before it created followers
Coatue invested in ByteDance when China exposure meant something different. What first impressed Laffont was the founder’s simple question: why should excellent content remain invisible merely because its creator lacks an existing social graph?
TikTok’s answer was to show every upload to at least one user, then expand distribution according to viewing behavior and duration. “Your content will be shown, and if people like it, it’ll be shown to a lot of people.”
Laffont estimated roughly 100 million TikTok US DAUs, about half the combined Facebook-and-Instagram audience, but with equivalent aggregate time spent. That makes engagement—not merely user count—the franchise’s most valuable asset.
His valuation bridge began with Meta near $1.5 trillion, assumed half represented the US, then halved that $750 billion for TikTok’s smaller audience to reach $375 billion. After return requirements and discounts for uncertain access to users, data and the algorithm, roughly $100 billion remained “a reasonable scenario”; $1 trillion was not impossible over time.
5. Chinese consumer technology became a national-security asset class
Coatue’s historical China investments largely backed Chinese companies serving Chinese demand: Huawei in smartphones, Meituan in delivery and restaurant software, Tencent in gaming and ByteDance in content. Tencent was once public in Hong Kong at a market capitalization around $200 billion.
The opportunity ran across public and private markets for roughly two decades, but technology’s conversion into a national-security issue changed the permissible landscape. Coatue’s response was not to set policy itself: “We rely on government and regulators to do so.”
Laffont’s regret was the lost entrepreneurial cohort. He singled out Meituan founder Wang Xing and Tencent’s Pony Ma as unusually strong builders; after Jack Ma’s disappearance, Jason’s interpretation was that ambitious founders received an unmistakable message: “Don’t do this.”
Laffont would not pretend to know the Chinese government’s internal rationale. For an investor, the regime was “a black box”: Coatue could evaluate founders and businesses, but not reliably determine whether the pendulum would swing back toward innovation.
6. Chamath wants taxpayers to own part of government-created upside
Chamath compared TikTok with Bernard Arnault’s 1984 purchase of the failing Boussac empire, including Christian Dior, for one franc. After four decades of execution produced an approximately $350 billion LVMH, a hypothetical retained 50% French interest would be worth $175 billion.
Applied to TikTok, Trump controls a permission without which the US asset may be worthless. Chamath therefore expected any transaction price to sit far below intrinsic value—perhaps $25 billion or $10 billion, not necessarily one dollar—while the Treasury retained 50% of the upside.
His broader template would exchange federal land, accelerated permits or other economic concessions for 5%-10% royalties or equity. A battery-materials company he helped start received a $150 million Energy Department grant; he said he would readily have surrendered 5%-10% equity.
Tesla supplied the missed-opportunity analogy: it received $465 million from the Department of Energy through the ATVM program, but taxpayers received no equity. Chamath argued that government should stop distributing capital “with no consequence and no accounting” when modest participation would not change private underwriting.
7. Public ownership could distort competition and invite retaliation
Laffont’s objection was not that taxpayers deserve no return; it was that government ownership could disadvantage other companies. A 50% interest in TikTok might disadvantage Meta, while ordinary taxation captures value without selecting a favored company.
Chamath conceded that open competition is essential and proposed public RFPs available to every bidder. Once a winner receives a scarce public benefit, however, he still saw no reason taxpayers should forgo “a small stake in the upside”; Laffont compared alternative mechanisms with spectrum auctions.
Friedberg worried most about precedent abroad. If Washington can demand half of TikTok for security reasons, Beijing could demand half of Tesla’s Chinese operation, while other governments could target Google, Apple or American food manufacturers under similarly broad rationales.
Jason answered that reciprocity already exists: when Apple introduced iCloud in China, Guizhou-Cloud Big Data told Apple that it would get all the data, and Apple had to comply. His narrower claim was pragmatic—when government inevitably intervenes, it should retain a stake if that intervention creates durable private value.
8. The January 6 pardons fractured the panel’s reconciliation story
Trump pardoned roughly 1,500 January 6 participants despite JD Vance previously saying violent offenders “obviously shouldn’t be pardoned.” Friedberg, from a law-enforcement family, said the treatment “betrayed the blue” and argued that individual cases should have received granular review.
Friedberg accepted that individual sentences may have been excessive, but that was precisely why he wanted a careful process. Elevating violent Proud Boys or Oath Keepers as heroes could validate future violence; Jason cited Enrique Tarrio’s warning that perceived enemies “need to pay for what they did.”
Jason called January 6 “a stain” and agreed that methodical review would have been better. His contextual defense was accumulated unequal enforcement—from BLM and Antifa riots to decarceration and violent undocumented offenders—creating the belief that identical conduct drew different punishment according to politics.
Jason also cited a 6-3 Supreme Court ruling that, in his account, cast doubt on at least 350 convictions, plus misdemeanors elevated into felonies and sentences that did not match crimes. His preferred settlement was to close the chapter, then “apply the law equally to everybody” going forward.
9. Birthright citizenship and pardons now face constitutional stress tests
Jason framed the birthright order around the Fourteenth Amendment’s phrase “subject to the jurisdiction” of the United States. With 22 state attorneys general already suing, he expected the Supreme Court to decide quickly whether that qualifier limits citizenship for children of undocumented immigrants or visa holders.
Jason said that 6,500 of the 8,000 people Biden pardoned appeared to have been convicted of marijuana possession, while Friedberg argued more broadly that Biden and Trump had used pardon power politically. Friedberg argued that legislators and courts should change unjust law instead of presidents repeatedly superseding them.
Yet Federalist No. 74 complicated his own objection. Hamilton defended pardons both as mercy for excessive punishment and as a way, during insurrection, to “restore the tranquility of the commonwealth”—meaning a pardon can serve national reconciliation even when the outcome does not look individually just.
10. OpenAI’s consumer franchise is the strongest case for Stargate
Stargate was announced as a new company intending to invest $500 billion over four years in US AI infrastructure. SoftBank and OpenAI were the lead partners, with Oracle and MGX participating and Oracle, NVIDIA and OpenAI expected to build and operate infrastructure.
Laffont disclosed that Coatue owns some OpenAI while rejecting venture-style tribalism: he can be pro-OpenAI, pro-Tesla and pro-X simultaneously. His thesis separates the model, API and ChatGPT businesses rather than treating OpenAI as one undifferentiated bet.
ChatGPT was the centerpiece: roughly 300 million weekly active users, more than one million enterprise users and, by his cited download charts, greater than 80% share against Gemini and Grok. It had also become indispensable to his own phone-and-desktop workflow.
Jason’s pushback—worth keeping—was that digital AI competition is unlike ride-sharing. Gemini, Grok, ChatGPT and Claude already felt close in quality, Google, Microsoft and Elon Musk were simultaneously challenging the incumbent, and enterprise buyers often preferred open-source systems.
11. Stargate’s capital structure works only if compute earns a return
Laffont dismissed the idea that SoftBank must supply $500 billion in equity upfront. Infrastructure can be financed data center by data center and GPU by GPU, combining sponsor equity, site-level debt and other capital over four years.
His admiration for Masayoshi Son informed that confidence. Laffont credited Masa with a SoftBank-Alibaba win that peaked around $200 billion and, he thought, netted more than $80 billion, plus buying Arm for about $30 billion and later sitting on roughly $135 billion of gains.
Scale alone was not disqualifying: the top five US players were expected to spend $312 billion in 2025, while American internet companies collectively spent about 20x their Chinese counterparts on capex. Stargate’s annualized ambition was therefore at least in the same order of magnitude.
The underwriting question remained: “Do we ultimately believe that you can get an ROI on that $500 billion?” If the market and revenue opportunity justify the facilities, Laffont expected capital to appear; if not, announcements and prestigious sponsors cannot rescue the economics.
12. A Chinese model broke the assumed link between AI spending and progress
Chamath’s contrary evidence was a Chinese model released under an MIT license, runnable on a laptop and, by his assessment, as capable as OpenAI’s o1. He said it cost millions to develop versus billions for the model it challenged.
Unlike oil fields or rockets, AI costs cannot be projected from stable physical inputs. Advances can make an equivalent task cheaper “almost overnight,” sometimes at one-thousandth of the cost associated with an approach only six months earlier.
That made Stargate’s $500 billion sound more like “marketing and sizzle” than a technical requirement. Chamath did not doubt SoftBank, Oracle or OpenAI as companies; he doubted that spending could be directly linked to custom vaccines, cancer cures or whatever model follows o3, o4 or o5.
Jason read the number more charitably as Trump’s moonshot style: Masa reportedly offered $100 billion, Trump pushed for $200 billion and the announcement reached $500 billion. Even $50 billion-$100 billion of realized investment would still increase US capability, though it would fall far short of the headline.
13. America’s AI race is becoming an electricity race
Friedberg estimated US electricity at roughly 1.5x-3x China’s price, with roughly half China’s electricity-production capacity. China could add new capacity for between one-tenth and one-half of the comparable US cost.
Laffont’s chart showed both countries around 1,000 terawatt-hours in 2000, followed by the US reaching roughly 1,600 and China roughly 9,000. He connected the divergence to offshoring: fewer American factories meant less reason to expand the grid.
AI reverses that logic because the new factories are GPU and data-center facilities. Laffont’s chain was explicit: “You can’t power a GPU without a data center, and you can’t use a data center without power.”
The consumer constraint matters too. Adding AI demand cannot be perceived as raising household electricity bills during an inflationary period; supply must expand fast enough that industrial growth does not create a political backlash through utility prices.
14. Nuclear is the proposed bridge, but America has forgotten how to build
Asked whether the US could catch up without nuclear, Laffont answered simply: “No.” Constellation Energy illustrated the emerging market structure as Amazon, Microsoft and federal agencies contracted directly for behind-the-meter or facility power.
US nuclear capacity has been essentially flat for more than 25 years while China’s climbed rapidly. France provided the counterexample inside Europe: it “held the line” as Germany retreated, and more than 70% of French electricity still came from nuclear.
Friedberg argued that Generation IV designs differ fundamentally from the architectures associated with Chernobyl or Three Mile Island. He highlighted meltdown-proof systems such as pebble-bed reactors already generating electricity in China while the US deployed none.
His conclusion was urgent: capital, physics, talent and technology exist, but regulation blocks deployment. He called energy a greater national-security threat than the border and suggested emergency action may be necessary. Laffont’s caveat was that decades without construction also mean less practice building safely and quickly.
15. Netflix’s success became a proxy debate about disengagement
Chamath overlaid Netflix’s rising share price with increased SSRI use and falling labor-force participation, while noting fertility had also declined. He expressly denied causation, but saw a “dark underbelly”: entertainment platforms perform best when more time is available for distraction and withdrawal.
Laffont challenged that reading with substitution: measured television time had not obviously risen over 20 years, while traditional media companies such as Fox, Comcast and CBS had lost ground. Netflix might be replacing linear viewing rather than creating more passivity.
Laffont’s personalization argument was that algorithmic products “hypertune” what people think they want, producing weekend-long binge loops at the expense of exercise, sleep, food, friends and family.
Jason generalized the mechanism to TikTok, YouTube, music and MrBeast: the damage is less the screen itself than what it replaces. Friedberg suggested scheduled real-world rituals, while Jason emphasized limiting screen time and focusing on “sleep hygiene, diet, exercise, meditation.”
16. Broadcom shows how disciplined M&A can compound without hype
Laffont called Broadcom the trillion-dollar company “nobody talks about.” Hock Tan built it through rollups beginning with Avago, itself a “double orphan” spun from a Hewlett-Packard spinout, rather than relying on one breakthrough product.
When Avago acquired the original Broadcom, founders Henry Nicholas and Henry Samueli valued the name enough to concede price. Hock accepted the discount but retained ticker AVGO because, as he later told Laffont, “They forgot to negotiate for the ticker.”
Broadcom had grown roughly 8x over five years, versus NVIDIA around 20x and Intel down approximately 70%. Thomas attributed Intel’s failure to “a complete abdication of corporate governance at the board level and the CEO level.”
17. Sacks returned with formal mandates across crypto, AI and science
Joining immediately after the Oval Office signing, Sacks said three orders covered crypto, AI and PCAST. The crypto order created an internal group chaired by Sacks, with Treasury, the SEC and other relevant agencies, to recommend how America becomes “the world capital of crypto.”
Chamath’s priority within that mandate was stablecoin infrastructure: instantaneous, effectively costless payments could accelerate GDP and reduce fraud. He was less committed to the proposed government crypto stockpile than to modernizing transactional rails.
The AI order added detail to the rescission of Biden’s roughly 100-page framework and directed an action plan led by Mike Waltz, Michael Kratsios after Senate confirmation and Sacks. Its stated goal was global AI “dominance,” including models that are as politically unbiased as possible.
Sacks and Kratsios would also co-chair PCAST, assembling science and technology advisers around “truth-telling in science.” Sacks cautioned that ethics onboarding still limited him largely to listening rather than shaping policy, but the signed orders established the responsibilities he would assume.