Featuring
Spice rack
Was the Fed holding rates because the data demanded patience, or because Powell wanted political leverage over Trump?
Original point: The Fed was ignoring yellow lights in consumer credit and liquidity because a rate cut would help Trump, making its decision as politically motivated as financially motivated.
What everyone argued
Chamath Palihapitiya
Chamath accepts that liquidity matters, then says subprime-lender valuations were warning of a rollover and the Fed was choosing not to get ahead of it. He makes the allegation explicit: Powell was preserving the rate lever because cutting would help Trump.
David Friedberg
Friedberg says mortgage delinquencies were flat, March CPI was still 2.4%, and the Fed needed to see both incoming inflation data and the shape of tariff agreements. A persistent 10% tariff layer could affect prices, growth, employment, and fiscal revenue, so cutting into that uncertainty was premature.
Philippe Laffont
Laffont says weak sentiment coexisted with resilient spending and that a hold could signal a strong economy rather than policy failure. He agrees with Friedberg that the Fed would wait for evidence, while arguing that businesses may absorb part of tariffs rather than pass every dollar to consumers.
Winner circle
Friedberg wins, with Laffont alongside him. They explain the hold using observable inputs that the Fed itself documented, while Chamath moves from 'I would cut' to 'Powell is political' without bridging the evidentiary gap. Later cuts as employment risks rose—and another pause with inflation still elevated—fit their data-dependent model.
Commentary
Chamath Palihapitiya
Assumptions and fact checks
Political hostility was the only plausible reason for the Fed not to cut in May 2025.
Why it mattersInflation was still above target, the labor market was solid, tariff policy was unsettled, and the Fed's minutes document serious two-sided risks. A different preferred weighting of those risks does not establish political motive.
Subprime-equity valuation signals justified an immediate rate cut.
Why it mattersCredit stress can lead the cycle, but one relative-valuation chart is not enough to override inflation, employment, expectations, and financial-conditions data. Chamath never identified the series precisely enough to test its historical signal.
The May 7 FOMC press release used 'waiting' or a synonym 22 times to justify not cutting.
CheckThe official statement is brief and does not repeat 'waiting' 22 times. It says the Committee will assess incoming data, the evolving outlook, and the balance of risks, while noting that risks of both higher unemployment and higher inflation had risen.
The May 2025 U.S.-UK deal eliminated Britain's 2% Digital Services Tax.
CheckThe UK government's announcement said the Digital Services Tax remained unchanged; the countries agreed only to continue work on digital trade.
David Friedberg
Friedberg does the unglamorous central-bank work: name the variables, admit the sign of their effects is uncertain, and wait for observations. His weakest sentence turns tariff receipts into growth too mechanically, but it does not damage his core answer.
Assumptions and fact checks
Waiting for tariff and inflation data was worth the risk of reacting late to credit weakness.
Why it mattersWith both mandates at risk and the tariff regime changing weekly, preserving optionality was defensible. Later cuts show waiting did not mean refusing to respond.
Headline CPI inflation was 2.4% in March 2025.
CheckBLS reported a 2.4% increase in CPI-U over the 12 months ending March 2025; core CPI was 2.8%.
The May 2025 Fed record showed policymakers were weighing both higher unemployment and higher inflation risks.
CheckThe statement says both risks had risen, and the minutes describe solid labor conditions alongside tariff-related inflation risks and uncertainty about later weakening.
Philippe Laffont
Laffont earns credit for saying where his knowledge stops. His best contribution is not a rate forecast but a warning that scary surveys and actual spending can tell different stories.
Assumptions and fact checks
Consumer sentiment was more lagging market performance than forecasting a recession.
Why it mattersThe economy did not enter the immediate crisis feared in spring 2025, and activity remained solid into 2026. Sentiment still contains information, but this episode's divergence was real.
Roughly half of tariffs would be absorbed rather than passed into retail prices.
Why it mattersPass-through varies by product, contract, inventory, exchange rate, margins, and time horizon. One unnamed retailer's plan cannot support a national percentage.
Did Google need to cannibalize Search aggressively, or could it integrate Gemini without detonating the ad business?
Original point: Google had the models, users, and products to manage a gradual transition; it did not need to replace profitable Search with a costly chat interface overnight.
What everyone argued
Chamath Palihapitiya
Chamath says Google should assume a severe share loss, red-team the death case, and make Gemini the front door before competitors dictate the timetable. Spending tens of billions on AI while stage-gating the product is, in his view, the worst of both worlds.
Jason Calacanis
Jason is bullish on Google's distribution and suggests targeted Gemini experiences in YouTube, Calendar, Gmail, and Workspace. He argues that richer intent across those surfaces could create a better advertising product without sacrificing core Search queries too quickly.
David Friedberg
Friedberg says the old search-click-repeat pattern will change, but Google owns competitive models, huge distribution, and a culture of experiments. Because chat queries cost more and Search funds the company, the rational move is gradual integration followed by larger changes once tested.
Philippe Laffont
Laffont treats Google as a classic innovator's dilemma: Search supplies nearly all economic value while Gemini threatens to cannibalize it. He refuses a simple buy-or-sell answer, argues against harvesting the cash cow, and says founders may have the credibility to impose the painful redesign.
Winner circle
Friedberg wins the central strategy question. Google ultimately moved fast in product terms while preserving an incremental deployment model, and the first year of hindsight shows expanding usage and monetization rather than a collapsing franchise. Chamath deserves credit for demanding urgency, but his 99%-to-75% emergency scenario was not the path that materialized.
Commentary
Chamath Palihapitiya
Chamath correctly demands a death-case plan, then treats the death case as the base case. That is useful boardroom pressure, but not yet a probability-weighted strategy.
Assumptions and fact checks
Waiting for internal data would leave Google structurally unable to respond to an external product shock.
Why it mattersCompetitor surprises are real, but distribution, telemetry, cash, and model quality give Google unusually strong adaptation tools. The 2025 rollout shows internal testing and fast shipping were compatible.
Google had effectively 99% of general-search share before the AI shift.
CheckThe DOJ describes Google's U.S. query share as approximately 90%, not 99%. Google was plainly dominant, but the extra nine points materially exaggerate the starting position.
Alphabet expected about $75 billion of capital expenditures in 2025.
CheckAlphabet's February 2025 earnings call gave approximately $75 billion as its 2025 CapEx expectation, primarily for technical infrastructure.
Jason Calacanis
Jason's YouTube and Workspace examples turn a vague 'use AI' demand into product surfaces with users, context, and a migration path. That specificity is his edge.
Assumptions and fact checks
Google could monetize broader AI intent at least as effectively as traditional keyword search.
Why it mattersBy late 2025 Alphabet said Gemini improved query understanding and monetization of longer, more complex searches. Agentic ads remain early, so superiority across all use cases is not settled.
Google Search was about 56% of Alphabet's 2024 revenue.
CheckAlphabet reported $198.1 billion of Google Search & other revenue on $350.0 billion of total 2024 revenue, or about 56.6%.
David Friedberg
Friedberg wins by changing the denominator. If AI creates far more queries and tasks, losing some old-style share can coexist with a larger business; Alphabet's later numbers look remarkably like that thesis.
Assumptions and fact checks
Google could discover the transition path through incremental tests without surrendering the new category.
Why it mattersThrough late 2025 the company expanded both AI usage and Search economics. Longer-run competitive share remains open, but the strategy cleared its first major test.
Google Search generated roughly $200 billion of annual revenue at the time.
CheckGoogle Search & other produced $198.1 billion in 2024, making $200 billion a fair rounding.
Philippe Laffont
Laffont is most useful when he declines false certainty. He sees the incentive trap clearly, but his disruption analogies understate the option value of already owning Gemini.
Assumptions and fact checks
Only founders would have enough moral authority to force a self-cannibalizing transition at Alphabet.
Why it mattersFounder support can overcome internal resistance, but the claim is organizational and counterfactual. Alphabet's disclosed execution does not isolate whether founder authority was necessary.
Should antitrust put a market-cap ceiling on Big Tech acquisitions?
Original point: Let sub-$750 billion or sub-$1 trillion companies merge freely while applying much tighter limits to the largest platforms, whose distribution lets them bundle acquired products and crush standalone rivals.
What everyone argued
Jason Calacanis
Jason proposes a bright-line size rule to unleash a 'max 70' below the Mag 7 while restraining dominant platforms. His mechanism is cross-subsidy: Apple or Google could buy a paid product, make it free through a bundle, eliminate rivals, and exploit the resulting market later.
David Friedberg
Friedberg rejects a size cutoff and says antitrust should ask whether the specific deal harms consumers, choice, and market competition. A large platform buying an unrelated company should not be blocked merely because the acquirer is already valuable; the same conduct risk can arise inside a smaller market.
Philippe Laffont
Laffont backs Friedberg's case against capping corporate success and argues that large-company acquisitions provide the upside that finances risky startup bets. Existing law should punish actual anticompetitive conduct, while orthogonal growth and holding-company expansion should remain available.
Winner circle
Friedberg wins. Jason's bundling concern is legitimate, but Friedberg shows why it should trigger deeper review rather than an automatic ban: the same conduct can harm a narrow market regardless of the acquirer's total valuation. Philippe supports the correct no-cap conclusion, though he gives too little weight to nascent-competition and platform foreclosure risks.
Commentary
Jason Calacanis
Jason finds the right disease and prescribes the wrong screening test. His own Coinbase-Robinhood-E*Trade example shows why market definition beats a trillion-dollar velvet rope.
Assumptions and fact checks
Total market capitalization is a useful proxy for the risk that an acquisition will suppress competition.
Why it mattersA giant firm's orthogonal purchase may be harmless while a much smaller firm's horizontal acquisition can eliminate its closest rival. Market cap is easy to measure but poorly matched to the statutory question.
A dominant company making an acquired product free is automatically illegal price dumping.
CheckLow or even below-cost pricing is not automatically an antitrust violation. Federal predatory-pricing doctrine generally requires below-cost pricing plus a dangerous probability that the firm can later recoup losses through monopoly pricing.
David Friedberg
Friedberg has the best rule: investigate the mechanism in the relevant market. He only overreaches when 'market cap is not the rule' becomes 'scale is irrelevant.'
Assumptions and fact checks
Scale should never affect merger scrutiny if the acquisition appears orthogonal.
Why it mattersExisting dominance, data, defaults, tying, and ecosystem control can turn an apparently adjacent deal into a foreclosure risk. Scale is not a verdict, but it can be material evidence.
U.S. merger review asks whether a transaction may substantially lessen competition or tend to create a monopoly.
CheckThat is the Section 7 Clayton Act standard summarized by the FTC, applied through a fact-specific review of likely competitive effects.
Philippe Laffont
Laffont is right that 'large' is not an antitrust theory. He is less convincing when investor incentives quietly become the purpose of antitrust rather than one consequence the policy should consider.
Assumptions and fact checks
Looser acquisition review would materially increase early-stage investment by restoring power-law exits.
Why it mattersExit options affect expected returns, but startup funding also depends on valuations, rates, IPO markets, technology cycles, and fund overhang. The magnitude is not established here.
Was venture capital's exit drought a healthy market reset or an artificial regulatory strangulation?
Original point: Regulation and blocked exits were starving the risk-capital flywheel at exactly the moment AI created extraordinary opportunities, threatening long-run American innovation.
What everyone argued
Chamath Palihapitiya
Chamath says venture needs mid-to-high-20% net returns to compensate for long lockups, but recent funds look more like three- or four-year hedge funds. He attributes much of that gap to policy-created illiquidity and estimates that lighter regulation could add 500 to 1,000 basis points of return.
Jason Calacanis
Jason documents the post-2021 collapse in IPOs and acquisitions, then largely joins the regulatory account: boards stopped considering deals because enforcement risk made exits feel futile, harming founders, employees, endowments, and pension beneficiaries downstream.
David Friedberg
Friedberg asks whether exit volume should discipline how much capital LPs allocate to venture. If the economy only produces a certain number of viable outcomes, lower fundraising and lower entry prices may be market normalization rather than policy failure.
Philippe Laffont
Laffont says scarce M&A makes early-stage investing less attractive because acquisitions provide essential singles and doubles around rare home runs. He views low exits as historically abnormal and argues that countries without a strong risk-capital compact stagnate.
Winner circle
Friedberg wins a low-confidence ruling. His normalization thesis explains more of the cycle with fewer unsupported causal leaps, while Chamath and Jason correctly identify a genuine liquidity feedback loop but over-assign it to regulators. Laffont makes the strongest pro-exit incentive case, yet none of them quantifies how much enforcement, rather than prices and rates, caused the drought.
Commentary
Chamath Palihapitiya
Chamath is persuasive about the plumbing and speculative about the culprit. His return ladder makes the liquidity premium tangible; the 500-to-1,000-basis-point counterfactual is precision without a model.
Assumptions and fact checks
Removing recent regulatory constraints would raise venture returns by 5 to 10 percentage points.
Why it mattersNo model or counterfactual supports that range. Exit policy matters, but rate changes, entry valuations, company performance, IPO appetite, and fund selection also drive returns.
A prolonged shortage of exits will reduce new risk capital and future company formation.
Why it mattersDistributions influence LP re-ups, and a sustained denominator and liquidity problem can shrink fundraising. The link is credible even when regulation is not the sole cause.
Jason Calacanis
Jason's chart proves the drought, not the rainmaker. His best move is tracing distributions to LPs; his worst is converting frustration with Lina Khan into a complete causal model.
Assumptions and fact checks
FTC posture was the main reason corporate boards stopped pursuing startup acquisitions.
Why it mattersEnforcement risk can chill marginal deals, but acquisition demand also fell with financing conditions and valuation resets. Anonymous adviser reports cannot allocate the causes.
U.S. venture exits spiked in 2021 and public-listing activity remained far below that level through early 2026.
CheckPitchBook-NVCA's Q1 2026 data show the exceptional 2021 peak and only a measured recovery; 15 VC-backed IPOs in Q1 2026 remained insufficient to clear the backlog.
David Friedberg
Friedberg wins by asking the uncomfortable denominator question: perhaps the asset class raised too much at the wrong prices. His account is incomplete, but it needs fewer heroic assumptions than the claim that regulators erased ten points of return.
Assumptions and fact checks
Lower venture fundraising and valuations can restore prospective returns after an overfunded cycle.
Why it mattersPaying lower entry prices and reducing marginal capital are standard adjustment mechanisms. They do not guarantee exits, but they improve the arithmetic for future vintages.
Observed exit volume reflects the economy's natural innovation rate.
Why it mattersExit volume also reflects interest rates, public-market risk appetite, regulation, and temporary windows. It is a market signal, not a pure measure of underlying innovation.
Philippe Laffont
Laffont explains why exit optionality matters without pretending every startup becomes OpenAI. He still treats all additional M&A as fuel, when some acquisitions can also remove the challengers that create the next cycle.
Assumptions and fact checks
More permissive M&A would restore enough singles and doubles to justify more early-stage risk.
Why it mattersMore credible exits would raise expected value, but the benefit depends on acquisition prices, competition effects, and whether funding had already exceeded the supply of strong companies.
Federal agencies challenged only a small minority of reportable mergers in fiscal 2023.
CheckFTC and DOJ reported 1,805 HSR transactions and 28 merger enforcement actions in fiscal 2023. That does not measure informal deterrence, but it contradicts a literal picture in which large-company M&A was generally prohibited.
Chamath has a legitimate early-cut case and buries it under an accusation he cannot prove. The tell is 'the only reason I can come up with': failure to name another explanation is not evidence that no other explanation exists.