Fed Hesitates on Tariffs, The New Mag 7, Death of VC, Google's Value in a Post-Search World

Philippe views the Fed’s decision not to cut as consistent with a solid real economy, noting strong consumer spend despite weak sentiment. Chamath agrees the Fed is watching liquidity, not stock prices, but argues key subprime and credit indicators are ‘blinking yellow,’ and that Powell may be delaying cuts to avoid appearing to help Trump politically rather than reacting purely to data.

Friedberg highlights the UK trade deal’s 10% tariff on imports as a template: sustained tariff income could fund tax cuts, impact GDP and employment, and complicate the Fed’s incentive to cut rates into what might become a structurally different economy. Philippe adds that retailers expect only ~50% of tariffs to pass to consumers, muting the inflation impact while still raising government revenue.

Laffont frames ‘tokens >> tariffs’: Microsoft processed ~100 trillion tokens in Q1 (50T in March alone), and there is a broad chip and compute shortage across his portfolio. He argues that initial market anxiety about an ‘AI bubble’ is giving way to evidence of real usage and CapEx, suggesting AI’s growth can outweigh damage from tariffs over the next 1–2 years.

Eddy Cue’s revelation that Apple’s search volume fell for the first time in 20 years—because of ChatGPT and Perplexity—triggered a $100B wobble in Google’s market cap. The group agrees Google’s Gemini models are strong, but Chamath warns that losing even a modest share from a ‘99%’ dominance, while under-investing in front-end Gemini integration, risks ceding the new AI interaction layer to OpenAI and others. They argue Google must aggressively front-load Gemini into properties like YouTube, Workspace, and Gmail, ideally with founder-led ‘taste’ and courage.

Philippe notes that markets moved from ‘FANG’ to ‘Mag 7,’ but rapid AI-driven change means those baskets are outdated. He informally targets a list of ~25 companies—some public, some private like SpaceX, Stripe, OpenAI—as the true future leaders. The right way to own them is via vehicles that can span public and private markets rather than being forced into public-only, index-heavy allocations.

The hosts show post‑2021 IPO and M&A data flatlining and argue Lina Khan’s FTC has scared boards away from even discussing acquisitions, removing ‘singles and doubles’ that traditionally returned capital. Chamath contends that when LPs can’t get liquidity via IPOs or M&A, they rationally stop funding VC, choking off risk capital exactly when AI requires it most. Philippe adds that capping big-company upside or blocking M&A also kills the power-law lottery that motivates founders, VCs, and the institutional LPs (pensions, endowments, foundations) behind them.

Laffont describes launching an ‘interval fund’ seeded with $1B from the Bezos and Dell family offices, targeting a Berkshire-like mix: roughly thirds in cash, public equities, and private investments over time. The fund charges lower fees (around 1.25% and 12% carry) in exchange for quasi-permanent capital, allows periodic redemptions instead of 10-year lockups, and is accessible at much lower minimums (around $50K), aiming to democratize access to the future ‘Mag 25’ while giving him flexibility to own both public comps and private leaders.