Question 1

DOGE is applying a startup-style zero-based budget to government

Accepted Answer

Antonio Gracias and David Friedberg frame DOGE as a federal-scale zero-based budgeting exercise: instead of starting from last year’s spending and adding more, every line item is reset to zero and must be re‑justified. This mirrors the early Twitter/X turnaround where they froze payments, turned off credit cards, and waited to see who complained—often exposing the worst grifters and unused software/services. The implicit playbook for founders and operators: freeze spend, audit every contract, and rebuild budgets from first principles to surface fraud, waste, and low-ROI activity.

Question 2

USAID’s media and NGO funding raises perceived corruption and ‘astroturfing’ risks

Accepted Answer

Data emerging from USAID shows sharp spending increases to outlets like Politico (~$8M/year under Biden vs. ~$1.3M under Trump), the BBC (~$2.7M, ~8% of income), and Thomson Reuters’ consulting arm (~$120M since 2011, half under Biden). Sacks and Chamath argue this looks like modern ‘payola’—top‑down funding of media and political opposition groups abroad that masquerades as grassroots sentiment. For policymakers and journalists, the actionable implication is to demand disclosure rules and independent audits whenever government money flows to media or political NGOs.

Question 3

Crypto is shifting from regulatory hostility to a bipartisan legislative framework

Accepted Answer

Sacks outlines a new legislative push involving four key committees (House/Senate Banking and Agriculture) to pass two major bills: (1) a stablecoin bill first, then (2) a market structure bill (akin to an updated FIT 21) defining when a token is a security vs. a commodity, how decentralization changes that status, and what disclosures are required. He contrasts this with Gary Gensler’s SEC, which he says “honeypotted” founders—inviting them in, giving no clarity, then issuing Wells notices. Founders and investors should prepare for: clearer onshore rules, more activity moving back to the U.S., and a sharper distinction between regulated projects and offshore “Bahamas-style” scams.

Question 4

A U.S. sovereign wealth fund could double as stealth industrial policy

Accepted Answer

Trump’s EO to explore a U.S. sovereign wealth fund (possibly seeded with a 50% TikTok US stake and seized assets) sparks a debate. Chamath wants a small, rotating group of billionaire ‘elder statesmen’ investors (e.g., Tepper, Druckenmiller, Griffin) running it, investing into American companies and strategic sectors. Antonio emphasizes it as a way to finally give the U.S. a real industrial policy, like China’s state-directed chip and manufacturing push. Friedberg counters that government is bad at capitalism and argues the fund should focus on smarter monetization of government-owned assets (TikTok equity, seized Bitcoin, land), and potentially better management of Social Security funds—returning cash to Treasury to reduce debt rather than speculating.

Question 5

Google’s massive AI capex is likely rational given its ROIC track record

Accepted Answer

Investors punished Google for announcing $75B in 2025 capex, but Friedberg and Antonio argue the spend is probably disciplined given Google’s history as the most efficient infra operator. Google has lengthened data center depreciation from ~2–3 to 6 years through custom hardware, energy efficiency, and AI‑assisted ops, and only needs roughly an incremental ~$27B/year in operating profit (depreciation + ~20% ROIC) for the plan to pencil out. They note Google’s models are likely the best general-purpose LLMs today; the challenge is translating that into products (search → chat, ads optimization, cloud). The lesson for tech operators: framing and ROIC storytelling around AI spend is as important as the raw dollar figure.

DOGE vs USAID, Crypto Framework, Google's $75B AI Spend, US Sovereign Wealth Fund, GLP-1s

DOGE vs. USAID, Crypto Pivot, Google’s AI Bet, GLP-1 Revolution

Key Takeaways

Notable Quotes

Questions Answered in This Episode

Transcript Preview

Get more from every podcast