All-In with Chamath, Jason, Sacks & Friedberg

Trump Brokers Gaza Peace Deal, National Guard in Chicago, OpenAI/AMD, AI Roundtripping, Gold Rally

October 10, 2025

Key Takeaways Copied to clipboard!

  • Former President Trump is credited by the hosts and commentators for engineering a significant, multi-stage ceasefire deal between Israel and Hamas, potentially averting wider regional conflict. 
  • The deployment of the National Guard in Chicago to protect ICE agents is framed by some hosts as a necessary federal intervention against local Democratic leadership enabling violent leftist protesters and hindering immigration law enforcement. 
  • The massive OpenAI/AMD GPU deal, which includes warrants for OpenAI, highlights the intense capital race in AI, where control over critical supply chain inputs like HBM memory and energy (electrons) is becoming the primary determinant of power, echoing historical control over money supply. 
  • The Total Addressable Market (TAM) for AI infrastructure, considering both CapEx and long-term OPEX, is significantly larger than initial estimates, potentially reaching tens of trillions of dollars over the useful life of the assets. 
  • The 'AI Roundtripping' concerns regarding companies like NVIDIA extending credit or making equity investments in customers like OpenAI are likely overblown, as the underlying demand for AI tokens and services provides genuine economic substance to these transactions. 
  • The recent gold rally is attributed to a combination of new demand sources like Tether Gold stablecoin issuance, central bank rebalancing (especially China diversifying away from USD), and a general loss of confidence in central bank policy globally. 

Segments

Bestie Intros and Market Check
Copied to clipboard!
(00:00:00)
  • Key Takeaway: Chamath’s wealth surge, driven by high-performing stocks like Robinhood, is noted amidst the return of guest Brad Gerstner.
  • Summary: Chamath expressed being ‘off the reservation’ financially due to stock performance, specifically mentioning Robinhood’s rise, which David Sacks attributed partly to the Trump administration’s crypto policies. Brad Gerstner returned to the podcast after a significant period, referencing past discussions on ChamasPACs and crypto events.
Trump Brokers Gaza Peace Deal
Copied to clipboard!
(00:02:48)
  • Key Takeaway: President Trump announced the first phase of a multi-stage Gaza ceasefire deal, which includes a ceasefire, unrestricted aid, and the release of all Israeli hostages.
  • Summary: The deal, based on Trump’s 20-point peace plan, involves Israel releasing 2,000 Palestinian prisoners and beginning troop withdrawal in exchange for hostage release. Diplomat Aaron David Miller credited Trump with demonstrating a unique degree of will and pressure on both sides, unlike previous administrations. Stability in the region is seen as crucial for Middle Eastern countries to focus on monetizing oil reserves before the transition to renewable energy accelerates.
National Guard Deployed in Chicago
Copied to clipboard!
(00:13:29)
  • Key Takeaway: Federal deployment of the National Guard in Chicago to protect ICE agents during raids is met with strong opposition from local Democratic leaders who claim authoritarianism.
  • Summary: DHS Operation Midway Blitz resulted in 800 arrests, prompting protests and executive orders from Mayor Brandon Johnson restricting ICE access to city property. Sacks argued that leftist judges and prosecutors enable repeat felons, leading to high crime, and that federal intervention to enforce laws is constitutionally supported, citing historical precedents like JFK’s actions. Brad noted that deportations have mean-reverted to pre-Biden levels, and suggested focusing enforcement on employers via steep fines rather than violent raids.
OpenAI/AMD GPU Deal Analysis
Copied to clipboard!
(00:44:06)
  • Key Takeaway: The massive OpenAI-AMD deal, featuring warrants for 10% of AMD stock, is viewed as a high-stakes bet by AMD’s Lisa Su to capture market share from NVIDIA’s near-monopoly in AI compute.
  • Summary: NVIDIA has captured nearly 100% of incremental AI data center revenue, leading AMD to make a ‘bet the farm’ move on its next-gen MI450 chip to compete. The shift in deal framing from counting chips to measuring gigawatts reflects energy constraints becoming the primary bottleneck for AI buildout. Control over critical supply chain elements, like HBM memory (where OpenAI is securing forward capacity), grants leverage to dictate the pace of AI expansion.
AI TAM Calculation Refinement
Copied to clipboard!
(01:00:04)
  • Key Takeaway: The true TAM for AI infrastructure spending must account for long-term OPEX, potentially multiplying the initial CapEx estimates by a factor related to the asset’s useful life (estimated at 20 years).
  • Summary: The initial TAM calculation for AI users, based on $3,000 average SaaS spend for 500 million business users and consumer spending analogies, suggested a multi-trillion dollar market. A counterpoint clarified that this only covers CapEx, and the total spend must include OPEX, which could be tens of trillions over 20 years. Data center hardware lifespan estimates vary, with some underwriting to six years while others suggest four.
Data Center Investment Scale
Copied to clipboard!
(01:01:44)
  • Key Takeaway: A rule of thumb estimates that each gigawatt of data center capacity requires approximately $50 billion in investment across chips, land, power, and shell.
  • Summary: Scaling to massive data centers, such as Elon Musk’s goal of 10 gigawatts, implies a potential $500 billion investment per facility. Current scaling plans suggest moving from 1GW to 2-3GW next year, then potentially to 5GW over the following couple of years. Brad Gerstner is convinced that demand for these resources will continue to exceed current projections.
NVIDIA’s Competitive Moat
Copied to clipboard!
(01:02:30)
  • Key Takeaway: NVIDIA’s sustained advantage in the GPU market relies on superior performance per watt, making their chips economically preferable even if competitor chips were priced at zero.
  • Summary: The key uncertainty for NVIDIA is whether the GPU market will bifurcate between training (where NVIDIA excels) and inference (which will constitute 99% of the market). However, CFOs of hyperscalers suggest that NVIDIA’s efficiency (perf per watt) is the deciding factor because power is the primary constrained resource. This efficiency advantage is seen as strong enough to withstand competition from AMD, custom ASICs, and in-house chips from companies like OpenAI.
AI Investment and GDP Impact
Copied to clipboard!
(01:03:42)
  • Key Takeaway: The massive investment in the US AI race, unlocked by policy enabling power and capital access, is viewed as creative destruction that could drive 4% to 5% GDP growth for years.
  • Summary: The current level of investment in US AI infrastructure is celebrated as crucial for maintaining global leadership in the AI race. Projections suggest generative AI revenue could exceed $100 billion by 2026 and grow to well over a trillion by 2030. A 10% productivity improvement across the $125 trillion global GDP represents $12 trillion in annual productivity gains, justifying the ‘heroic bets’ being placed.
Analyzing AI Roundtripping
Copied to clipboard!
(01:05:45)
  • Key Takeaway: AI roundtripping transactions, where NVIDIA effectively extends credit or invests in customers like OpenAI, are analogous to standard, well-established financing practices in legacy industries like auto manufacturing.
  • Summary: Concerns about conflicted party transactions are mitigated by the fact that these companies are heavily audited and compliance-focused, and similar forward flow agreements exist elsewhere. The transactions have economic substance because there is massive downstream demand for AI tokens and APIs, evidenced by OpenAI’s projected revenue ramp from $20 billion to potentially $100 billion. NVIDIA’s investments are tiny equity checks compared to their projected $450 billion in cash flow between 2025 and 2027.
Gold Rally Drivers
Copied to clipboard!
(01:15:00)
  • Key Takeaway: The gold price surge, outpacing Bitcoin and major indices, is driven by new speculative demand (like Tether Gold issuance) and geopolitical concerns leading central banks to diversify away from the US dollar.
  • Summary: Gold broke $4,000 for the first time, significantly outperforming the S&P 500 and NASDAQ year-to-date. Key drivers include Tether’s increasing custody of physical gold for its stablecoin and a loss of confidence in central bank policy globally. Furthermore, China’s central bank has increased gold reserves for 11 consecutive months, substituting away from US dollars and Treasuries due to the weaponization of the USD in foreign policy.
Polymarket’s Major Investment
Copied to clipboard!
(01:20:41)
  • Key Takeaway: The NYSE owner, ICE, invested $2 billion in Polymarket, signaling the professionalization and financialization of prediction markets, which are expected to converge with traditional asset classes.
  • Summary: Polymarket’s valuation has rapidly increased, and the partnership with ICE will distribute its data to thousands of financial institutions globally, suggesting an imminent US launch. The concept is that ’everything is becoming a market,’ potentially tokenizing equities and debt, and professionalizing the wisdom of the crowd by putting money behind predictions. This trend could profoundly impact industries like sports betting (FanDuel, DraftKings) and film studio hedging.