Gavin Baker

Gavin Baker has spent the summer asking every AI leader “Can you tell me one quantitative data point in your business that’s getting worse? Just one” — and through July and August, nobody could. AI broadly accelerated both months even as some AI names fell into significant drawdowns; the index calm is misleading because “you can drown crossing a river that’s on average two feet deep.” His caveat: Anthropic is in an IPO quiet period.

Both men reject zero-sum framing: this cycle is “not an or thing, it’s an and thing,” where frontier labs, open source, neoclouds, applications, and Nvidia all win. Every LP conversation starts with “How is this all gonna go wrong?”, but the supply-side data shows a roughly nine-to-ten-month payback for Nebius ($50B/gigawatt, 50-60% prepaid by customers), Blackstone/KKR/Apollo financing at low cost, and useful lives extending — true equity payback “might be way inside of a year.”

The demand side is “absolutely nowhere”: the companies’ roughly $180B of revenue rests on maybe sub-10 million heavy users against 1.5 billion knowledge workers. Baker’s fund Atreides grew internal token consumption 100x from March to August; some AI-native companies already spend 10%+ of human compensation on tokens versus ~1% at old-economy firms. Both worry more about undersupply than overbuild through ’28 — Dwarkesh’s scenario of token prices rising 10x is “the opposite direction of where everybody thinks this is gonna go.”

Public markets will have to digest lab revenue as a dial, not a stream: a lab monetizing 8 gigawatts of inference at ~$60B per gigawatt per year could cut revenue from $480B to $120B in this example by reallocating to training — “and I actually think they would do that.” Satya “blinked” on capex and regrets it; Dario chose bankruptcy-avoidance over share, “and OpenAI was aggressive, and now OpenAI is back in the game.”

Baker and George argue that the AI industry must tell its own truth: data centers are “probably the best thing that has ever happened to working-class Americans.” Town tax revenue “10Xs,” Loudoun County pairs America’s highest income with its highest data-center density, and cheap natgas ($2-3 vs ~$20-25 in Europe/Asia) is reindustrializing America — while Baker alleges “an organized CCP-funded campaign… laundered through TikTok” against data centers. His favorite Dario line: “stop talking about curing cancer and actually cure cancer.”

Orbital compute flips on Starship reusability: of $50B per gigawatt, ~$35B is IT either way, while the $15B of terrestrial power/cooling/labor is inflationary — and reusable launch takes the space alternative under $1B. Elon and Jensen have co-designed a Reuben rack targeted for a Q4 ’27 launch; even two quarters late, “that’s 2028,” and per Brad Gershner it’s “happening in plain sight.” Training stays on Earth — latency and speed of light are real.

The endgame is an ensemble of models behind routers, and the “arbiter of intelligence” abstraction layer is the most vied-for position “in the history of business.” Enterprises will RL open-source base models (soon likely Nvidia’s, via Nemotron and the Poolside acquisition) on their own data rather than hand context to frontier labs; Fireworks Nexus is the best instantiation today; Kirkland & Ellis’s $500M in-house build validates the category but understates the difficulty.

On Nvidia, Baker’s rule for semiconductor CEOs is “the only thing you should ever say is ‘Thank you, Jensen’” — while George’s rule of thumb is that every 1% of accelerator share is worth ~$100B, so plug into Jensen’s ecosystem rather than tugging on Superman’s cape. Baker estimates Jensen has locked up roughly 70-80% of the supply chain; his data centers are the most financeable ($15B equity on $50B), and George’s deal hierarchy reads true customer preference — equity investments beat RVGs beat token-priced warrants beat naked warrants.

🔗 Original source & video: Why AI Demand Is Outrunning Compute Supply

Gavin Baker sees the FT’s $2 trillion Anthropic IPO figure as potentially banker theater, not a reliable clearing price. Losers of the lead-left mandate “leak to the press” and set the bar high to make the winner look bad; he thinks $2T may be where you’d price to absorb the lockup, while his earlier 3–4T call is “where it might trade” — against a $100–120B exit run rate (16–20x sales) after three straight years of 10X growth.

Both Sacks and Baker converge on roughly $400–500B exit ARR for Anthropic in 2027, constrained by physical supply rather than lack of demand. Sacks: “the over-under really is 500 billion exit ARR… and I think they think they’ll do the over”; Gavin heard that at $60B ARR the company believed it could reach $600B within a year — and that Dario has said Anthropic “might be the only private company in the world,” hubris Gavin flags as “getting into SBF land a little bit.”

Sacks says OpenAI has pivoted to coding and reportedly reaccelerated to over 20% month-over-month growth with GPT-5.6 — a 10X annual pace matching Anthropic’s. Anthropic’s original coding bet came from watching Cursor’s utilization on its own platform (“We’re gonna take that business”), and moving vertically into what users build on you is now the industry playbook.

The episode’s ideological core: Gavin says Anthropic and the effective altruists think AI is “too dangerous to distribute”; Zuckerberg — and, he thinks, Elon and Jensen — think it’s “too dangerous to centralize,” and Gavin says “history has spoken.” Sacks adds the commercial irony: had Dario won his “FAA for AI” in Washington, the six-month frontier lead underwriting Anthropic’s pricing power “will be gone like that. Your model will be commoditized.”

Gavin’s counter to the bubble crowd: the overwhelming majority of tokens are profitable across the chain, and the Anthropic S-1 “is gonna break a lot of people’s brains.” Macro and value investors assuming subsidized tokens are like someone bearish because “oil’s at $500 a barrel… It’s just not.” Sacks frames Anthropic’s quarterly prints as the industry’s pace car — ~$100B revenue per gigawatt funding SpaceX at ~$50B/GW and NVIDIA at ~$30B — where a demand wobble means a pile-up.

NVIDIA’s $500B financing effort with Goldman, BlackRock, Blackstone, KKR and Apollo turns GPU compute into an asset class — Gavin calls NVIDIA “the central bank of AI.” Residual-value guarantees plus revenue shares above the floor could make it “a very large cloud with a capital light business” (Morgan Stanley); CoreWeave renting 2020 Amperes profitably out to 2029 implies nine-year GPU lives. Sacks’ lone risk: dark GPUs à la dark fiber if buildouts assumed $30–50/watt spot — though anti-data-center politics ironically insure against a glut.

Grok 4.6 breaks Sacks’ “frontier duopoly” frame: Pareto-dominant on intelligence-per-cost, only 1.5T parameters, with Grok 4.7 weeks away. Gavin’s positioning tell: SpaceX investors hold nuanced views on Starlink and orbital compute, but “very few of these investors talk about Grok at all” — while Elon runs a call option on the frontier and a put option of selling compute at “spot minus 90.”

A rare genuine bestie fight closes the show: JCal calls Amazon’s DSP subcontracting capitalism gone “a little too clever” and predicts “Mamdani and New Jersey are gonna win these lawsuits”; Sacks defends freedom of contract at $5.20 per package and $664 per household per year. Plus breaking news: Silver Lake may be buying Workday (+17%) — Gavin guesses the returning private-equity bid reflects the fact that open source is “an absolute godsend for the American software industry.”

🔗 Original source & video: Anthropic’s $2T IPO, Zuck’s AI Manifesto, Nvidia’s $500B AI Bet, Grok’s Comeback

July was “2022 in a month” — AI names down 40-60% in a straight line — yet Baker’s week of pressure-testing Silicon Valley for “one negative quantitative metric” turned up no clear one: GPU availability, GPU rental pricing, spot DRAM, and token growth are all accelerating. Nvidia now trades at its lowest forward P/E in ten years; “the market 100% thinks they are significantly over-earning… maybe they are.”

The core long thesis is compute repricing: everyone in ‘24-‘25 modeled GPU prices declining, but old-GPU prices are “going vertical in 2026” — one startup rented an identical B200 cluster at mid-$2/GPU-hour and, seven months later, hoped to pay just under $4, while an inference cloud plans to pay 100% more at renewal. As below-market contracts roll off, the hyperscalers’ installed base reprices higher: “essentially all the hyperscalers are under-earning.”

Credit is the only bearish catalyst he calls real — hyperscaler CDS blowing out, a Meta bond pricing poorly, real yields up — but consensus monetizes incoming Blackwell/Reuben gigawatts at Aier-generation rates ($1.3-1.4T of hyperscaler operating cash flow); monetize at merely a discount to current Blackwell and it’s ~$2T, “taking 700 billion of credit demand out.” Failsafe: “if credit’s not there, it just means the flops that are there are going to be even more valuable.”

The open-source freakout was backwards: “a token is a token” — GLM 5.2 and Kimmy K3 shift tokens from ~90%-gross-margin frontier to ~30%-margin open source on the same flops, memory, and watts, moving margin dollars into the infrastructure layer. The tell: Jensen wouldn’t be “the world’s biggest supporter of open source if it was bad for his business.”

Memory LTAs are now franchise-defining: with four buyers at scale and market share set by supply allocations, “you might blow up your entire business and your franchise by breaking an LTA.” Nvidia’s answer — a “credit wrapper with a revenue share if GPU prices are above a floor” plus equity stakes everywhere — is misunderstood, and exactly what he’d copy if he ran likely Hynix.

Market microstructure has changed: everyone feeds every headline into Claude, and “Claude is kind of Walter Cronkite for the stock market” — smart but not always right, compressing a three-year Japanese capacitor cycle into six weeks. His anchor: “the three most important words in investing aren’t margin of safety, but I don’t know.”

Biggest risk is regulatory, not fundamental: New York’s data-center moratorium looks like the first of many while the industry loses a PR war partly built on an admitted 10,000x water-usage error — despite data centers being “the best thing to happen for blue-collar wages in my lifetime.”

SpaceX is the underpriced compute machine: monetizing ~$50B per gigawatt against $73B consensus next-year revenue, with a public report claiming 8GW in 18 months he “almost doesn’t believe,” fundamentals better since IPO (Grok 4.5, Cursor), and “orbital compute feels more real every day.”

🔗 Original source & video: The AI Selloff Doesn’t Match the Data | Top AI Investor Explains

New York’s Democratic-socialist sweep showed how disciplined activists can capture safe seats and pull an entire party left. Polymarket put the three-candidate sweep at just 26% before election day, yet Mamdani-backed challengers won in NY-7, NY-10, and NY-13, including two defeats of established incumbents. David Sacks’s key investor-relevant point was organizational leverage: the DSA views Democrats as “a ballot access vehicle,” so even members who survive may alter votes and rhetoric to avoid primaries.

The panel attributed socialism’s momentum less to working-class demand than to downward mobility, institutional failure, and exceptional political communication. Gavin Baker described the emerging base as relatively affluent, highly educated white liberals whose outcomes differed sharply from peers in industry, while Chamath Palihapitiya emphasized housing costs, college debt, healthcare dysfunction, and inflation as proof to younger voters that “the system’s rigged.” Baker’s changed view was explicit: he once considered AOC the Democrats’ standout communicator, but now calls Zohran Mamdani “one of the most talented politicians I’ve ever seen.”

The proposed cure for youth radicalization exposed a fundamental split between child protection and anonymous speech. Chamath Palihapitiya argued that under-16 social-media bans could reduce early exposure and produce “a far less radicalized youth”; Travis Kalanick agreed that social media is “brain rot for real” but said age verification is really a mechanism to deanonymize adults and construct “a full-scale censorship regime.” Baker supported the child-safety objective but judged Kalanick’s free-speech cost “really high.”

Israel has become a primary-election fault line whose intensity is increasingly explained by age rather than party alone. Sacks presented Brad Lander’s defeat of pro-Israel incumbent Dan Goldman as close to a single-issue test, citing 80% Democratic disapproval of Israel and 57% disapproval among Republicans under 50. Chamath insisted critics must separate Jews, Israelis, the state of Israel, and Benjamin Netanyahu: collapsing them into one object of blame is “insane.”

GLM 5.2 turns China’s open-model progress from a strategic warning into an immediate commercial constraint. The 744-billion-parameter, one-million-token-context MIT-licensed model scored 51 on Artificial Analysis, beat GPT 5.5 on Frontier SWE, trailed Claude Opus 4.8 by under one point, and was described as 85% cheaper than GPT 5.5 at comparable performance. With Opus 4.8 rolled back and GPT 5.6 navigating approvals, Sacks warned that America is “on a shot clock”: stopping domestic deployment will not stop China.

Open weights may compress frontier-model margins while expanding the infrastructure opportunity. Baker expects enterprises to run a “council of LLMs,” routing perhaps 85% of work through customized open models and escalating only difficult tasks to frontier systems; his rough current split was open models processing 80%-plus of tokens while frontier tokens capture about 90% of economic value. His conclusion was not that frontier labs disappear, but that composability shifts value toward chips, memory, hosting, and routing.

Micron’s quarter confirmed that DRAM—not exotic components—is the binding AI bottleneck and a source of consumer inflation. Revenue rose from roughly $9 billion to $42 billion, Q4 guidance reached $50 billion versus $43 billion expected, and 2026 HBM supply was already sold out; Baker expects DRAM to absorb 30%-40% of hyperscaler capex next year. Apple’s $699 MacBook Neo moving to $799 and Mac Studio pricing rising 25% illustrated the spillover: “Inflation has come to the desktop.”

Scarce terrestrial power strengthens both orbital-compute economics and the value of large energized sites, while public markets remain able to fund the buildout. Baker put a one-gigawatt terrestrial data center at $35 billion of semiconductors plus $25 billion of power and cooling, versus approximately $40 billion in orbit if reusable Starship lowers launch to $5 billion. He separately valued Anthropic at roughly $3 trillion and argued that public markets need absorb only the offered slice, but Cerebras showed why execution matters: once an IPO breaks deal price, mechanical selling and opportunistic shorts can turn disappointment into “a pile-on.”

🔗 Original source & video: Socialists Sweep NYC, China Catches Up in Coding, AI Memory Crunch, Micron’s Blowout Quarter

SpaceX prices Friday at $135/share, $1.77T — and the table’s verdict is unambiguous: Brad calls it “a must buy, a must own, a set it and forget it” for any investor who is “AI pilled.” Gavin’s two levers: how fast SpaceX brings on terrestrial data centers (Elon stands them up in 122 days — “speed is literally cost”) and whether the Cursor acquisition puts xAI on the coding Pareto frontier, where “all frontier model revenue will accrue.”

SpaceX became the #4 hyperscaler in 30 days — “EWS” wasn’t in many people’s models six months ago. The Anthropic deal monetizes at $22-23B per gigawatt per year, Google at $50B, versus only ~$14B implied in the leaked $160B 2028 revenue number — meaning the Street math clears before any orbital leap of faith, and Altimeter’s Freida pegged a 55% IRR on the Anthropic deal (“if you can borrow at 6, 7, 8% and invest at 55%… that math maths”).

Clark’s view: orbital compute is a call option, not a requirement: with rapid two-stage Starship reusability ($1,500/kg on Falcon → $250/kg or lower), the math backs into ~$5B per gigawatt of capex in space vs $20-25B terrestrially for the non-silicon half of the bill of materials — a 5x reduction — conditional on satellite reliability, since “GPUs melt and lasers fail.”

The least-talked-about upside is the model itself: Cursor’s Composer 2.5 (built on Kimi K2.5 plus proprietary coding data exceeding the public internet) was Pareto dominant 12 days ago, and Grok 4.3’s 1.5T-parameter run is now training with Cursor data injected into pre-training. Brad: if there’s an upside surprise in the IPO, “this is the place getting the least amount of attention.”

Fable 5 and Mythos reset the compute bull case: the unlock is long-running tasks, and per Noam Brown’s “polynomial” post, “we do not know how smart these models are” because nobody has run one continuously for a year — imagine Einstein thinking about physics 24 hours a day. Gavin: “however bullish I was on compute before, I’m just a lot more bullish.”

Frontier captures ~90% of revenue even as open source may take 80% of tokens — the “cheap tokens catch up” thesis was “decisively wrong” on revenue. Twist: open source is bearish for frontier labs but bullish for compute providers, and Gavin thinks Nvidia could weaponize it against ASICs — “That’s a cute ASIC you’ve built there… How would you like open source to join the frontier?”

The capex math maths — but both PMs have dialed risk down. ~$1.5T of 2027 capex against ~$300B+ inference revenue at 60-70% gross margins works, especially with per-gigawatt monetization up from ~$20B to $30-40B this year and Anthropic hitting “accidental profitability.” Still, with CPI back at 4.2, semis having “gone straight up a cliff,” Altimeter has cut from large to medium-small exposure — “consolidation on the way to much higher highs.”

🔗 Original source & video: The SpaceX IPO, Fable 5, AI Capex Update & Market Check w/ Gavin Baker, Andrew Fox & Clark Tang

Starship V3’s first flight last week “was a huge success” — full-system demo of the new V3 Raptors plus a soft second-stage splashdown gives Johnsen conviction “not just in years to come, but really even in the next couple of flights.” Once the second stage returns to the tower for rapid reuse “in the next couple of years,” he expects another 10x drop in cost per kilogram vs. Falcon — chasing “the holy grail of rocketry”: aircraft-like operations at 100 metric tons to LEO.

Starlink is attacking a ~$1.6T telecom market with what Baker calls the first genuinely differentiated product in telecom history — 10M+ customers, 10,000+ satellites, 160+ countries in six years, and Johnsen says 10M “can become hundreds of millions.” Next-gen direct-to-cell will be “5G quality… in the next two years,” and each Starship carries 20x the Starlink V3 capacity of a Falcon launch.

Orbital compute is “racks in space,” not buildings — larger Starlink V3 satellites with more solar, Nvidia GPUs, and a radiative-cooling sheet, virtually networked “literally like another constellation.” First-principles edge: 5x+ solar energy per cell, sun-sync 24-hour power, no land lease, and avoidance of the “data center in my backyard” concern — and while terrestrial data-center costs are rising, the satellite’s cost “is mostly silicon” and benefits from silicon cost reductions, process node to process node.

The launch math today: ~200 Starship launches per gigawatt of orbital compute (first-gen satellite and rocket), so SpaceX is “capacitizing for thousands of launches a year” across four towers — two in South Texas, Cape Canaveral, and LC-37 within the next year — with capabilities demonstrated “as soon as next year.”

SpaceX’s AI business is already a top-5 AI infrastructure operation by Baker’s napkin math — the Anthropic hosting deal puts it at a $3.75B-per-quarter run rate, “50% larger than the company just outside the top four,” while its own models train and do inference on bleeding-edge GB300s and the Cursor deal brings over half the Fortune 500 as enterprise accounts.

Terafab exists because SpaceX fears silicon supply won’t scale — Nvidia, AI5, or TPU all resolve to TSMC one layer down, and the target is “ideally 100 gigawatts a year.” The Intel partnership plus captive customers (“we will take every wafer that you can yield out”) reduces it to “a capital risk only.”

The capital-allocation timing is the quiet thesis: Starship flies Starlink V3, which unlocks cash-generative Starlink growth and presumably high-margin direct-to-cell exactly as orbital compute needs heavy capex. Johnsen: “I wish I could take credit for that, but that’s certainly Elon.” And the vertically integrated stack is open at every layer — competitors can buy launch, Starlink, terrestrial compute, or the model.

🔗 Original source & video: Gavin Baker interviews SpaceX CFO Bret Johnsen at Mission Control

SpaceX’s filing supplied a plausible $2 trillion case, but the near-term engine is terrestrial compute rather than orbital data centers. The proposed $75 billion raise at a $1.75 trillion valuation rests on Starlink’s $11.4 billion revenue and $4.4 billion operating income, plus Anthropic’s cancellable $1.25 billion-a-month Colossus contract. Friedberg described the basic internet-infrastructure case; Chamath’s underwriting emphasized a “capital moat” accelerating technology, execution, and learning, with terrestrial data centers alone potentially reaching $100-$200 billion of revenue by 2030-32.

Nvidia’s $81.6 billion quarter strengthened the fundamentals even as its valuation lagged richer AI-adjacent trades. Revenue rose 85% year-over-year and 20% sequentially, with $58 billion of net income, $48 billion of free cash flow, and 75% gross margins; Gavin argued Nvidia may also be growing faster than Broadcom in comparable Western AI infrastructure. The unanswered bear case is benchmarking: rival ASICs are not entering MLPerf or comparable tests, leaving the share-loss argument “fighting shadows.”

Anthropic’s profitability is beginning to answer the AI-ROI objection, while Cursor’s proprietary coding data strengthens the model-development case. Gavin estimated OpenAI and Anthropic together at roughly $100 billion of ARR with approximately 80% inference gross margins, and saw $200-$400 billion of aggregate LLM ARR as plausible by year-end. Adding Andrej Karpathy to pursue recursive self-improvement and continual learning could put model development on Chamath’s combination of “overdrive and autopilot.”

The AI backlash is being driven as much by concentrated gains and careless corporate messaging as by the technology itself. Friedberg argued that people see a small group collecting outsized returns before benefits diffuse, while Jason highlighted Cloudflare cutting more than 20% of staff and Meta laying off 8,000 people while recording work on employees’ computers to train models. Chamath’s verdict on branding former workers “measurers”: tech CEOs “don’t understand the moment.”

The panel rejected a unilateral AI slowdown while leaving room for narrow, internationally coordinated guardrails. Friedberg emphasized state competition and the risk of foreign efforts to slow technological progress; Chamath used postwar nuclear proliferation as an analogy for why the U.S. and China may need to preserve strategic balance. Kevin Weil gave the cautious non-answer on testing frontier models, while Gavin argued that courts, liability, and self-regulation already create incentives for responsible conduct and that government power is a “one-way ratchet.”

Autonomy and public-safety AI will likely advance through safety economics and competition between jurisdictions. Gavin said human-driving mandates will invite wrongful-death suits and predicted that cities without Waymo or cybercabs will eventually feel “barbaric and unsafe,” while Chamath said policymakers should first ask warehouse workers whether jobs with 35%-40% churn are actually worth preserving. Friedberg framed Flock Safety and Las Vegas’s drone-enabled policing as evidence that “crime is now a choice”; he estimated $30-$40 million annually could make Las Vegas much safer.

The macro backdrop is unequivocally harder, but the panel split between systemic alarm and selective ownership. Oil-driven inflation forecasts reached 4.2%-6%, the U.S. 10-year hit 4.6%, and Japan’s 30-year was cited inconsistently in the transcript as both 3.1% and 5.2%. Friedberg warned that 310% global debt-to-GDP could turn rising yields into a credit crisis. Chamath’s response was to own five or fewer businesses for a decade, while Gavin held that higher rates, extraordinary AI growth, and America’s relative energy advantage can all be true.

U.S.-China talks produced commerce rather than a grand bargain, yet continued contact may itself reduce geopolitical tail risk. Friedberg saw no durable de-escalation, whereas Chamath suggested there may have been useful private alignment on the geopolitical “tic-tac-toe” and said that America and China talking is “only good.” Gavin supported selling depreciated Nvidia GPUs to China because it could discourage a rival, more power-hungry ecosystem and keep America’s platform embedded.

🔗 Original source & video: SpaceX’s $2T Case, Nvidia’s Shock Selloff, America Turns on AI, Trump Pulls AI Order, Bond Crisis?

March/April was a lean-in drawdown, not a take-your-medicine drawdown: while the Nasdaq sold off, Anthropic added $11B of ARR in a single month — equal to the combined businesses Palantir, Snowflake and Databricks took a decade to build. “Nothing like that has ever happened in the history of capitalism,” and tech simultaneously got as cheap versus the market as at any point in the last 10 years. “All you had to do in March was simply observe what was happening to Anthropic.”

Anthropic at a rumored $900B on $50B ARR is cheaper than it looks: Baker argues compute constraints and a degraded Claude (Opus generating ~70% fewer tokens per question) mask an “unconstrained run-rate revenue” of $100–200B — so you may be paying ~5x “URR.” They could raise at “at least a 100% premium” but are following Elon’s lesson: “never being greedy on valuation” buys 20–30 years of capital access.

The single bubble tell is TSMC’s capacity decisions. If TSMC gave Jensen what he wants, Nvidia could sell “$2 trillion of GPUs in ‘26 or ‘27. Maybe 2½. Maybe 3” — and that’s overbuild territory. History (Carlota Perez, canals, railroads, 2000) says expect a bubble; the offsets are a build-out still funded from operating cash flow and GPUs at 100% utilization vs. 99% dark fiber. Watch for Intel or Samsung breaking discipline: “they’re not going to stay disciplined. They will break.”

Orbital compute is “racks in space,” not Pentagon-sized data centers — a Blackwell rack is 3,000 lb and 100kW; Starlink V3 is expected to operate at 20kW, and SpaceX is targeting 100–120kW satellites linked by lasers through vacuum. Watts shortage alleviates ‘27–‘28, then orbital solves it — a direct warning to terrestrial power/cooling suppliers whose capacity ramps land “just as all of the silly skeptics start to understand that orbital compute is very real.”

Frontier tokens are capturing the overwhelming majority of model-layer value — Gemini 3.1 Pro went from “mind-blowing” to “intolerable” — and the Pareto frontier flipped: Google dominated 9 months ago; now it’s Anthropic and OpenAI, with Grok 4.3 on the frontier and Gemini clinging on, likely “subsidizing out of pride.” Meanwhile AI’s shift “from all you can eat to pay by the drink” (his telecom-analyst analogy) is probably why OpenAI and Anthropic should exceed well over $200 ARR this year.

Prefill/decode disaggregation extends GPU lives to 10–15 years — put Cerebras or Groq LPUs in front of Hoppers and Amperes and run them “until it melts” — which “may single-handedly save private credit” and, by cutting GPU financing from CoreWeave’s low-7s toward 5–6%, mathematically lowers the cost of the whole build-out.

Cross-sectional valuations “flat out do not make sense”: semi-cap at 40x next quarter’s annualized earnings vs. DRAM at mid-single-digits (last cycle’s peak gap was ~5 vs. 12); the lowest-quality, highest-cost “sellers of shortage” are mooning on X-driven flows while quality lags. AI intra-correlations blew apart in January — you can no longer hedge memory with semi-cap — and miscategorized names like Astera (a switch company stuck in “copper loser” baskets) are the opportunity. “I just wish there were more AI bears.”

The three questions that decide everything: does the frontier-token premium persist, does the bitter lesson survive contact with ASI (“who knows if the bitter lesson holds for 400-IQ models”), and when does continual learning arrive — “if we get that, then we have a really fast takeoff.” Plus the dark coda: rising political violence around AI, and “the machine gun is here… if we do not all become masters of the machine gun, we’re going to get mastered.”

🔗 Original source & video: Watts, Wafers, and the Future of AI Infra | Gavin Baker

Baker treats March’s AI selloff as “pent-up alpha,” because prices fell while Anthropic added $11 billion of ARR in one month and a 500% NDR statistic was shared on the show. He argues that this one-month addition rivaled the combined businesses of Palantir, Snowflake, and Databricks, while DeepSeek had already demonstrated that reasoning models increase inference demand: Asian AWS availability-zone prices doubled, GPU availability fell, and DRAM went vertical. “I’ve just never seen an exponential like this.”

Reported revenue may radically understate Anthropic’s compute-constrained economics. Against roughly $50 billion of ARR and a $900 billion valuation, Baker estimates unlimited compute might support $100 billion to $200 billion of revenue, or roughly five times his invented “URR, unconstrained run-rate revenue”; he also believes Anthropic burned perhaps 80% less capital than OpenAI at comparable scale. His expectation that Anthropic could generate cash this year remains a forecast, not a certainty.

The terrestrial watts shortage should begin easing in 2027 or 2028, but zoning and approvals may become more binding than turbines or fuel. Longer term, Baker’s answer is “racks in space”: roughly rack-sized satellites with immense solar wings, radiators, and laser links forming virtual data centers, with inference moving orbital while training stays on Earth. Cooling appears solvable to SpaceX engineers; repair remains the honest weakness “until you have probably floating Optimuses.”

TSMC’s capacity decisions are Baker’s single best indicator of whether AI becomes a classic infrastructure bubble. Unlike 2000, today’s build-out is funded overwhelmingly from operating cash flow and GPUs run at 100% utilization, versus 99% of fiber sitting unused; yet if TSMC supplied everything Jensen Huang wanted, Baker thinks NVIDIA might sell $2 trillion to $3 trillion of GPUs in 2026 or 2027 and eventually overbuild. The “Goldilocks zone” is enough expansion to keep Intel or Samsung below roughly 30% share, but not enough to remove wafer scarcity.

Terafab could challenge normal fab timelines without alienating TSMC. Baker describes a SpaceX joint venture, with Tesla possibly involved, that would use Intel’s institutional knowledge and recruit the equipment companies’ A-teams. He says the relevant process gap may be roughly 9 to 15 months, or three to five quarters, behind the frontier, but presents that timing approximately.

Frontier-token economics remain unusually durable, but three uncertainties could overturn the trade: whether frontier tokens retain their premium, whether ASI violates the “bitter lesson,” and when continual learning arrives. Baker says Gemini 3.1 Pro went from mind-blowing to intolerable as the frontier advanced, while Anthropic, OpenAI, and Grok 4.3 now dominate the intelligence-versus-cost Pareto frontier. A model that updates from one experience rather than “put its hand in the fire a million times” could produce a very fast takeoff—and perhaps optimize away some compute demand.

Usage-based pricing could steepen AI revenue as capped subscriptions obscure what frontier systems can do. Baker says capped $250-to-$300 plans increasingly deliver rate-limited, “lobotomized” models, whereas enterprise and usage plans expose the token budget and agent harness; he compares the shift to telecom moving from all-you-can-eat to “pay by the drink.” His aggressive call is that OpenAI and Anthropic will exceed well over $200 billion in ARR this year as compute availability, frontier pricing, and fleets of agents expand usage.

New chip companies need to be both different and hard, because “a better GPU” invites NVIDIA to copy the trade-off with superior economics and customer knowledge. Disaggregating prefill from decode creates openings for specialized memory-capacity and memory-bandwidth architectures; Baker’s venture rule is that 1% market share could be worth $100 billion, with Cerebras’s wafer-scale design as the exemplar. The same disaggregation could extend Hopper and Ampere lives to 10 or 15 years, potentially lowering GPU financing from the low sevens toward 5% or 6%.

Application investors need exposure to the “token path,” defensible scale, or a niche the model companies will not absorb. Baker credits Cursor and Cognition’s coding focus and cites Replit founder Amjad Masad’s view that coding may be the shortest path to useful AI or ASI. He says AI has still destroyed trillions of dollars of application-layer value and that economic gains currently favor businesses with the highest ratio of utilized GPUs per employee. Meanwhile, public-market selection has become finer-grained: semiconductor-equipment companies at 40 times annualized next-quarter earnings and DRAM companies at mid-single-digit multiples “cannot all be true.”

🔗 Original source & video: Gavin Baker - Watts and Wafers - [Invest Like the Best, EP.473]

The memory call: maybe don’t sell. Prices are up 60% and Micron’s margins are high-60s vs a historical average closer to 16% — and Baker, Micron’s analyst in 2000, admits “based on every memory cycle we have had for the last 25 years, this is the time to be selling memory. 100%.” Except one: the mid-90s, “the last true capacity cycle,” and on that template “we may still be very early.” On the numbers other Sohn panels gave: “I take the over on every number. Every single number.”

Maybe no bubble this time — TSMC limits supply. Every profound technology bubbles, but AI has physical constraints past manias lacked: TSMC’s “flinty old men and women in their 70s,” guardians of Morris Chang’s legacy, may expand 5% while Jensen — visiting every 3 months — wants double or triple. If they double or triple, “Nvidia could probably sell 1, 1.5, 2 trillion dollars worth of chips next year. I really believe that.”

He takes the over on OpenAI + Anthropic at $200B combined revenue near term. The engine: the shift from $250/month all-you-can-eat to usage-based pricing with frontier capability “locked behind harnesses” on enterprise plans — “wildly bullish” for token pricing, the same overage model that made cellular a great growth industry. Only “10 basis points of the world’s population” uses models properly amid an insane shortage; at 5% it’s “unimaginable.”

Trainium is “by far” the most underestimated custom chip — Trainium 3’s ramp in the second half of this year will make Trainium to 2026 what TPUs were to 2025. Google made “very conservative design choices” with TPU V8, and the only two functioning switched scale-up networks for mixture-of-experts inference are Nvidia’s and Amazon’s. Google won’t submit TPUs to MLPerf — its own benchmark — which is “visibly driving Jensen crazy.”

Neoclouds are durable, not a CapEx arbitrage. Running a cluster is Formula 1 — looks easy, would kill an amateur — and CoreWeave’s GPUs get utilized 2-3x more per hour than bottom-of-barrel providers, justifying the premium. His regret: The Trade Desk could have invested over $50M into CoreWeave at a $1.1B valuation but was conflicted out by Crusoe, which he says they still hold a large position in.

The underappreciated short: terrestrial power and cooling industrials. Orbital compute becomes provably “possible, going to work, and economical in the next 2 years” and takes meaningful share by end of decade — the years before that will be “very painful” for industrial names that “massively flexed up capacity” for a build-out that “could really come to a screeching halt.”

Process edge is overwhelmingly reading: he rarely meets public companies — “they never say anything that’s not in a transcript or 10-Q. And I can read much faster than they can speak.” And even non-coders get better investment answers from Claude Code or Code X than the regular model: coding may be “the ultimate AI app and it subsumes more and more.”

🔗 Original source & video: Inside the Mind of a Tech Investor with Gavin Baker

Gemini 3’s real significance: pre-training scaling laws held — stated “unequivocally,” which matters because “no one on planet Earth knows how or why scaling laws for pre-training work.” Baker’s contrarian frame: given the 200,000-Hopper coherence ceiling and Blackwell’s brutal transition, “there really should have been no progress in ‘24 and ‘25” — “reasoning kind of saved AI,” bridging an 18-month gap (ARC-AGI: 0→8% in four years, then 8%→95% in three months). The scaling laws are multiplicative, so “the Blackwell models are going to be amazing.”

Google’s low-cost-token advantage is temporary. Gemini 3 was trained on 2024-25-era TPU v6/v7 — “F-4 Phantoms” next to Blackwell’s F-35. Once GB300s (drop-in compatible with GB200 racks) shift to inference, vertically integrated Blackwell users become the low-cost producers, and Google’s rational strategy of “sucking the economic oxygen out of the AI ecosystem” at negative-30% margins gets painful: “it might start to impact their stock.” When Reuben lands, “the gap is going to expand significantly.”

The ASIC field likely narrows to TPU and Trainium. Broadcom takes 50-55% gross margin on the TPU back end — ~$15B of a ~$30B 2027 program against ~$5B of divisional opex — so in-housing is “absolutely inevitable” (MediaTek is the warning shot). It takes three generations to make a good chip, and Nvidia’s answer to every in-house ASIC is annual cadence: “you cannot keep up with us.”

Only four labs matter — OpenAI, Gemini, Anthropic, xAI — and the gap is compounding. Reasoning restarted the data flywheel (user feedback as verifiable reward), internal checkpoints train the next model, and Meta’s failure shows how hard it is: Zuckerberg “was as wrong as it was possible to be.” China’s refusal of Blackwell will blow the gap out — DeepSeek admitted the compute shortfall in its v3.2 paper — and China realizes “whoopsy daisy, we do need the Blackwells” around late ‘26, by which point rare-earth leverage is solved.

The board repositions around token cost: OpenAI is a high-cost producer paying margins for compute — “you go from $1.4 trillion rough vibes to code red pretty fast” — while Anthropic is “burning dramatically less cash than OpenAI and growing faster” and its $5B Nvidia deal gives Jensen three fighters against Google.

ROI is “empirically, factually, unambiguously” positive — audited ROIC at the big GPU buyers is up, and Q3 was the first quarter non-tech Fortune 500s quantified uplift: C.H. Robinson quoting truckloads in seconds at 100% of inbound requests (vs 15-45 minutes at 60%), stock +~20%. The risk he watches: a Blackwell ROI “air gap” while the chips do training, since “there’s no ROI on training — the ROI comes from inference.”

Application SaaS is repeating brick-and-mortar’s e-commerce mistake: clinging to 80% gross margins while AI natives run agents at sub-35-40% — “you are guaranteeing that you will not succeed at AI.” It’s “a life-or-death decision that essentially everyone except Microsoft is failing”; the activist playbook (show low-margin AI revenue) is open to Salesforce, ServiceNow, HubSpot, GitLab, Atlassian.

Bear cases and bubbles: edge AI — a free, pruned Gemini 5 or Grok 4.1 on-phone at 30-60 tokens/second, “clearly Apple’s strategy” — is “by far the most plausible and scariest bear case.” The rolling bubble has moved from EVs and meme stocks to nuclear and quantum, where no public vehicle is a real leader. Data centers in space are his “most important thing in the next 3-4 years” — “in every way… superior to data centers on earth.”

🔗 Original source & video: GPUs, TPUs, & The Economics of AI Explained | Gavin Baker Interview

Gavin Baker’s answer is no: today’s AI buildout does not resemble the 2000 bubble by either valuation or utilization. Cisco peaked around 150–180 times trailing earnings versus roughly 40 times for NVIDIA; 97% of peak-era fiber was dark, while “there are no dark GPUs” and training clusters are pushing chips until they melt. The largest public GPU buyers have gained roughly 10 points of ROIC since ramping capex—though whether that persists through Blackwell spending remains “an interesting and open debate,” and Baker personally thinks it will.

The infrastructure bill is enormous, but its buyers have an unusually deep balance-sheet buffer. David George framed roughly $1 trillion of existing US data centers, another $3–4 trillion planned over five years, and estimated more than $1 trillion of OpenAI commitments; against that, the major spenders generate about $300 billion of annual free cash flow and hold $500 billion of cash. At $40–50 billion per NVIDIA-powered gigawatt, George sees “an $800 billion buffer growing $300 billion every year,” even if near-term buildout creates some mismatch.

The feared round-tripping is real but, in Baker’s view, small and strategically rational. NVIDIA funding OpenAI while OpenAI buys NVIDIA chips looks circular because “money is fungible,” but Baker says the real driver is competition with Google’s TPU, DeepMind and Gemini—not weak underlying demand. Baker estimates Gemini had taken roughly 15–20 points of traffic share in two or three months and suspects Google may already have more AI traffic than OpenAI or Anthropic on an actual-traffic basis; the cited share gain did not include AI Overviews.

AI could reinforce much of the Mag 7, but execution failure remains existential. Incumbents possess the essential inputs—data, distribution, compute, capital and talent—so AI might be a sustaining innovation if they execute; otherwise, “IBM might be a good fate.” David George called ChatGPT “Pearl Harbor for Google,” while Baker cautions that frontier labs will structurally carry lower gross margins than SaaS because scaling laws and test-time compute keep the products compute-intensive.

Application SaaS is not necessarily dead, but winning requires embracing margin compression. Baker has softened his early-2024 view that all application SaaS “might be a zero,” especially for vendors serving fragmented SMB customers; his warning is that protecting 80–90% gross margins can sacrifice the AI opportunity. The operative choice is “10 bucks of revenue with 90% gross margins or 50 bucks of revenue with 60%,” while incumbents can subsidize break-even AI products before leaders such as Cursor accumulate enough tokens to make catching up difficult.

Distribution and reasoning have made consumer AI less hostile to durable platforms. AI-browser launches could let Google watch the pioneers for three to six months before responding through Chrome’s roughly 5 billion users. Reasoning and RL can turn a large user base into the classic product-data flywheel: users improve the algorithm, which improves the product. David George says GPT-5 is not evidence that scaling laws ended because it was “a smaller model” designed to run more economically, not to maximize capability.

Outcome pricing is the likely business-model shift, with robotics as the physical extension of the same logic. Customer support can charge per resolved task because success supplies a verified reward; personal agents may collect affiliate fees for completed purchases, squeezing the advertiser overpayment that made Google search so lucrative. Baker calls robotics “very real,” expects Tesla versus China, and thinks the humanoid debate is effectively over because robots can learn from video or human demonstrations and receive clean task-level feedback.

🔗 Original source & video: “Is there an AI bubble?” Gavin Baker and David George

GPT-5 ended OpenAI’s run of decisively superior launches, even if calling it a “flop” overstates a mixed result. Gavin Baker cited Grok 4 at 44.4% on Humanity’s Last Exam versus GPT-5 at 42%, plus a meaningful Grok lead on ARC-AGI-2; GPT-5 was only “very narrowly ahead” on Artificial Analysis. His signal for investors: this was “the first time” OpenAI failed to clear every rival decisively, while Grok 5 and a new Gemini remain pending.

OpenAI’s more defensible GPT-5 advance may be product simplification, not frontier intelligence. Friedberg found that GPT-5 routed a simple retrieval request to a fast response and a complex research task into thinking mode, eliminating the model-picker alphabet soup that ordinary users cannot interpret. The router was reportedly inconsistent for the first 12–16 hours. Jason relayed Benchmark investor Eric Vishria’s view that even if OpenAI “never releases another model,” its users, distribution and product quality could still make it highly valuable.

The AI capex cycle already shows economic returns and full utilization, weakening comparisons with the telecom bubble. Gavin said return on invested capital at the hyperscalers has risen during the buildout, first through “silicon opex to replace human opex” and now through revenue acceleration: Meta’s engagement and ad targeting improved, while Microsoft reported significant Copilot uptake. Dot-com capital funded dark fiber that sat unused; today’s GPUs run so intensively “that they melt,” meaning supply remains behind observable demand.

Power, not model access alone, is becoming the binding asset in the AI race. Phil Deutch cited NVIDIA’s Climate in a Bottle and said the sector remains in the first half of the first inning for solving energy’s hardest problems. The discussion cited Anthropic’s request for 50 GW of US power over three years—roughly 5% of current US electricity production and comparable to an entire year’s recent capacity additions—while data centers need continuous, location-specific power and are relatively price-insensitive. The discussion also cited China adding one terawatt every 18 months, with solar additions at extraordinary scale and more than 150 nuclear reactors in some stage of deployment.

China’s advantage is coordinated manufacturing and infrastructure, but its own suppression of entrepreneurship and weak rule of law may cap it. The existential case presented in the discussion spans AI, military power, TikTok, BYD and state-backed brands; Friedberg instead asked, “Who’s going to win the internet?” and argued that advantage will vary across models, drones, supply chains and markets. China can “throttle up and throttle down entrepreneurship,” but company formation peaked around 2018, DeepSeek personnel are tightly controlled, and Phil’s factory experience was that “every day was a renegotiation on terms.”

The socialism debate converged on housing costs and economic insecurity as the combustible inputs, but not on whether failure will cure the politics. Friedberg warned that rising rent, groceries and government dependence create a “slow-burning” ratchet in which every failed program becomes an argument for more intervention; David Sacks warned that a downturn could produce a backlash “much further to the left than Obama ever was.” Jason’s counter-program was abundant housing, broader ownership and cheaper education: build 10 million homes, loosen regulation, and reduce the anxiety of households spending 60% of their budget on shelter.

Trump’s tariff experiment is producing real revenue but still lacks a stable test of its inflation, demand and growth effects. July receipts reached $30 billion and 2025 receipts about $127 billion, but Ben noted that weak inflation could reflect falling demand—and that maximizing tariff revenue ultimately conflicts with reshoring, which eliminates imports. Phil Libin put success at “50/50.” David Sacks and Chamath Palihapitiya argued that retaliation and early damage had been lighter than expected, while Gavin’s more cautious hedge was that Trump appears responsive to market feedback but needs to restore policy stability.

Apple’s $700 billion of buybacks crystallized the episode’s capital-allocation argument: the company could fund returns and still compete far more aggressively in AI. Gavin rejected the false choice between repurchases and R&D but suggested redirecting even $200 billion toward data centers, calling Apple’s AI products “terrible” and its inaction astonishing. Gavin also offered the upside case for augmented-reality glasses, while Friedberg suggested Apple could eventually become the operating system for homes, EV batteries and distributed power.

🔗 Original source & video: OpenAI’s GPT-5 Flop, AI’s Unlimited Market, China’s Big Advantage, Rise in Socialism, Housing Crisis

Nvidia’s accounts-receivable expansion is best explained by a Blackwell installation bottleneck—unless it persists beyond the July quarter. Gavin Baker called the rise from roughly $1.5 billion to $5.5 billion “never good,” but argued that Hopper-to-Blackwell is an unprecedented physical transition: racks move from about 1,000 pounds, 60 kW and air cooling to 3,000 pounds, 120 kW and liquid cooling. He rejected the round-tripping thesis because Nvidia could have sold those GPUs to Meta, Amazon or Microsoft without investing in neoclouds.

CoreWeave’s IPO is a referendum on whether GPU-cloud operations are a commodity or a scarce execution capability. The proposed deal was cut to $1.5 billion at a $23 billion valuation after $2 billion of revenue, alongside nearly $8 billion of debt and Microsoft contributing more than 60% of sales. Baker’s counter to the bearish consensus: synchronizing tens of thousands of failure-prone GPUs is hard, and CoreWeave “runs these big GPU clusters as well as anyone.”

The AI infrastructure cycle did not look overbuilt to Baker after DeepSeek-R1. China was buying every available GPU, DRAM pricing was rising, and OpenAI had begun gating image generation for lack of compute; meanwhile, hyperscalers face a prisoner’s dilemma in which underspending on Blackwell might concede a durable advantage. His analogy: Blackwell is a Formula 1 car from ten years in the future that operators must learn to drive.

Export controls buy the US time while intensifying China’s incentive to build around Nvidia. Baker described enforcement as a “game of cat and mouse”: friction can preserve an American advantage, but necessity encourages domestic semiconductors and algorithmic advances such as DeepSeek. He put China’s five-year chance of matching or beating Nvidia at “zero,” then added, “Over the 10 years, who knows?”

Agents could make compute the binding constraint while potentially collapsing incumbent cost structures. Baker saw MCP becoming a standard that lets services such as Stripe connect once and work across models; if agents become real, Blackwell ROI rises even though it is “not good for human employment.” Friedberg envisioned three or four people tackling projects previously requiring 300 or 400 specialists, while Chamath argued that one- or two-order-of-magnitude Opex reductions could “blow a hole” in the $3.5 trillion–$4 trillion software complex.

Trump’s tariff program is a high-risk policy stack requiring reciprocal rates, deregulation, tax relief and domestic ownership of IP to move together. Baker framed reciprocity as “if you charge 5%, we charge 5%,” while saying every mention of tariffs should be accompanied by deregulation “two or three times.” Friedberg described the proposed elimination of taxes below $150,000 as the consumer offset to higher import prices; Jason likened the full maneuver to “four freight trains” exchanging sandwich ingredients through their windows.

The deficit arithmetic leaves DOGE carrying far more weight if tariff revenue disappoints. Against a stated $1.9 trillion deficit, Chamath contrasted Howard Lutnick’s trillion-dollar revenue ambition with Wall Street estimates of only $170 billion–$300 billion, concluding that Musk might need to find roughly $1.75 trillion rather than $1 trillion. Baker thought the administration had a coherent theory and would adapt within six to nine months if it failed, because “the one thing they can’t afford is a recession.”

Signalgate and the CECOT deportations turned execution, accountability and due process into threats to the administration’s larger mandate. The Red Sea operation had material stakes—Chamath cited shipping-price increases of 30%–40% into the US and 300%–400% into Europe—but Chamath and Jason regarded sensitive planning on personal Signal devices as a preventable mistake. On deporting 238 alleged gang members, Friedberg rejected a 4%–5% innocent-error trade-off; Jason emphasized due process and the prison’s conditions, while Baker’s warning was simple: “Execution matters.”

🔗 Original source & video: The AI Cold War, Signalgate, CoreWeave IPO, Tariff Endgames, El Salvador Deportations

Autonomous hardware and scarce compute components were central technology winners. Friedberg called 2025 “the year of the robot,” citing Unitree’s $1,600 Go2 quadruped and $16,000 G1 humanoid; Baker added mainstream FSD adoption and picked high-bandwidth memory from SK Hynix and Micron as the best-performing asset. HBM, he said, is a larger share of GPU input costs than TSMC and has been sold out for two years.

Full-stack AI providers were cast as winners, while enterprise software and independent model labs face brutal economics. Baker argued that o3-style reasoning and test-time compute let a large company spend $1 million answering its most important question over six weeks, while cloud owners enjoy lower infrastructure costs than labs renting compute. Chamath called legacy SaaS the “software-industrial complex.” Jason argued Google’s Deep Research is already outperforming rivals and predicted OpenAI’s cited $57 billion valuation could be its peak, with a nonzero chance that court cases block the transfer of $157 billion in value from nonprofit to for-profit.

Chamath’s stablecoin call was that dollar-denominated stablecoins could quadruple or quintuple during 2025. He cited roughly 1.1 billion transactions and $8.5 trillion of second-quarter 2024 volume—more than twice Visa’s—and argued that removing 300 basis points of payment friction could be worth $1 trillion in the United States alone. Baker’s pushback was geopolitical: dollar rails are advantageous, but a stablecoin constellation replacing the dollar as reserve currency would be “very bad for America.”

The political forecasts favored fiscal restraint and a younger governing class while betting against Putin, neoconservatives and progressivism. Chamath saw austerity testing the case for fiscal conservatism; Friedberg contrasted Trump’s roughly 40–45-year-old cabinet with Biden’s nearly 60-year-old one. Baker predicted Europe’s rearmament would free American resources for the Pacific and induce Xi Jinping to distance China from Putin. Jason expected aggressive Trump rhetoric before negotiated deals.

The macro outlook was a barbell between an AI-driven growth boom and a low-probability banking rupture. Baker predicted at least one year above 5% real GDP growth within four years, powered by AI and deregulation; Chamath warned that 5% rates on roughly $70 trillion of aggregate Pax Americana debt can impose the dollar burden that 10% rates once did. His hedge was long CDS protection—a trade he expects to lose most of the time but that could return 100–1,000x in his scenarios; Baker said a genuine bank failure could produce 1,000–10,000x.

Friedberg’s contrarian social call was that accelerating growth could revive socialism rather than defeat it. DOGE cuts, contracting disruption and AI replacement of white-collar labor could leave large groups behind even as some billionaires become $100 billionaires and eventually trillionaires. Baker’s framing was that before AI makes money irrelevant, test-time compute means “money will matter more than it’s ever mattered before.”

A post-Lina-Khan M&A wave could consolidate autos, AI, robotics manufacturing and autonomous transportation. The panel treated Honda–Nissan as a warning for legacy OEMs caught between Tesla and Chinese producers, while Baker expected something significant at Intel and said independent frontier AI labs could go quiet as full-stack economics favor compute owners. Waymo’s San Francisco share reportedly reached 22% in 15 months—matching Lyft—making a financing, IPO or strategic transaction plausible alongside combinations involving Uber, DoorDash, Amazon, Tesla and drone-delivery operators.

🔗 Original source & video: 2025 Predictions: Tech, Business, Media, Politics!