2025 Predictions with bestie Gavin Baker

Evidence from 2025 shows that fiscal conservatives did not achieve the kind of dominant, austerity‑driven influence Chamath predicted.

1. Major 2025 legislation increased, not restrained, federal borrowing. The One Big Beautiful Bill Act made 2017 tax cuts permanent, added new tax breaks, and sharply increased funding for defense and immigration enforcement; CBO and independent estimates project it will add several trillion dollars to the debt over the next decade, even though it includes delayed cuts to Medicaid and SNAP that begin after 2026. This is a net deficit‑expanding Trumpist agenda, not austerity. (en.wikipedia.org)

2. The 2025 funding deal mostly froze existing spending and still raised long‑run deficits. Congress relied on a full‑year continuing resolution that largely extends FY2024 levels with only limited non‑defense trims and a modest defense bump; non‑partisan analysts estimated the framework would still add trillions to the debt over ten years, and only a couple of Republican fiscal hawks voted no. A one‑year nominal freeze in a high‑inflation, high‑debt context falls well short of the “austerity” Chamath described. (en.wikipedia.org)

3. Actual cuts were narrow and symbolic relative to the overall budget. The Rescissions Act of 2025 canceled roughly $9 billion (mainly foreign aid and public broadcasting) out of a multi‑trillion‑dollar budget—highly visible but fiscally minor, illustrating how limited concrete austerity has been. (en.wikipedia.org)

4. Fiscal conservatives often lost internal fights. Hard‑line Republicans did temporarily stall Trump’s megabill over debt concerns, and some opposed budget resolutions they saw as too loose, but leadership ultimately pushed the core package through; the final law still substantially increases projected deficits, indicating that growth‑and‑culture priorities overrode strict fiscal restraint. (washingtonpost.com)

5. Political salience in 2025 has centered on immigration and trade, not austerity. The year’s defining battles have been mass‑deportation policies and related protests, deployments of federal forces in U.S. cities, and aggressive tariff actions against Canada, Mexico, and China—while state budgets show case‑by‑case adjustments or impasses rather than a nationwide turn to deep spending cuts. That issue mix does not match a landscape where fiscal conservatism and austerity are the dominant political themes. (en.wikipedia.org)

Taken together, 2025 has featured louder fiscal‑hawk rhetoric and some future‑dated cuts, but no broad move to austerity or clear "day" for fiscal conservatives as the primary political winners, so the prediction is best judged as wrong.

As of November 30, 2025, the timeframe for the prediction (“by the end of 2025”) has not fully elapsed, and the 2026 U.S. midterm fields—especially for the House—are still forming. Many candidates have not yet filed, and there is no comprehensive, comparable age dataset across cycles, so we cannot reliably measure whether the 2026 candidate pool is “noticeably younger” than prior midterms.

There are visible signs of a generational-change push and some prominent younger entrants. For example, 36‑year‑old Texas state Rep. James Talarico launched a 2026 U.S. Senate campaign explicitly framed around “generational change” and presenting a youthful contrast to older Republican leadership, getting national write‑ups and viral attention.(statesman.com) Within the Democratic Party more broadly, major coverage has emphasized a post‑2024 push for younger, more relatable leaders, and the retirements of senior figures like Sen. Dick Durbin, Tina Smith, Jeanne Shaheen and others have been framed as opening the door to a younger Senate caucus.(time.com) High‑profile primary challenges such as Rep. Seth Moulton (47) running against 79‑year‑old Sen. Ed Markey in Massachusetts also center explicitly on “generational change.”(washingtonpost.com)

At the same time, many of the most important 2026 Senate prospects for Democrats—like Sherrod Brown (72), Janet Mills (79), and Roy Cooper (68)—are themselves well over 65, and reporting notes that Democrats’ hopes of regaining the Senate majority still depend heavily on these older candidates.(en.wikipedia.org) Across early candidate lists for 2026 Senate and House races, there is a mix of long‑tenured incumbents in their 60s–80s and newer figures in their 30s–40s, but no clear, quantifiable evidence yet that the overall field is younger than in prior midterm cycles.

Given (1) the remaining time in 2025, (2) incomplete candidate slates, and (3) the lack of robust cross‑cycle age data, it’s too early to definitively say whether Friedberg’s prediction about a noticeably younger emerging 2026 field has come true, even though there are qualitative signs pointing in that direction. Therefore the prediction is best classified as inconclusive at this point.

Across 2025, the balance of power in U.S. Republican and executive‑branch foreign policy clearly shifted toward the JD Vance / “America First” restraint camp and away from the traditional pro‑Ukraine, pro‑intervention neoconservatives, with multiple concrete setbacks for the neocon position.

Concrete policy/legislative setbacks for the neocon line

• JD Vance became vice president and a central architect of foreign policy, explicitly skeptical of continued Ukraine aid and favoring economic deals over long‑term security guarantees, which is at odds with neocon priorities of robust, alliance‑based military support. (reuters.com)
• The Trump–Vance administration first paused and then effectively ended U.S. funding for Ukraine’s war effort, pressuring Kyiv toward a negotiated settlement; this is a direct defeat for the neocon goal of sustained, large‑scale U.S. military assistance to Ukraine. (cnbc.com)
• Trump’s Executive Order 14169 imposed a sweeping 90‑day pause on most foreign development aid, and the Rescissions Act of 2025 cut nearly $8 billion from international assistance programs. Traditional neoconservatives have generally supported foreign aid as a tool of American influence, so these moves represent material reversals of that agenda. (en.wikipedia.org)
• On Ukraine diplomacy, an Army secretary seen as a proxy for JD Vance has led a controversial peace initiative closely aligned with Vance’s more accommodationist vision, sidelining more hawkish voices; reporting emphasizes this as evidence of Vance’s “significant influence” over policy. (theguardian.com)
• Prominent neocon figures like Liz Cheney and Bill Kristol publicly describe themselves as out of power and explicitly frame JD Vance’s rise as a repudiation of their interventionist approach inside the GOP, underscoring their loss of intra‑party influence. (breitbart.com)

Countervailing evidence and why the prediction still holds The Trump–Vance administration has pursued very hawkish actions in some theaters—major arms sales and maximalist proposals on Gaza, airstrikes in Yemen, and expanded strikes against alleged traffickers around Venezuela—showing that “America First” is not pure pacifism. (en.wikipedia.org) But these operations are framed in transactional or narrowly defined security terms, not in the neocon vocabulary of democracy promotion and open‑ended commitments, and they coexist with a decisive rollback of the flagship neocon project of this era: arming and financing Ukraine at scale.

Given (1) the elevation and clear policy impact of JD Vance and like‑minded figures, (2) the termination of Ukraine war funding and cuts/pauses to foreign aid, and (3) the explicit marginalization acknowledged by leading neocon voices, the specific prediction—that the pro‑war neocon faction would lose influence relative to a Vance‑aligned restraint camp, with concrete setbacks for their agenda in 2025—has been borne out.

The prediction is wrong because none of the specified right‑leaning outcomes materialized by late 2025. In Canada, Justin Trudeau announced his intention to resign in early January 2025, Mark Carney became Liberal leader and prime minister in March, and the 28 April 2025 federal election returned another Liberal minority government under Carney; Conservative leader Pierre Poilievre lost both the election and his own seat, so Trudeau did not lose massively to Poilievre and Canada did not shift to a right‑wing government. (en.wikipedia.org) In Germany, the February 2025 federal election made the CDU/CSU the largest bloc, AfD finished second with about 20–21 percent of the vote, and a CDU/CSU–SPD grand coalition installed Friedrich Merz as chancellor in May, while other parties reiterated a cordon sanitaire against AfD; AfD did not enter government or “win national power.” (en.wikipedia.org) In France, Emmanuel Macron remained president throughout 2025 and no early presidential election was held; instead, Marine Le Pen was convicted in March 2025 over misuse of EU funds and given a five‑year ban on holding office, making it impossible for her to become president in 2025, while late‑2025 polling focused on Jordan Bardella as the far‑right favourite for the regular 2027 race. (en.wikipedia.org) In the United Kingdom, Labour won the July 2024 general election and Keir Starmer has served continuously as prime minister since 5 July 2024; as of November 2025 the Starmer ministry is still in office, and Nigel Farage remains leader of Reform UK in opposition, publicly pressing Starmer rather than replacing him, so the Labour government has not fallen and Farage has not taken national leadership. (en.wikipedia.org) Since all four concrete electoral predictions failed and progressive or centrist parties still govern in Canada, France and the UK, the broader claim that 2025 would bring a string of right‑leaning electoral victories and a major repudiation of progressive, identity‑politics‑oriented parties in the G8 is contradicted by actual outcomes by late 2025. (en.wikipedia.org)

Evidence from 2025 shows a clear surge in humanoid-robot R&D, investment, and pilot deployments, but not the level of mainstream commercial use and broad public recognition as a major platform that the prediction requires.

Surge and visibility (prediction gets this part roughly right):

• Multiple outlets explicitly described 2025 as a pivotal year for humanoid robots. Wired called it the year that multipurpose humanoid robots "go commercial," highlighting Boston Dynamics’ Atlas entering a Hyundai factory and Agility’s Digit and Figure’s robots beginning paid work in warehouses and other sites. (wired.com)
• Business Insider’s late‑2025 overview of six leading humanoid-robot companies notes that Agility’s Digit is already deployed commercially at GXO Logistics and Amazon, and frames humanoids as rapidly moving from science fiction toward practical reality. (businessinsider.com)
• Governments and industry coalitions treated humanoids as strategic: China’s government work report and trade-fair materials describe 2025 as the first year of mass production for humanoid robots and integrate "embodied intelligence" into national industrial policy; South Korea launched the K‑Humanoid Alliance to build a national humanoid ecosystem. (globalomp.com)
• China hosted the first World Humanoid Robot Games in August 2025 with 500+ robots competing, and conferences like the Humanoids Summit expanded globally, underscoring growing public and investor visibility. (en.wikipedia.org)

But deployment is not yet mainstream commercial use (prediction fails here):

• The scale of real-world deployments remains small and experimental. GXO’s own disclosures describe its Digit deployment at a Spanx warehouse as the first commercial application of a humanoid robot and emphasize that only a small fleet is in use; Business Insider notes GXO has just two Digit units active and that widespread deployment "is not imminent." (therobotreport.com)
• A 2025 industry white paper on humanoid applications characterizes 2025 as a “critical starting phase” for industrial deployment, but explicitly projects 2027 as the year of small‑batch production and 2030 as the starting point for real commercialization, implying that 2025 is still pre‑mainstream. (sohu.com)
• Chosun Ilbo’s 2025 analysis likewise calls this the inaugural year of mass production but quotes Unitree’s CEO predicting a "ChatGPT moment" for humanoids 2–3 years in the future, and cites Goldman Sachs’ forecast of a relatively small $1.5B market in 2025 versus $37.8B by 2035—again suggesting that broad, mainstream commercial penetration has not yet arrived. (chosun.com)
• Even where commercial pilots exist (e.g., Digit at Amazon and GXO, Apptronik’s Apollo at Mercedes factories, Kepler’s K2 robots entering mass production), the coverage consistently describes them as early trials or milestones rather than ubiquitous, routine tools across warehouses, factories, or consumer settings. (reuters.com)

Public recognition as a “major technology platform” is still emerging, not achieved:

• Media, investors, and policymakers are heavily focused on humanoids, and 2025 saw new summits, national alliances, and even international robot games, which indicates rising awareness. (en.wikipedia.org)
• Yet the same industry reports and news pieces frame humanoids as on the verge of breakthrough rather than already widely adopted—explicitly forecasting a ChatGPT‑like popularization in several years and placing true large‑scale commercialization closer to the late 2020s. (chosun.com) There is no evidence that the general public in 2025 sees humanoid robots as a mature, mainstream platform on par with smartphones or even consumer AI assistants.

Because by late 2025 humanoid robots are high-visibility but still in early, small‑scale commercial pilots, and industry consensus pushes genuine mainstream deployment and broad public recognition into the later 2020s, Friedberg’s stronger claim—that by year‑end 2025 humanoid robots would be in mainstream commercial use cases and widely recognized by the public as a major new technology platform—has not come true.

1. Volume didn’t 4–5x vs 2024

• Baseline 2024:* The joint Artemis–Dune report State of Stablecoins 2025 shows that stablecoins facilitated about $35 trillion in total transfers over 2024.(holder.io) Separate summaries of the same report repeat this ~$35T 2024 figure.(cryptonews.net) PANews notes that USD‑denominated stablecoins are ~99% of the market, so this is essentially all USD stablecoin volume.(panewslab.com)

Data for 2025 so far: Multiple 2025 analyses based on Artemis / Dune and a16z data report that:

• Adjusted stablecoin payment volume in 2025 is about $9T, up ~87% year‑on‑year.(kucoin.com) That implies 2024 adjusted volume around $4.8T — roughly a 1.9x increase, not 4–5x.
• SQ Magazine, summarizing 2025 data, says “adjusted annual stablecoin transaction volume reached approximately $9 trillion in 2025, up about 87% year‑on‑year,” and that gross transaction flows over the last 12 months are about $46T.(sqmagazine.co.uk) Relative to the ~$35T 2024 gross figure, that’s roughly a 1.3x increase.

The prediction was for 4–5x growth in aggregate transaction volume vs 2024. By late November 2025, the best on‑chain and adjusted metrics show somewhere between ~1.3x (gross) and ~1.9x (adjusted) growth, orders of magnitude below 4–5x. Even if December 2025 were unusually high, it cannot plausibly close the gap to 4–5x relative to 2024.

2. USD stablecoins did not materially erode Visa/Mastercard’s card‑payments duopoly by 2025

Card network performance in 2025:

• Visa’s FY2025 results show net revenue up 11%, EPS up 14%, and total payments volume of about $14T, up 8% YoY, with processed transactions up 10%.(finance.yahoo.com) These are strong, not eroded, economics.
• Mastercard’s 2025 results to date show similarly robust growth: Q3 2025 net revenue up 17% YoY, net income up 20%, gross dollar volume up 9%, and switched transactions up 10%.(finance.yahoo.com) Again, this indicates expanding volumes and healthy margins.

How industry analysts and the networks themselves describe the impact:

• A mid‑2025 Jefferies analysis concludes that stablecoins do not pose a material threat to Visa or Mastercard, framing the risk as more perceived than real.(investing.com)
• A separate report on Visa and Mastercard’s earnings notes that the networks "report stablecoins have minimal market impact" so far, with usage still “minimal” relative to their overall payment volumes.(coinlive.me)
• Reuters’ Breakingviews column in July 2025 acknowledges rising concern about future competition from stablecoins but emphasizes that Visa and Mastercard still process roughly 70% of U.S. purchases, with resilient earnings; any fee pressure is characterized as a potential long‑term vulnerability, not present‑day erosion.(reuters.com)

Meanwhile, mainstream coverage of stablecoins stresses that, despite explosive on‑chain volumes and surpassing card networks in raw transfer value, much of that flow is trading, arbitrage, or other crypto‑internal activity rather than consumer/merchant card‑like payments. The a16z and Artemis/Dune reports themselves distinguish between “gross transaction flows” and a much smaller subset that qualifies as actual payments, which they put on the order of single‑digit trillions annually, comparable to systems like ACH or PayPal, not yet displacing card rails.(kucoin.com)

So by late 2025:

• Visa and Mastercard’s market share of card spending remains dominant, with U.S. card spending in 2024 exceeding $10.7T and still growing.(rss.globenewswire.com)
• Their revenues and profits are up, not down, and independent equity research explicitly says stablecoins are not yet a material threat.(investing.com)

That is very different from “materially eroding the market share and economics” of the Visa/Mastercard card‑payments duopoly by the end of 2025.

3. Summary

By the end of 2025 (with 11 months of data in hand):

• USD stablecoin transaction volumes have grown strongly but are up on the order of 30–90% vs 2024, not the 300–400% (4–5x) Chamath predicted.
• Visa and Mastercard’s card‑payment volumes, revenues, and profits have continued to grow, and mainstream financial analysis characterizes stablecoin impact as minimal so far, not materially erosive.

Taken together, both core components of the prediction are not borne out by the data available as of November 30, 2025, so the prediction is wrong.

It’s too early to evaluate this prediction as of November 30, 2025, because:

• The claim is specifically about "measured over full‑year 2025" dollar‑denominated stablecoin transaction volume vs Visa’s global payment volume. Full‑year 2025 data for both:
  • On‑chain stablecoin settlement volumes, and
  • Visa’s FY2025 global payment volume will not be fully reported until calendar Q1–Q2 2026.
• Public data currently available (on‑chain stablecoin analytics and Visa’s most recent quarterly/annual filings) only cover periods up to part of 2025, not the entire calendar year, and therefore cannot confirm whether the full‑year 2025 condition is satisfied.
• Even if mid‑2025 or year‑to‑date numbers suggest a certain trend (e.g., that stablecoin settlement volumes are on track to exceed or not exceed Visa’s payments volume), that would still be projection, not a completed measurement of the whole year.

Because the prediction is explicitly about full‑year 2025 totals, and those totals are not yet available in official or widely accepted datasets, the correctness of the prediction cannot yet be determined.

The prediction is framed to be evaluated "by the end of 2025," and today is November 30, 2025, so the deadline has not yet passed and market caps could still change materially in December. Current evidence indicates that the Magnificent Seven’s combined market cap peaked around $18.2T in 2024 (e.g., on Dec. 11, 2024). (benzinga.com) In 2025 the group has actually exceeded that level, reaching record highs around $20.9T–$21.1T in October 2025, well above the 2024 peak rather than at least $2T below it. (linkedin.com) However, because the year has not yet ended, we cannot definitively say whether a $2T+ drawdown from the 2024 peak will or will not exist by December 31, 2025, so the prediction is currently too early to judge.

Evidence on this multi‑part prediction is mixed:

• Legacy primes’ stock performance in 2025:

  • Boeing (BA): After a steep 32% drop in 2024 and an additional 9% slide early in 2025, Boeing’s shares recovered somewhat; as of late November, they are only up “mid‑single digits” year‑to‑date and remain volatile, with ongoing regulatory and production issues. This lags the S&P 500’s ~16% gain in 2025. (barrons.com)
  • Lockheed Martin (LMT): By early July, LMT was down about 4% YTD and by mid‑October it had a one‑year return of around –19%, materially underperforming the S&P 500. Lockheed also cut its 2025 profit outlook due to F‑35 technology‑upgrade delays and losses on fixed‑price classified programs, clear business headwinds. (benzinga.com)
  • RTX (Raytheon): In contrast, RTX has had an excellent 2025. StatMuse and Macrotrends data show its 2025 return in the ~37–50%+ range, far ahead of the market. The company reports a record ~$218 billion backlog and “unprecedented” missile demand, with 4–6% expected sales growth in 2025. (statmuse.com)
    Net: Two of the three cited primes (BA, LMT) did face notable headwinds and at best modest or negative stock performance, but RTX is a major counterexample, enjoying strong fundamentals and outsized returns rather than being “deeply challenged.”

• Performance of newer tech‑driven defense firms:

  • Palantir (PLTR): Multiple sources show PLTR as one of the best‑performing U.S. defense‑adjacent stocks in 2025. A Benzinga comparison table has Palantir up ~79% YTD by early July, versus Boeing ~26%, RTX ~26%, and Lockheed –4%. (benzinga.com) Later analysis notes Palantir up ~75% YTD by early July and roughly 300% YTD by late October, vastly outpacing both the legacy primes and the S&P 500. (nasdaq.com) Over the past year through November, PLTR is up about 150%. (investing.com)
  • Anduril: Anduril raised about $2.5 billion in 2025 at roughly a $30–30.5 billion valuation and is reported to have more than $2 billion in active government contracts. It won high‑profile programs such as Australia’s A$1.7 billion Ghost Shark XL‑AUV fleet, is a contractor on the U.S. Space Force’s Golden Dome missile‑defense effort, and its Fury (YFQ‑44) drone is one of the winning designs in the USAF’s Collaborative Combat Aircraft program. (businessinsider.com) These wins and its funding trajectory support the idea that tech‑centric “new primes” are gaining momentum.

• Shift in defense spending and contracting patterns:

  • Policy and funding signals clearly favor AI‑ and software‑heavy solutions. Commentators note firms like Palantir and Anduril winning multi‑hundred‑million‑dollar data, AI, and autonomy contracts and becoming central to a “new breed” of defense tech suppliers. (aoav.org.uk) Venture funding into defense tech startups roughly doubled in 2025 to over $19 billion, driven by Pentagon modernization priorities around AI, autonomy, and software‑defined systems. (businessinsider.com)
  • However, at Baird’s 2025 Defense & Gov’t conference, Anduril’s own leadership emphasized that the Department of Defense still isn’t procuring at scale from new entrants—most of their work is in prototypes and early fielding, not yet in large production programs with the same volume and stability that legacy primes enjoy. (wash100.com) This suggests the “swing” of spending away from incumbents toward startups is underway but far from complete.

Putting it together:

• The relative stock‑performance call—that newer tech‑driven defense companies like Palantir would dramatically outperform legacy primes—did happen in 2025. Palantir’s returns tower over Boeing, Lockheed, and even the strongly performing RTX. (benzinga.com)
• The claim that legacy primes would be “deeply challenged…across the board” is only partially supported. Boeing and Lockheed faced real operational and earnings headwinds plus subpar share performance, but RTX experienced record demand, strong financial results, and very strong stock gains—directly contradicting a blanket “headwinds” thesis. (marketwatch.com)
• The idea of a wholesale, near‑term shift of government defense spending away from primes toward ROI‑driven startups is also overstated. While policy, venture funding, and some major contracts favor AI/autonomy players, official statements from both the Pentagon and new‑entrant executives indicate that large‑scale production dollars still overwhelmingly flow to the traditional primes, with new firms mostly in prototype and early deployment stages. (defensenews.com)

Because the prediction bundles together (1) a relative stock‑market bet that largely worked, (2) an industry‑wide “deep challenge” claim that is contradicted by RTX’s strong 2025, and (3) a structural spending‑shift claim that is only partially and gradually playing out, the overall truth value depends heavily on which component you weight most. The data are sufficient, but they support some parts and clearly contradict others, so the outcome for the prediction as stated is best characterized as ambiguous rather than cleanly right or wrong.

Jason argued that OpenAI’s roughly $157 billion private‑market valuation achieved in 2024 would mark the company’s peak and that its enterprise value would be materially lower than that level by the end of 2025. That peak claim has already been falsified.

1. In late 2024, OpenAI completed a funding and tender offer led by SoftBank that valued the company at about $157 billion; this is the reference valuation Jason was talking about. (cnbc.com)
2. In 2025, OpenAI went on to close a much larger SoftBank‑led round/tender offer that valued the company at roughly $300 billion. (washingtonpost.com)
3. Later in 2025, OpenAI conducted secondary share sales for employees at a valuation of about $500 billion, expanding the sale size to over $10 billion while keeping that $500 billion pricing. (cnbc.com)

Because OpenAI’s private‑market valuation subsequently rose first to around $300 billion and then to about $500 billion, the ~$157 billion 2024 valuation clearly did not prove to be the peak. That is enough to render the prediction wrong, regardless of where OpenAI’s value ends up on December 31, 2025.

OpenAI spent 2024–25 working with regulators on a plan to convert its capped‑profit subsidiary into a standard equity structure and for‑profit public benefit corporation (PBC), while facing lawsuits and scrutiny over the treatment of its nonprofit assets. (arstechnica.com)

On 28 October 2025 OpenAI announced and implemented a new structure: the original nonprofit became the OpenAI Foundation and the operating business became OpenAI Group PBC, a for‑profit corporation with ordinary stock, majority investor ownership and a path to an eventual IPO, while the foundation retained a large equity stake and the power to appoint and remove all directors. (openai.com) Regulators in Delaware and California formally approved this restructuring, and major outlets reported that OpenAI had completed its conversion to a for‑profit PBC. (apnews.com)

Although litigation and policy debates continue, no court has invalidated the restructuring, and the new for‑profit PBC structure is in force as of late 2025. (theverge.com) That means OpenAI was in fact able to complete a legally approved transition to a conventional for‑profit corporate vehicle during 2025, so Friedberg’s prediction that it would not be able to convert from nonprofit to for‑profit in 2025 is contradicted by events.

By November 30, 2025, there has not been a wave of auto mega‑mergers among large legacy OEMs.

• The one clear example pointing in that direction was the proposed Honda–Nissan (with possible Mitsubishi participation) mega‑merger. On December 23, 2024, they announced an in‑principle agreement to pursue a merger under a holding company by 2026, explicitly framed as a response to electrification pressures and competition from Tesla and Chinese EV makers like BYD. Analysts and trade press repeatedly described this as a mega‑merger aimed at surviving EV disruption and achieving scale. (trendforce.com)
• However, those talks collapsed quickly. By February 2025, Honda, Nissan, and Mitsubishi formally ended negotiations and terminated their integration framework; Nissan’s board rejected the proposed terms, and the companies reverted to looser strategic cooperation instead of merging. Nissan’s own summary of events notes that the planned 2026 merger was abandoned in February 2025. (upi.com)
• Crucially, no further comparable mergers among major traditional automakers (GM, Ford, Toyota, VW Group, Stellantis, Hyundai–Kia, etc.) were announced or completed in 2025. Contemporary analysis around the Honda–Nissan plan even argued that such deals are rare and were unlikely to trigger a broader merger mania, which is exactly what played out. (forbes.com) Instead, we saw partnerships and JVs (e.g., GM–Hyundai co‑development, VW–Rivian technology JV) rather than corporate mergers. (reuters.com)
• Some consolidation occurred inside Chinese groups (e.g., Geely fully taking Zeekr private), but these are intra‑group restructurings in China, not the cross‑company mega‑mergers of Western/Japanese legacy OEMs that Chamath was describing. (ft.com)

Since 2025 produced at most one short‑lived mega‑merger attempt and no sustained, multi‑firm wave of auto mega‑mergers or clear “collapse of the traditional auto OEMs,” the prediction that 2025 would bring such a wave is best judged as wrong.

By late 2025, multiple large‑scale, multi‑billion‑dollar packages for U.S. hardware manufacturing had been announced, matching the prediction’s direction. In April 2025, the White House reported over $5 trillion in new U.S.-based investments in Trump’s first 100 days, emphasizing a series of blockbuster advanced‑manufacturing commitments: Apple’s $500 billion U.S. manufacturing and training plan, TSMC’s additional $100 billion for U.S. chip manufacturing, Johnson & Johnson’s $55 billion, Novartis’s $23 billion to build or expand ten U.S. plants, and Hyundai’s $21 billion U.S. investment including a $5.8 billion new steel plant—each framed as major U.S. manufacturing build‑outs rather than purely digital or services spending.(whitehouse.gov)

In parallel, semiconductor‑focused industrial policy continued to drive enormous hardware manufacturing capex. By 2025 the Commerce Department had distributed over $30 billion in CHIPS Act subsidies across 17 firms (including Intel, TSMC, Samsung, Micron), catalyzing more than $100 billion in new U.S. fab projects.(ainvest.com) TSMC specifically expanded its Arizona plans to a $65+ billion complex and, alongside earlier awards, received a $6.6 billion U.S. subsidy plus up to $5 billion in loans, while separately announcing an additional $100 billion U.S. chips‑manufacturing program and a multi‑fab Arizona “gigafab” cluster.(investing.com) These are clearly multi‑billion financing/subsidy‑linked build‑outs of physical manufacturing capacity in the U.S.

On the robotics/“physical AI” side, 2025 also saw U.S.-based manufacturing expansions tied directly to robotics: Figure AI announced BotQ, a high‑volume U.S. facility intended to manufacture 12,000 humanoid robots per year; Nvidia and Foxconn launched a new Houston AI manufacturing campus where humanoid robots will assemble Nvidia AI servers; and SoftBank unveiled “Project Crystal Land,” a proposed trillion‑dollar‑scale AI and robotics industrial hub in Arizona in partnership with TSMC, explicitly framed as a U.S. manufacturing center for AI and robotics hardware.(en.wikipedia.org) While not every robotics project discloses a precise multi‑billion capex figure, they are part of the same wave of highly publicized industrial‑scale hardware and robotics build‑outs.

Taken together, these 2025 announcements constitute several multi‑billion‑dollar financing and subsidy–driven build‑outs of U.S. hardware (and related robotics) manufacturing. Their headline sizes (tens to hundreds of billions, with some projects framed at the trillion‑dollar level) and the way they dominate policy and business news place them in the same general league of scale and prominence as the 2023–24 AI compute megaprojects, so Friedberg’s prediction is best judged as having come true in substance.

Evidence from 2025 shows that at least one major strategic partnership matching Jason’s criteria has already been announced, so the prediction is satisfied even before year‑end.

On 16 October 2025, DoorDash announced a new strategic partnership with Waymo to launch an autonomous delivery service in Metro Phoenix, using Waymo’s fully autonomous vehicles integrated into DoorDash’s AutonomousDelivery Platform. This is explicitly framed as a large‑scale autonomous delivery rollout for food and groceries and includes a multi‑city promotion for DashPass members using Waymo’s ride‑hailing service in Los Angeles, San Francisco, and Phoenix. (businesswire.com) DoorDash is a leading food‑delivery/e‑commerce logistics platform, and Waymo is a top autonomous‑vehicle company, so this partnership fits the predicted pattern: a major autonomy player teaming with a major logistics/food‑delivery platform to integrate autonomous delivery at scale.

Separately, Uber and Waymo expanded their existing partnership to bring fully autonomous ride‑hailing, via Waymo’s driverless Jaguar I‑PACE fleet, onto the Uber app in Austin and Atlanta starting in early 2025—again, a large autonomy provider integrating with a large‑scale ride‑hailing/logistics platform. (businesswire.com) Uber has also struck a large, multi‑hundred‑million‑dollar robotaxi deal with Lucid and Nuro to deploy tens of thousands of autonomous vehicles on its platform over the next several years. (theverge.com) These additional deals further reinforce that the type of partnership Jason described is in fact occurring.

While the specific speculative examples he mused about (e.g., Tesla buying DoorDash or Uber) have not happened, his normalized prediction—that by the end of 2025 at least one major strategic partnership or M&A deal among the big autonomy/logistics players would be announced to integrate AVs with large‑scale e‑commerce, food, or ride delivery—has clearly come true, most directly via the DoorDash–Waymo partnership.

Available reporting through late 2025 shows no major new capital or corporate event for Waymo in 2025 of the type Friedberg described (large external financing round, IPO, or merger/acquisition).

• Waymo’s last major capital raise was a $5.6 billion Series C funding round announced on October 25, 2024, led by Alphabet with participation from outside investors such as Andreessen Horowitz, Fidelity, Silver Lake, Tiger Global, and T. Rowe Price. This is consistently dated to October 2024, not 2025. (cnbc.com)
• Follow‑on coverage in 2024–2025 continues to reference that same $5.6 billion Series C as Waymo’s big round, putting its valuation above $45 billion, with no subsequent 2025 financing round disclosed. (gurufocus.com)
• Throughout 2025, news about Waymo focuses on geographic expansion and partnerships (e.g., expansion of robotaxi services in Silicon Valley, Austin, and Miami; partnerships with Uber and Lyft in markets like Austin and Nashville), but these are commercial/operational agreements, not mergers or acquisitions. (en.wikipedia.org)
• As of late November 2025, major outlets still describe Waymo as Alphabet’s self‑driving unit/subsidiary, indicating no IPO or spin‑off and no sale to or merger with another company. (axios.com)

Because the prediction was specifically “this year” (2025) and pointed to a new massive financing, IPO, or M&A event, and none of those occurred in 2025 (the big funding round happened in 2024), the prediction is best scored as wrong.

Evidence from 2025 shows that Jason’s prediction did occur.

On October 16, 2025, DoorDash announced a new partnership with Waymo to launch an autonomous delivery service in Metro Phoenix, plus a Waymo ride promotion for DashPass members in Los Angeles, San Francisco, and Phoenix. The announcement explicitly describes this as a “new partnership” between the two companies focused on using Waymo’s autonomous vehicles for DoorDash deliveries, with broader commercial operations planned for later in 2025. (ir.doordash.com)

Waymo’s own blog and multiple tech/business outlets likewise describe the deal as a partnership where DoorDash customers can opt in to have orders delivered by fully autonomous Waymo vehicles, initially from DoorDash’s DashMart convenience and grocery locations, with plans to expand to more merchants. These reports emphasize that this collaboration is part of DoorDash’s and Waymo’s broader strategies to automate local commerce and reshape last‑mile logistics, and some characterize it as the first collaboration between the two firms. (waymo.com)

This satisfies the normalized prediction criteria:

• It occurred in 2025.
• It involves two of the specified “cohort” of large autonomy/e‑commerce/logistics players: DoorDash and Waymo.
• The partnership is explicitly centered on autonomy and delivery/food delivery (autonomous vehicles handling DoorDash orders), making it a clear strategic partnership in the autonomy/logistics space.

Therefore, the prediction that at least one major strategic partnership among that group (Amazon, DoorDash, Uber, Tesla, Waymo and similar players) would form in 2025 around autonomy, delivery, food delivery, or e‑commerce is fulfilled.

Available evidence shows that Waymo did not complete a new “major capital markets transaction” in 2025.

Waymo’s last large external raise was a $5.6 billion Series C funding round that closed on October 25, 2024, led by Alphabet with participation from major institutional investors; this round is consistently described in contemporary coverage and transaction records as a 2024 deal, not 2025. (cnbc.com) Overviews of Waymo’s financing history as of March–June 2025 list total funding of about $11.1 billion and identify that 2024 Series C (plus earlier 2020 and 2021 rounds) as the latest funding, with no additional 2025 equity or debt rounds recorded. (en.wikipedia.org) Analysts discussing a possible spinoff or separation from Alphabet in 2025 treat it purely as speculation and explicitly reference the 2024 Series C as the most recent capital raise. (forbes.com)

At the same time, 2025 news about Waymo focuses on operational expansion and commercial partnerships (e.g., launching services via Uber and Lyft in cities like Austin and Nashville), but these are structured as service/operations agreements, not mergers or acquisitions of, or by, major ride‑hailing companies. (en.wikipedia.org) Waymo also remains a wholly owned (and non‑public) subsidiary of Alphabet throughout 2025, with no IPO or spin‑off completed by November 30, 2025. (en.wikipedia.org)

Because the prediction specifically required that in 2025 Waymo would execute a major capital‑markets event (large financing, IPO, or M&A with a big ride‑sharing company), and no such 2025 transaction occurred, the prediction is wrong.

It’s too early to evaluate this prediction based on the timeframe given.

Jason’s claim is that U.S. ridesharing’s share of total rides (the overall ride/transportation TAM) will rise from roughly ~1.5% to about 20% in a “very short period of time” / “within a few years” from early 2025. As of today (Nov 30, 2025), fewer than one year has elapsed since the prediction was made (Jan 4, 2025). Even if we interpret “very short period of time” aggressively as 2–3 years, that window clearly extends well beyond 2025, so we don’t yet know whether the prediction will be correct.

Public data also doesn’t yet show anything close to 20% of all U.S. trips being rideshare. Most available sources only estimate rideshare as a low‑single‑digit share of overall passenger miles or trips in the U.S. in the early‑to‑mid‑2020s, and none indicate a sudden jump to anything near 20% of all rides by 2025. But because the prediction’s horizon is “within a few years” after 2025, the relevant deadline has not arrived, so we can’t label it right or wrong yet.

Given that:

• The prediction explicitly depends on a multi‑year future period starting in 2025, and
• We are still inside that period as of Nov 30, 2025,

…the only fair status is “inconclusive (too early)” rather than right or wrong.

As of November 30, 2025, no widely‑recognized major U.S. "mainline" bank (e.g., JPMorgan Chase, Bank of America, Citigroup, Wells Fargo, Goldman Sachs, Morgan Stanley, etc., as listed among the largest U.S. banks by assets) has experienced an outright banking crisis involving failure, FDIC resolution, or an acute, public liquidity/solvency event driven by reserve or balance‑sheet stress in 2025.(en.wikipedia.org)

Public indicators instead show:

• Only two very small banks (Pulaski Savings Bank and The Santa Anna National Bank, each with <$100M in assets) have failed in 2025, per the FDIC list of bank failures; no large or systemically important bank appears on that list.(en.wikipedia.org)
• The Federal Reserve’s 2025 stress tests report that all 22 of the largest U.S. banks remained above regulatory capital minimums even under severe recession scenarios, implying no immediate capital or solvency crisis at the big "mainline" institutions.(ft.com)
• Disclosures from individual majors (e.g., Citigroup’s Q1 2025 call) describe strong capital and liquidity positions, with CET1 ratios comfortably above requirements and substantial available liquidity resources.(alphaspread.com)
• System‑level commentary and forecasts (e.g., Swift Centre’s probabilistic assessment) treat a major‑bank liquidity crisis as a low‑probability future risk, not as an event that has already occurred.(swiftcentre.org)

However, Chamath’s prediction window runs “by the end of 2025.” Since there is still one month remaining and no such major‑bank crisis has occurred yet, the prediction has not been confirmed, but it also cannot yet be definitively ruled out. Therefore, the fairest status as of today is inconclusive (too early to tell) rather than clearly right or wrong.

Evidence from 2025 strongly indicates that socialist movements in the U.S. did grow substantially in both membership and political influence relative to 2024.

1. Organizational membership (breadth): The Democratic Socialists of America (DSA), the largest socialist organization in the U.S., saw membership fall to about 64,000 by October 2024, but then experience a sharp rebound to over 80,000 by October 2025 and over 90,000 by November 2025—roughly a 40% increase from the 2024 trough in a single year. (en.wikipedia.org) A September 2025 analysis of the DSA convention likewise reports the group at “more than 80,000 members” and on track to possibly exceed 100,000. (solidarity-us.org)

2. Youth and campus organizing: Young Democratic Socialists of America (YDSA) reported 47% growth in chapter membership since November 2024, indicating rapid expansion among student and youth organizers. (notus.org) An October 2025 Axios/Generation Lab poll found U.S. college students now view socialism more favorably than capitalism (34% vs. 17%), suggesting a widening base of young sympathizers. (axios.com)

3. Electoral and political influence (depth): DSA‑aligned officeholders expanded from “over 200 elected officials” in early 2024 to more than 250 by 2025. (en.wikipedia.org) NYC‑DSA and DSA nationally were central to Zohran Mamdani’s 2025 New York City mayoral win, widely described as the biggest electoral victory in DSA history and a major boost to democratic socialist influence in a flagship U.S. city. (en.wikipedia.org)

Taken together—large year‑over‑year membership gains from a 2024 low point, strong growth among youth and campus groups, and increased electoral clout—the data support Friedberg’s prediction of a “dramatic rise” in socialist movements in the U.S. during 2025 compared with 2024.

As of 30 November 2025, the prediction’s deadline (“by the end of 2025”) has not yet passed, so it’s too early to give a definitive right/wrong verdict.

What we can say so far:

• Enterprise value / market position: OpenAI completed a large employee share sale in October 2025 that valued the company at about $500 billion, making it the most valuable private company in the world, ahead of other major AI labs such as Anthropic (~$183B) and xAI (~$80B). This clearly does not place OpenAI in fourth position by enterprise value among AI-focused labs; it is currently #1 by that metric. (en.wikipedia.org)
• Technical / market leadership: Multiple industry snapshots still describe OpenAI as an AI or generative-AI leader. A Statista-based summary from May 2025 reports that ChatGPT (OpenAI) together with Microsoft Copilot hold roughly three‑quarters of the global generative AI chatbot market, far ahead of rivals like Google Gemini, Claude, and others. (mtlc.co) Gartner’s 2025 Innovation Guide lists OpenAI as an “Emerging Leader in Generative AI,” with more than 1 million companies using its products. (openai.com)
• Competitive pressure (but not a collapse): Several analyses do suggest that Anthropic has overtaken OpenAI in certain enterprise segments, with estimates around 32% enterprise LLM share for Anthropic vs ~25–34% for OpenAI, and recent Claude Opus releases beating OpenAI and Google models on some coding benchmarks. (businessinsider.com) Even under those more aggressive Anthropic-favorable numbers, OpenAI looks like a close #2, not a distant #4, and certainly not in a “total collapse.”

Putting this together: as of late November 2025, OpenAI has not lost its overall market/valuation primacy among AI labs nor fallen to “at least fourth place” on any broad, objective ranking by enterprise value or general technical stature, and it is still widely described as a leading AI company. However, because the prediction explicitly runs through 31 December 2025, it might still be fulfilled or definitively falsified in the remaining time. That’s why the current status is best labeled “inconclusive (too early)” rather than clearly right or wrong.

As of November 30, 2025, there is no record that Jeff Bezos has sold The Washington Post. Major coverage of the paper throughout 2025 continues to describe it as owned by Jeff Bezos, including Reuters on January 7, 2025, reporting layoffs at “the organization, owned by Amazon founder Jeff Bezos,” and referring to no change in ownership. (reuters.com) The Washington Post’s own February 26, 2025 story about shifting its opinion section explicitly calls Bezos the paper’s owner with no mention of a sale. (washingtonpost.com) Business Insider’s May 2025 coverage of buyouts similarly treats Bezos as the current owner during restructuring. (businessinsider.com)

In addition, reporting on speculation about a possible sale consistently notes that Bezos or his representatives say the Post is not for sale, including a New Yorker article in May 2025 stating he recently told an interested buyer that it is not for sale, and earlier coverage of Commanders-related rumors where a Bezos spokesperson said the Post is not for sale. (newyorker.com) There is no credible report of any completed sale in 2025, which would almost certainly attract extensive media coverage.

However, because the prediction specifies “during calendar year 2025” and the year has not yet ended as of the current date (November 30, 2025), it is still theoretically possible—though seemingly unlikely—that a sale could occur in December 2025. Therefore, with the deadline not yet passed, the prediction cannot be definitively graded as right or wrong and is best classified as inconclusive (too early to tell).

Chamath’s claim was conditional: if there were severe credit stress or a banking crisis in 2025, being long selected CDS on vulnerable credits would be the best‑performing asset of 2025; otherwise that CDS trade would “lose money.”

1. Trigger condition (severe credit stress / banking crisis) is not clearly met or clearly absent.

   • 2025 did see a sharp but tariff‑driven equity crash in early April, with the S&P 500 and Nasdaq suffering their biggest declines since 2020, but markets largely recovered by May. This episode is generally described as a stock market crash linked to trade policy, not a banking‑system meltdown.
     • (en.wikipedia.org)
   • Credit markets did come under notable stress: CDS spreads on high‑yield and crossover indices jumped to multi‑year highs in early 2025, and Asia/Europe credit indices and sovereign CDS widened materially around the tariff shock, reflecting increased default concerns.
     • (ft.com)
   • However, major official and media sources through late 2025 talk about rising risks and regional bank/CRE stress, not an actual systemic banking crisis. The Russian central bank explicitly said there was no looming banking crisis, and IMF commentary on AI and markets specifically distinguished current stresses from 2008‑style systemic events. (reuters.com)
   • In the U.S., only two small banks failed in 2025 as of late June, far from a broad wave of failures. (en.wikipedia.org)
   • Because terms like “severe credit stress” and “banking crisis” are qualitative, it is debatable whether 2025 meets Chamath’s threshold. Reasonable observers could disagree, so we cannot definitively say the if‑crisis branch of his prediction did or did not trigger.

2. We cannot observe the performance of the specific CDS trade he had in mind.

   • Public data confirm big swings in CDS spreads (which would have made some long‑CDS positions very profitable during the April stress), but spreads partially retraced as markets stabilized. (kelo.com)
   • Without knowing which credits he meant by “vulnerable,” the exact timing of entry and exit, or whether the position was held all year, we cannot say whether his particular long‑CDS trade ended up making or losing money over full‑year 2025.

3. “Best‑performing asset of 2025” is not practically checkable for CDS.

   • Many listed assets in 2025 (e.g., certain AI/tech and defense stocks/ETFs, as well as some high‑beta names) show very high year‑to‑date returns, often 50–200%+ by late 2025. (onedayadvisor.com)
   • CDS trades are over‑the‑counter, highly bespoke, and not reported in a way that allows a clean, market‑wide ranking of “best‑performing asset.” Even if some niche CDS position had enormous percentage returns, there is no public, comprehensive dataset that would let us verify that it outperformed all other assets.

Because (a) whether the crisis condition truly occurred is subjective, (b) the exact CDS instruments and P&L for Chamath’s trade are unknown, and (c) there is no way to rigorously rank that CDS position against all other assets in 2025, the prediction cannot be conclusively scored as right or wrong based on available evidence.

Multiple broad Chinese tech / internet benchmarks had strong double‑digit gains in 2025 and clearly outperformed many other equity sectors.

• The KraneShares CSI China Internet ETF (KWEB), a widely used proxy for Chinese tech/internet stocks, shows a YTD return of 35.2% (NAV) as of Oct 31, 2025.【1search4】 A separate performance commentary notes KWEB had returned about 49–50% year‑to‑date later in 2025, highlighting a very strong run.【3search2】
• Broader commentary on Chinese tech confirms a major rally: Axios reports Chinese tech stocks up over 44% year‑to‑date from their April lows, outpacing leading U.S. tech stocks and the “Magnificent 7.”【3news13】 The Financial Times likewise notes the Hang Seng Tech Index entered a bull market in 2025, rising about 25% from its January low and beating the Nasdaq 100, with large caps like Alibaba up ~43% and Xiaomi ~34%.【3news14】
• In contrast, the S&P 500 total‑return YTD as of Oct 31, 2025 is about 17.4%,【2search2】 and the U.S. Technology Select Sector SPDR (XLK)—the best major U.S. sector—shows a 29.9% YTD return over the same period.【2search1】 A sector breakdown for the S&P 500 shows most other sectors (healthcare, financials, materials, real estate, energy, etc.) with mid‑single to low‑20s YTD returns, or less, all below KWEB’s mid‑30s+ YTD gains.【3search0】

Taken together, these data show that broad Chinese tech equities (as represented by ETFs like KWEB and indices like Hang Seng Tech) appreciated significantly in 2025 and outperformed the S&P 500 and the majority of U.S. equity sectors, including most non‑tech sectors. That matches the normalized prediction that Chinese tech stocks would have “a pretty good run” and beat many other sectors, conditional on reasonably supportive macro conditions. Therefore, the prediction is best judged as right.

The prediction is about performance over the full calendar year 2025 (“is going to be the best performing asset”). Today’s date is November 30, 2025, so the year has not ended and final 2025 performance rankings across major asset classes (including the Magnificent 7 basket) are not yet known. Because the outcome depends on returns through December 31, 2025, it is too early to determine whether this prediction ultimately proves right or wrong.

Available 2025 data show that major incumbent enterprise‑software vendors are still growing and profitable, but they have begun to exhibit the kind of “fissures” Chamath described—slower growth, restructuring, and mounting pressure from faster‑growing AI‑native rivals—by late 2025.

On the incumbent side: Salesforce’s revenue growth has decelerated to the high‑single‑digit range (about 8–9% guidance for FY26) with slowing subscription growth across core clouds, even as margins remain high; its stock is the worst performer among large‑cap tech in 2025 and analysts explicitly worry that AI could “eat away” at SaaS business models. (investor.salesforce.com) SAP missed Q3 2025 revenue expectations, with total revenue up only 7% and its cloud business growing 22%—still solid but the slowest cloud growth since 2023—prompting guidance to the low end of its cloud‑revenue range. (reuters.com) Workday announced layoffs of about 1,750 employees (8.5% of staff) in early 2025 to refocus spending on AI and later reported “lukewarm” subscription revenue that merely met expectations, causing a share‑price drop and reflecting softer enterprise demand. (reuters.com) Adobe, despite posting ~10–11% revenue growth and record results, has been downgraded on the thesis that “AI is eating software,” with analysts highlighting intensifying competition from AI‑powered tools and newer design platforms; its shares are down materially year‑to‑date. (barrons.com) Altogether, these trends show slowing growth, workforce cuts, and increased investor skepticism across several flagship vendors, even as a few (e.g., ServiceNow, Oracle) continue to post strong AI‑driven growth and margin expansion. (reddit.com)

In contrast, AI‑native competitors are growing dramatically faster. Databricks projects about $4 billion in 2025 revenue, more than 50% year‑over‑year growth, with >140% net revenue retention and hundreds of $1M‑plus enterprise customers. (reuters.com) Sector‑wide benchmark data for 2025 indicate AI‑native startups have median annual growth around 100%, versus roughly 20–25% for traditional SaaS, and are reaching $100M ARR in 1–2 years with far higher revenue per employee. (deepstarstrategic.com) Industry analyses also describe customers delaying some SaaS renewals and experimenting with AI‑native tools, explicitly framing this as pressure on legacy software economics. (stansberryresearch.com)

Because Chamath only predicted that “fissures” would start to appear in 2025—rather than a full‑blown collapse—the combination of (a) visible growth deceleration, restructurings, and investor multiple compression at several large incumbents, and (b) much faster growth and capital allocation toward AI‑native enterprise platforms, matches the spirit of his call. The incumbents remain large and profitable, but clear early cracks relative to AI‑native competitors are observable by late 2025, so the prediction is best judged as broadly right rather than wrong or purely inconclusive.

Available 2025 data show vertical SaaS has generally been one of the stronger parts of software, not among the worst-performing, and there is no clear evidence that AI‑driven in‑house tools and pricing compression uniquely damaged its valuations or revenues.

Key points:

• Valuations and relative performance. Multiple independent valuation and M&A analyses in 2025 find that vertical SaaS companies still trade at a premium to horizontal SaaS and are viewed as especially attractive:

  • AGC Partners’ early‑2025 report shows vertical SaaS trading at about 7.0× EV/Revenue vs. 4.8× for horizontal SaaS, with similarly strong growth and margins, implying investors pay up for vertical focus rather than shun it. (agcpartners.com)
  • A Baker Tilly / AInvest summary of Q3 2025 data likewise reports vertical SaaS firms at 7.0× EV/revenue vs. 4.8× for horizontal, explicitly saying vertical SaaS companies "continue to outperform horizontal peers." (ainvest.com)
  • A 2025 valuation overview notes that AI‑first and vertical SaaS are the most sought‑after categories, often rewarded with 8–12× revenue multiples, while broader horizontal platforms without clear ROI sit around 3–5×. (ful.io)
  • A broad SaaS metrics benchmark (2,000+ companies) highlights "Vertical SaaS outperforming": vertical players are 1.5–3.3× more likely to be outlier winners and account for 54% of Q3 2025 SaaS M&A, with the conclusion "Vertical‑first wins." (rockingweb.com.au)
  • A July 2025 sector note from JM Financial contrasts vertical vs. horizontal SaaS since the 2021 peak: horizontal SaaS endured sharper multiple compression, while vertical SaaS names like AppFolio, Procore, Veeva, Samsara, and Guidewire generally maintained 5–15× EV/Sales, with "more orderly and muted de‑rating" and structurally stronger pricing power due to sticky, mission‑critical workflows. (fr.scribd.com)

• Stock and growth benchmarks do not show vertical SaaS as a bottom tier.

  • A First Analysis quarterly review does show that in the June 2025 quarter, the vertical SaaS group’s shares rose only about 5% vs. 13% for its broader SaaS universe, lagging categories like cybersecurity and data visibility. That is mild underperformance for that quarter, not "worst‑performing software assets" for the year. (firstanalysis.com)
  • The Software Equity Group’s SEG SaaS Index reports that "Vertically Focused" public SaaS names saw EV/revenue medians fall from 4.5× (3Q24) to 3.2× (3Q25) while the overall index went from 5.9× to 5.3×, indicating some relative multiple compression—but again, several other categories (e.g., analytics, sales & marketing) also sit in the low‑multiple range, so vertical is not uniquely or clearly the "worst" segment. (softwareequity.com)
  • By contrast, other 2025 performance roundups emphasize that industry‑focused, vertical platforms are among the winners: a 1H’25 SaaS stock review notes that industry‑focused platforms like Veeva and Guidewire outperformed horizontal solutions, and calls out "Vertical Specialization Advantage" as a key driver. (linkedin.com)
  • A separate SaaStr analysis, "The Vertical SaaS Gold Rush," shows many vertical or non‑tech B2B names (Samsara, ServiceTitan, Toast, Shopify, etc.) growing 2–3× faster than generic enterprise SaaS; most horizontal incumbents like Salesforce are in single‑digit growth. (saastr.com)
  • Multiple benchmark summaries (e.g., OpenView/WeBuildSaaS, LinkedIn analyses) show median growth of ~45% for vertical SaaS vs. ~28% for horizontal, with top‑quartile verticals at ~100%+ YoY growth—again describing vertical SaaS as "crushing" horizontals, not as a laggard. (webuildsaas.com)

• AI’s effect: more tailwind than systemic damage to vertical SaaS.

  • A 2025 vertical‑SMB SaaS benchmark from Tidemark concludes that "Vertical SaaS has arrived" and that AI is now a revenue driver, with companies that adopt AI in their products and GTM seeing stronger growth, rather than mass churn to in‑house tools. (tidemarkcap.com)
  • Coverage of the "third wave" of vertical SaaS (cloud + fintech + AI) from investors like a16z describes AI as expanding vertical SaaS TAM by automating labor‑intensive workflows and enabling higher net revenue retention—again positioning AI as a core growth catalyst for vertical SaaS firms such as Toast, not as a force that is broadly eroding their revenue base. (yourstory.com)
  • Two detailed essays from Reformation Partners (Jan 2024 and Nov 2025) explicitly rebut Friedberg’s thesis that internal AI‑built tools would make vertical SaaS "doomed." They argue, with examples from construction, hospitality, legal, and other sectors, that most vertical SaaS customers lack the in‑house engineering depth to build and maintain comparable tools; that internal builds carry higher ongoing risk and cost than simply buying the vertical standard; and that AI productivity gains accrue to vertical SaaS vendors themselves, allowing them to ship better products or cut prices while preserving margins. The author concludes, nearly two years into the AI boom, that the original “vertical SaaS is dead” prediction has not been borne out. (medium.com)
  • More general commentary on AI and SaaS economics in 2025 acknowledges some margin and pricing pressure across SaaS as AI features become table stakes, but does not single out vertical SaaS as uniquely impaired; instead, it typically notes that commoditization risk is greatest for generic horizontal tools, while domain‑specific, data‑rich platforms retain differentiation. (implicator.ai)

Putting this together:

• Vertical SaaS did not become "among the worst‑performing software assets" in 2025. Across valuations, M&A share, and growth metrics, it is more often described as outperforming horizontal SaaS and as a favored category for investors.
• While there is some evidence of moderate multiple compression and a few quarters of relative underperformance for certain vertical cohorts, that is part of a broader SaaS repricing and does not match the severity or uniqueness implied by the prediction.
• The specific causal story—that enterprises would rapidly adopt AI‑driven in‑house tools, causing widespread churn and pricing compression that disproportionately harms vertical SaaS—has not materialized at scale. Available data and investor commentary instead show AI being integrated into vertical SaaS products and business models, largely as a competitive advantage rather than an existential threat.

On balance, Friedberg’s prediction that vertical SaaS would be one of the worst‑performing software assets in 2025 due to AI‑driven in‑house tools and pricing compression is not supported by the observed 2025 data, so it is best classified as wrong.

Looking at 2025 performance through late November, neither leg of the prediction holds at the major‑asset‑class level.

Legacy auto OEM equities
U.S. legacy automakers have been strong performers in 2025:

• General Motors is up about 30.7% year‑to‑date as of Oct. 31, 2025, according to StatMuse’s 2025 GM price series. (statmuse.com)
• Ford is up roughly mid‑20s% year‑to‑date by early October 2025, with StatMuse showing a 25–28% gain and Zacks noting +23.4% over the prior six months, outpacing many peers. (statmuse.com)
  These returns are better than, or at least broadly in line with, the S&P 500’s mid‑teens gain over a similar period, not “terrible” nor “among the worst.” (ftportfolios.com)

Some non‑U.S. legacy OEMs have indeed struggled—Stellantis’ stock is down on the order of 30–40% in 2025, and the European autos sector is described as one of the worst‑performing sectors in Europe this year. (investopedia.com) But because several large legacy OEMs (GM, Ford, Toyota) are solidly positive and even leading their industry groups, “legacy auto OEM equities” as a whole cannot be classified as one of the worst‑performing major asset classes in 2025.

Real‑estate investments / REITs, especially in overbuilt U.S. markets
Public real estate has been a relative laggard, but not a catastrophic one at the broad‑index level:

• State Street’s Q3 real‑assets review reports U.S. REITs up about 4.5% year‑to‑date by end‑Q3 2025. (ssga.com)
• A detailed REIT study shows the average REIT (equal‑weighted) at –4.69% YTD through Oct. 31, while the cap‑weighted Vanguard Real Estate ETF (VNQ) is +2.98% YTD, and the S&P 500 is solidly positive. (seekingalpha.com)
  So REITs underperform equities but still post low‑single‑digit positive index returns, and they are not clearly the worst major asset class (for example, inflation‑linked Treasuries and some bond segments have lower returns in some real‑asset scorecards). (ssga.com)

Within real estate, some overbuilt U.S. segments did face real pain—Sunbelt multifamily markets like Austin and Phoenix experienced oversupply, rent cuts, and heavy concessions, and a basket of large multifamily REITs is down roughly mid‑teens percent in 2025. (essexcapitalmarkets.com) That makes the directional call on specific overbuilt markets partly right, but the prediction was framed at the level of “major asset classes,” where broad REIT indices have only mildly lagged rather than collapsed.

Overall assessment
Because:

1. Major U.S. legacy auto stocks (GM, Ford) are up ~25–30% YTD and thus clearly not among the worst performers, and
2. Broad real‑estate/REIT indices have small positive or modestly negative returns and are underperformers but not clearly at the bottom among major asset classes,

the claim that both legacy auto OEM equities and real‑estate investments would be “the two worst‑performing major asset classes” in 2025 is not supported by the data available through late November 2025. The prediction, as stated, is therefore wrong.

Publicly available regulatory actions in 2025 show that U.S. bank regulators did implement exactly the kind of technical, leverage‑related relief Chamath was saying would “stop in its tracks,” particularly around rules that constrain banks’ balance‑sheet capacity for low‑risk assets like Treasuries.

Key facts:

• On June 27, 2025, the Federal Reserve, FDIC, and OCC jointly proposed modifying the enhanced supplementary leverage ratio (eSLR) for the largest banks, explicitly to ensure it functions only as a backstop and does not discourage low‑risk activities such as U.S. Treasury market intermediation. The agencies acknowledged the proposal would reduce Tier 1 capital requirements for affected bank holding companies (by <2%) and more for some subsidiaries. (fdic.gov)
• On November 25, 2025, those agencies finalized this rule, lowering the effective eSLR buffer from a flat 2% add‑on to a smaller, GSIB‑surcharge‑based buffer capped at 1% for key subsidiaries, and again stressing that the change is meant to give large banks more capacity for low‑risk activities like U.S. Treasury market intermediation and repo financing. The rule is estimated to reduce Tier 1 capital at GSIB holding companies by about $13 billion and substantially more at their bank subsidiaries. (fdic.gov)
• Financial press coverage characterized this package as one of the largest post‑crisis capital easings, centered on lowering the leverage requirements that had been criticized for penalizing Treasuries and other low‑risk assets and limiting banks’ Treasury‑market participation. (ft.com)
• In parallel, Fed officials publicly debated more direct relief for Treasuries in the leverage rule, including calls to exempt Treasuries from the eSLR entirely, citing the need to encourage banks to hold more of them, especially in stress. (reuters.com)

These actions are precisely the kind of “small, arcane regulatory changes” to leverage ratios that ease constraints on holding Treasuries and other low‑risk assets that Chamath was pointing to as examples of can‑kicking. Instead of stopping, they were actively pursued and then finalized in 2025.

Because such leverage‑rule tweaks did occur in 2025, the prediction that they would stop (and thus reveal something about who is really driving policy) did not come true.

Evidence from 2025 shows both regulatory reforms and major new nuclear project announcements in the U.S. that match the prediction:

• Regulatory reforms easing nuclear build‑out: In May 2025, the White House issued an executive order "Ordering the Reform of the Nuclear Regulatory Commission" directing the NRC to facilitate increased deployment of new reactor technologies, lower regulatory and cost barriers, reorganize to process licenses more quickly, and explicitly add facilitation of nuclear power to its mission, not just safety. (whitehouse.gov) In June 2025, the NRC implemented a new fee rule cutting hourly charges for advanced reactor applicants and pre‑applicants by more than 50%, as required by the ADVANCE Act, directly lowering the cost of licensing new reactors. (ans.org) Together, these are material regulatory changes intended to speed and cheapen advanced nuclear licensing.

• Significant new U.S. nuclear power plant projects announced in 2025: In July 2025, Westinghouse publicly laid out plans to build ten new AP1000 reactors in the USA by 2030, an explicit large‑scale build‑out, framed in the context of Trump’s May 23 executive orders to support the U.S. nuclear supply chain and quadruple nuclear capacity by 2050. (world-nuclear-news.org) In October 2025, Amazon and Energy Northwest announced the Cascade Advanced Energy Facility in Washington state, a new SMR power station using X‑energy’s Xe‑100 reactors (initially four units, expandable to 12) to supply Amazon’s data‑center demand—one of the first large, tech‑backed modular nuclear projects in the U.S. (tomshardware.com) In parallel, DOE approved a $1 billion federal loan in 2025 to restart Three Mile Island Unit 1 (the Crane Clean Energy Center) for long‑term power to Microsoft data centers, signaling federal support for bringing dormant nuclear capacity back online as part of a broader build‑out. (apnews.com)

Because 2025 saw both (1) meaningful regulatory reforms at the NRC that lower costs and aim to speed approvals, and (2) high‑profile announcements of new reactors and restarts in the U.S., the prediction that 2025 would bring announcements of nuclear build‑out and/or regulatory easing for nuclear power is substantially borne out.

Available 2025 data show that venture capital DPI did not experience the broad-based surge Jason predicted, even though M&A and IPO activity did rebound.

1. DPI / Distributions did not “shower down.”

   • An April 2025 PitchBook-based analysis described the industry as facing a “DPI crisis,” emphasizing that exits remained elusive and that secondaries were far too small to fix the shortfall in cash distributions to LPs.【6view0】
   • A September 2025 deep-dive on the 2025 liquidity crunch reports that, despite record committed capital, distributions (DPI) are “lagging well below expectations” and traditional IPO/M&A exits are still “nearly frozen” for most portfolios.【8view0】
   • Equitybee’s DPI commentary notes that for eight consecutive quarters, distribution rates have averaged single‑digit percentages of NAV, far below the decade average of 16.8%, highlighting a continuing distribution drought rather than a surge.【7search7】
   • A Q3 2025 venture market summary citing Carta data states that median DPI for 2019–2024 VC vintages sits at 0.00x, underscoring how little capital has actually been returned across recent funds.【7search9】
     Collectively, these point to modest, uneven improvement at best—not the “exits in DPI shower down” environment Jason envisioned.

2. M&A and IPO activity did rebound meaningfully.

   • Global M&A in 2025 is up sharply: Reuters and others report roughly a 10%+ rise in global deal value, with tech/media/telecom the largest sector, and other analyses put the 2025 rebound closer to a ~40% increase vs. 2024 as megadeals return.【1news12】【1news14】
   • VC-focused reports show exit value at multi‑year highs: KPMG’s Q3 2025 Venture Pulse notes that global VC exit value hit about $150B in Q3 2025, a 15‑quarter high, driven by a reopening IPO window and stronger M&A.【5search2】
   • In the U.S., Renaissance Capital data show 191 IPOs in 2025 vs. 150 in all of 2024, and AP reporting pegs 2025 IPO proceeds around $31B by October, the best year since 2021, with notable tech listings like Circle, CoreWeave, Klarna and others.【4news14】【4search3】
     So the claim that M&A and IPOs would “surge” is broadly consistent with 2025 deal and IPO statistics.

3. “Wrath of Khan ends” and antitrust pressure: partially but not clearly causal.

   • Lina Khan’s term as FTC chair ended on January 20, 2025, and she was replaced by Andrew N. Ferguson, widely described as a more deal‑friendly Republican chair.【3search17】【3search18】
   • Tech and VC commentary (e.g., TechCrunch) noted that Khan’s departure was welcomed in Silicon Valley, where her tenure was blamed for chilling M&A; Ferguson’s appointment was explicitly framed as the start of a more merger‑friendly era.【3search1】
   • However, a mid‑2025 US VC M&A outlook observes that the “anticipated regulatory shift that boosts large M&As has not yet materialized,” and that tech and healthcare deals still face substantial regulatory scrutiny.【5search3】
   • Major antitrust cases (e.g., against Meta and Microsoft’s AI activities) continued or evolved under the new leadership, suggesting that while the tone may be softer, antitrust pressure did not simply vanish.【2news14】【2search19】
     Thus, there is some alignment with Jason’s narrative (Khan out, a more permissive regime in), but the evidence for a clean cause‑and‑effect link between her exit and the 2025 deal surge is mixed.

Bottom line:
Even though tech‑heavy M&A and IPO activity did rebound in 2025 and Khan’s departure coincided with a more business‑friendly enforcement posture, multiple independent data points still describe a DPI “crisis,” liquidity drought, and near‑zero median DPI for recent VC vintages well into Q3 2025.【6view0】【8view0】【7search7】【7search9】 That contradicts the core of Jason’s prediction that VC DPI would surge and “exits in DPI shower down.” As such, the overall prediction is best classified as wrong.

Trump did in fact win a new term beginning in 2025 and was inaugurated for his second presidency on January 20, 2025, starting a new administration. 【(en.wikipedia.org)

Within days, he signed Executive Order 14176 on January 23, 2025, ordering the declassification of records related to the assassinations of John F. Kennedy, Robert F. Kennedy, and Martin Luther King Jr. 【(en.wikipedia.org) Pursuant to this, the National Archives released more than 60,000 previously withheld assassination-related documents in March 2025, totaling about 77,100 pages across over 2,300 PDFs—a very large volume of formerly secret material. 【(en.wikipedia.org) This directly matches the prediction of an "enormity of the files" being declassified and publicly released, specifically including JFK-related records.

On Epstein, Congress passed and Trump signed the Epstein Files Transparency Act on November 19, 2025, mandating the Department of Justice to release all unclassified DOJ records related to Jeffrey Epstein within 30 days, with the public release expected around December 19, 2025. 【(en.wikipedia.org) As of November 30, 2025, those files have not yet been fully released, but the law and ongoing DOJ actions show that large-scale Epstein disclosures are imminent and being driven by the Trump administration.

Because (1) a new Trump term began in 2025, and (2) that administration has already declassified and released very large troves of previously secret files on major historical controversies (notably JFK/MLK/RFK), with additional large-scale Epstein disclosures legally queued up, the core substance of Chamath’s prediction has already materialized. The remaining Epstein release affects completeness but not the basic correctness of the forecast that Trump’s new term would be marked by major declassification waves on controversies like JFK and Epstein.