E151: WW3 risk, War with Iran?, 4.9% GDP, startup failures growing, new Speaker & more

Israel did launch a full-scale ground invasion of Gaza on 27 October 2023, and that invasion is still formally ongoing more than two years later, with the campaign described in reference works as an "Israeli invasion of the Gaza Strip" running from 27 October 2023 to the present. (en.wikipedia.org) Large-scale ground and urban operations continued well beyond the initial months, including major offensives and sieges in Gaza City through January 2025 and in Khan Yunis, Beit Hanoun, and Deir al-Balah into mid–late 2025, contradicting the idea that the operation would be “not a long one.” (en.wikipedia.org)

A durable, structured ceasefire was only reached in stages starting with a temporary January–March 2025 armistice, which then collapsed when Israel resumed large-scale bombardment and ground attacks in March 2025; a more comprehensive ceasefire and withdrawal framework was not agreed until October 2025. (en.wikipedia.org) By then, the war had already caused tens of thousands of Palestinian deaths and massive destruction in Gaza, with major hostilities lasting roughly two years, far from the short campaign envisioned. (theguardian.com)

Moreover, the conflict produced substantial regional escalation before any lasting ceasefire: Iran carried out a large drone and missile strike directly against Israel in April 2024, and Yemen’s Houthi movement waged an extended campaign against international shipping in the Red Sea explicitly tied to the Gaza war, prompting US‑, UK‑, and EU‑led naval responses. (en.wikipedia.org) While this fell short of a classic multi‑front conventional “regional war,” it clearly contradicted the prediction that a brief operation would end in a ceasefire before the situation escalated in the region.

Because the Gaza ground campaign turned into a prolonged, years‑long war with only late and fragile ceasefires, and significant regional spillover occurred before those ceasefires, Sacks’s scenario of a short operation capped by an early, stabilizing ceasefire did not materialize.

This prediction is about a change in probability (“2% chance to 8% chance … in the next 20 years”) rather than a concrete event that must occur or not occur by a fixed date. That makes it inherently a statement of subjective risk assessment, not something that can be cleanly verified or falsified.

Empirically, there is evidence that many experts and institutions perceived elevated nuclear risk around and after 2023. For example, in March 2023 the UN High Representative for Disarmament Affairs told the Security Council that the risk of nuclear weapons use was higher than at any time since the end of the Cold War, largely due to the war in Ukraine and erosion of arms‑control architecture. (press.un.org) In June 2024, Pugwash (a leading disarmament NGO) similarly stated there is wide consensus that the risk of nuclear war is higher than at any time in the recent past, citing the ongoing wars in Ukraine and Gaza as drivers of "alarmingly high" risks of nuclear weapons use. (pugwash.org) Judgmental-forecasting work on nuclear risk (e.g., surveys of experts and superforecasters about the chance of large‑scale nuclear war by 2045) finds low but non‑trivial probabilities (around 1–5%), but these studies do not provide a time series detailed enough to say that the probability within 2023 rose from “a few percent” to “high single digits.” (hks.harvard.edu)

Two further issues prevent a clear verdict:

1. No observable ground truth for the probability change. We can observe wars, crises, and munitions shortages, but not the “true” probability of nuclear use or how it moved between early and late 2023; different experts could reasonably disagree on whether the change was “material” or on the order Friedberg suggested.
2. Long time horizon and probabilistic nature. The forecast concerns the chance of some nuclear use in the 20 years after late 2023 (roughly to 2043). We are only in 2025, and even if we waited to 2043, seeing either use or non‑use would not uniquely validate or falsify a specific 2% vs. 8% ex‑ante probability estimate.

Because the claim is a subjective, quantitative risk assessment about a still‑ongoing 20‑year window, and available data only show that many observers felt risk was higher (without pinning down the magnitude of change Friedberg posits), this prediction cannot be labeled definitively right or wrong. Hence: ambiguous.

Chamath’s claim was that in the months after October 27, 2023, both the Israel‑Hamas and Russia‑Ukraine wars would stay “slow and simmering,” with relatively de‑escalatory actions (even if rhetoric intensified), and without rapid military escalation to a much larger regional or world war.

What actually happened:

1. Israel–Hamas / Middle East theater:

   • Israel’s Gaza campaign expanded from airstrikes into a full‑scale ground invasion beginning in late October 2023, with very high casualties and destruction.
   • The conflict rapidly spilled over regionally: sustained cross‑border attacks between Israel and Hezbollah in Lebanon, with thousands of rocket and missile exchanges recorded and growing concern about a much wider Israel–Hezbollah war.
   • Iran‑aligned Houthis launched dozens of drone and missile attacks on commercial and naval shipping in the Red Sea and Gulf of Aden, explicitly tying them to the Gaza war, prompting repeated U.S.–UK and allied air and naval strikes in Yemen. These strikes and Houthi attacks persisted and intensified into 2024. (cnbc.com)
   • On 13 April 2024, Iran launched its first ever direct large‑scale missile and drone barrage on Israel from Iranian territory—hundreds of projectiles in a single night—widely described as a major escalation carrying serious risk of a region‑wide war. (aljazeera.com)
   • Subsequent analysis by CSIS and others characterized the Israel–Hezbollah–Iran front as a dramatically widened and more violent regional conflict, with a sharp increase in the number and geographic spread of strikes compared with late 2023. (csis.org)

   These developments contradict the idea that “actual actions are relatively de‑escalatory” and that there would be no rapid escalation toward a much larger regional war. The Middle East conflict clearly expanded beyond a “slow and simmering” bilateral fight.

2. Russia–Ukraine war:

   • From late 2023 through 2024–2025, the Russo‑Ukrainian war largely remained an attritional conventional conflict confined to Ukraine’s territory, with major but geographically limited offensives such as the Pokrovsk and Sumy campaigns and no direct NATO combat deployment. (en.wikipedia.org)
   • NATO states repeatedly stressed they would not send combat troops to avoid a direct NATO–Russia war, and that line has held, so this part of his intuition (no rapid escalation to world war via Ukraine) has been broadly accurate. (aljazeera.com)

However, his prediction was jointly about both conflicts and specifically about actions staying relatively de‑escalatory and not jumping into a much larger regional war. Given the substantial, fast‑moving regional escalation of the Israel‑Hamas conflict into a broader Israel–Iran–Hezbollah–Red Sea theater, that core claim did not hold. Even though the Ukraine portion was roughly right, the overall prediction—as stated—is best judged wrong.

The prediction has two linked parts—both miss what actually happened:

1. Recession timing (Q4 2023–Q2 2024)

   • The NBER, the official arbiter of U.S. recessions, still lists the most recent business‑cycle peak in February 2020 and trough in April 2020, with no later recession designated as of 2025. (nber.org)
   • BEA data show positive real GDP growth in all of the relevant quarters: Q4 2023 grew at an annual rate of 3.3%, Q1 2024 at 1.6%, and Q2 2024 at 2.8%. (bea.gov)
   • That combination (continued expansion, no NBER recession call) means the U.S. did not enter a recession in the 2–3 quarters after Q3 2023.

2. Fed rate‑cut timing (“starting in the middle of next year”)

   • The Federal Reserve held policy rates steady through mid‑2024 and did not begin easing in that window.
   • The first cut of the post‑hiking cycle came on September 18, 2024, when the FOMC lowered the federal funds target range by 50 bps to 4.75–5.00%, widely reported as the first rate cut since March 2020. (federalreserve.gov)
   • A further 25‑bp cut in December 2024 took the range down to 4.25–4.50%, confirming that the easing cycle started in late 2024, not mid‑2024. (federalreserve.gov)

Because (a) no recession occurred in the specified Q4 2023–Q2 2024 window, and (b) the Fed did not start cutting rates around mid‑2024 but instead in September 2024, Chamath’s combined prediction is wrong.

Stripe’s next large liquidity event after the Oct 27, 2023 podcast was a tender offer for employee and former‑employee shares announced on February 28, 2024. Stripe and new investors agreed to buy more than $1 billion of employee stock at a $65 billion valuation, up from a $50 billion Series I funding valuation in March 2023.(stripe.com)

Chamath’s prediction was that the next market‑clearing valuation for Stripe equity would be in the $25–30 billion range—roughly a 50% markdown from the prior ~$55 billion round. Instead, the first post‑podcast liquidity event cleared at $65 billion, which is above the earlier $50–55 billion level, not 50% below it.(ft.com)

Subsequent large secondary transactions have only moved Stripe’s valuation higher (e.g., a February 27, 2025 tender offer valuing the company at $91.5 billion, and reports of a later round taking it above $100 billion), confirming that the market never repriced Stripe down to the $25–30 billion range.(stripe.com)

Because the very next major liquidity event after October 2023 priced Stripe far above the predicted range, and later events pushed the valuation even higher, the prediction that Stripe’s next clearing valuation would be around $25–30 billion is wrong.

Multiple independent datasets show that startup failures were significantly higher from late 2023 through at least late 2024 than in the 2021 boom period, matching Sacks’s prediction both in timing (roughly the 18 months after October 2023) and in mechanism (2021‑era companies running out of runway and unable to refinance).

Key points:

• 2021 as easy‑money baseline: 2021 was a record year for global VC funding, with around $621–681 billion invested worldwide and more than $300 billion in the U.S., reflecting extremely loose capital and supportive conditions for startups rather than mass shutdowns.(news.crunchbase.com)

• Sharp rise in shutdowns starting 2023: Carta data (U.S. startups) show 467 shutdowns in 2022, rising to 770 in 2023, which Forbes describes as the highest number of shutdowns Carta had ever tracked and calls 2023 “the most fatal year for startups.”(forbes.com) Q3 2023 alone saw 212 shutdowns, the highest quarterly total since Carta began tracking this in 2012, up 50% year‑over‑year.(southernstartups.org) This is already a clear elevation versus the 2021 boom environment.

• Further acceleration into Sacks’s 18‑month window: Carta reports that shutdowns continued to accelerate into 2024, with Q1 2024 reaching 254 closures—a new quarterly record and 58% more than Q1 2023, after a 124% jump between Q1 2022 and Q1 2023.(carta.com) Full‑year data show 966 shutdowns in 2024 vs 769 in 2023, a 25.6% increase and again a new high.(techcrunch.com) Forbes and Carta both note that 2024 saw the highest absolute number of startup shutdowns ever recorded in Carta’s data.(forbes.com) TechCrunch and others expect 2025 to remain “another brutal year of failed startups,” indicating the elevated failure environment persisted into the first part of 2025.(techcrunch.com)

• Direct link to 2020–2021 funding cohorts and inability to raise: Multiple analyses explicitly tie these failures to startups that raised large rounds in 2020–2021 and then hit the 18–24 month mark in a far tighter funding environment. Carta’s head of insights notes that companies that raised in late 2020 and early 2021 were facing their next raise in 2022–23 just as capital dried up, forcing shutdowns.(bizjournals.com) The Financial Times similarly reports a 60% rise in startup failures as many firms that raised during the 2021–22 boom exhausted their cash and couldn’t secure new funding on acceptable terms.(ft.com) A 2025 Carta‑based review emphasizes that the wave of closures in 2024 was “concentrated” among startups funded in 2021–2022, exactly the dynamic Sacks described.(forbes.com)

• Global corroboration: Outside the U.S., Tracxn data on India (the world’s third‑largest startup ecosystem) show 15,921 shutdowns in 2023 and 12,717 in 2024, versus only about 2,300 shutdowns over 2019–2022 combined, a roughly 12‑fold jump, again indicating a post‑2021 correction with elevated failures.(magzter.com)

Across geographies and data providers, the pattern is consistent: 2023–2024 (and into early 2025) show a markedly higher level of startup shutdowns than the 2021 ZIRP boom, and analyses explicitly attribute this to companies that raised big 2020–2021 rounds running out of cash in a much harsher funding environment. That matches Sacks’s prediction on both direction and timing, so it is best classified as right.

Available data from late 2023 through 2025 show that the venture and growth equity markets did not revert to 2020–2021 peak pricing and instead went through an extended, broad valuation reset—especially at the later stages—matching Chamath’s thesis that “the hard work” of repricing still lay ahead.

Key evidence:

• Carta’s Q4 2023 "State of Private Markets" shows that by late 2023 median valuations at Series D were down nearly 42% from early‑2021 levels, and Series E+ valuations, despite some rebound, were still less than half of their 2021–2022 peaks, indicating a deep markdown at the growth stages rather than a return to bubble valuations. (carta.com)
• Analysis of the 2022–2024 funding environment finds that down rounds have risen structurally: Equidam reports down rounds reached 27.4% of all VC deals in Q1 2024—the highest in a decade—and still about 20% for 2024 overall, evidence of sustained valuation resets rather than transitory blips. (equidam.com)
• Commentaries in 2025 explicitly refer to a "Great Valuation Reset" and estimate that roughly 1 in 5 venture rounds since 2023 has been a down round, framing this as an ongoing correction in how startups are priced after the 2021 boom. (ericashman.com)
• 2025 enterprise SaaS funding data show late‑stage VC median pre‑money valuations at about $74M in 2025 vs. $278M in 2021—a ~73% decline—while venture growth valuations are flat vs. 2021, and large valuation step‑ups have become rare. This is a textbook “hard reset” for late‑stage and growth equity deals. (developmentcorporate.com)
• Broader VC market overviews in 2025 describe “valuation reset” as a core trend: later‑stage valuations remain below 2020 levels; flat and down rounds peaked in 2024, even for marquee companies (e.g., Plaid), underscoring that many private tech companies have been repriced downward rather than returning to their pandemic‑era highs. (caia.org)

There are nuances: seed and some AI‑focused companies have seen valuations hold steady or even exceed 2021 levels, so the reset is not uniform across every single stage or sector. (caia.org) But Chamath’s core claim—that the post‑October‑2023 period would require a difficult valuation reset instead of a smooth reversion to 2020–2021 prices, and that many private tech companies would be marked down—has been borne out by the prevalence of down rounds and steep valuation cuts at growth and late stages through 2024–2025. Thus the prediction is best categorized as right.

Jason’s prediction was that over the next several years a significant number of VC firms would either shut down or effectively become “zombie funds” that only manage existing portfolios and stop raising new funds or leading new investments.

By late 2025, multiple industry data points show this has indeed been happening at scale:

• PitchBook data (reported in October 2024) notes that about 45% of roughly 25,000 venture investors had made no investments in 2024, and explicitly characterizes this large inactive cohort as the “ballooning ranks of the zombie funds”—firms that can’t raise from LPs or write new checks but continue to exist managing old portfolios and collecting fees. (wolfstreet.com)
• Analysis of 2025 U.S. VC fundraising finds that hundreds of VC firms are now classified as “zombies,” with an estimated 574 zombie firms in 2025—a ~50% year‑over‑year increase—as fundraising fell ~34% in the first half of 2025. These zombies are described as unable to raise new funds or invest, fitting Jason’s description. (linkedin.com)
• A 2025 Forbes interview on private equity and VC’s “great cull” reports “hundreds of zombie funds that haven’t returned capital and can’t raise again”, and notes that fundraising for traditional early‑stage VC has become “nearly impossible” without realized returns—again matching the idea that many firms are stuck merely managing old portfolios. (forbes.com)
• The CEO of EQT, one of the largest private‑capital firms, stated in 2025 that out of more than 15,000 private‑capital firms globally, only about 5,000 have raised a fund in the last seven years and that up to 80% could become “zombie firms” over the next decade, implying that around 10,000 already can’t raise fresh capital. (ft.com)
• Broader fundraising data show a sustained slump: global private‑equity/VC fundraising in 2025 is well below 2021 peaks, with sharply fewer fund closings, indicating many managers are failing to raise successor vehicles and are effectively stuck running off existing funds. (wsj.com)

Jason did not give a precise cutoff date or a specific percentage; he only claimed that a lot of VC firms would lose their seats in the “musical chairs” and either shut down or end up as zombie funds over the ensuing years. By late 2025, the documented existence of hundreds to thousands of such zombie or non‑viable firms, alongside a clear consolidation and fundraising drought, matches that qualitative prediction. The trend he described has already materialized at meaningful scale and is widely recognized in industry analyses, so the prediction is best judged as right.

Jason’s prediction was that, after the 2021–2022 bubble, many venture firms launched in that cycle would either shut down or become effectively non‑operating “zombie funds” over the following several years.

By 2025 there is strong evidence that this dynamic is already well underway:

• PitchBook data cited by Fortune show the number of U.S. “zombie” venture firms (defined as firms that raised a fund in the prior six years but made no new investments over a multiyear window) rose from 382 at the end of 2021 to 574 by early 2025—roughly a 50% jump.(fortune.com)
• Analysis drawing on PitchBook data notes that around 45% of roughly 14,000 venture investors made no investments at all in 2024, described as the “ballooning ranks of the zombie funds” that can’t raise new LP capital and mainly just manage old portfolios for fees.(wolfstreet.com)
• In Europe, the number of active VCs (those doing at least one deal per year) fell about 30% between 2022 and 2024; newer funds in particular have “partially stopped doing investments,” which industry participants explicitly link to a rise in zombie funds as LPs grow impatient for returns.(sifted.eu)
• In Canada, industry reporting describes a growing set of VC funds that can’t raise new vehicles but continue to manage legacy portfolios—explicitly labeled “zombie funds” that linger until conditions improve or they quietly die.(thelogic.co)
• Broader private‑capital analysis from senior industry figures (e.g., EQT’s CEO) anticipates thousands more firms globally becoming zombies or disappearing amid a difficult fundraising environment over the rest of the decade, reinforcing that the shake‑out is real and ongoing.(ft.com)

While the full outcome through the “late 2020s” is not yet observable, the core claim—that a large number of venture firms from the prior boom would shut down or effectively become zombie funds in the ensuing years—has clearly been borne out by mid‑decade data and industry consensus. Therefore this prediction is best judged as right in direction and substance, even if the eventual scale may grow further as the cycle plays out.

There isn’t enough elapsed time or granular data to verify Friedberg’s prediction about final fund‑life multiples for this cohort of “mid‑tier” venture funds.

Key points:

1. Prediction concerns eventual, fully realized fund outcomes.
   Friedberg was talking about 10‑year‑plus fund lives: historically mid‑tier funds returning around ~1.8x gross over a decade, versus this cycle supposedly ending at only 0.5x–0.7x (i.e., net capital‑losing when all is said and done).(podscripts.co)

2. The vintages he’s implicitly talking about (roughly 2020–2022) are only ~3–5 years old.
   Carta and other benchmark data show that even the 2017 vintage (around 8 years old by 1Q25) still has a median TVPI of about 1.72x but a DPI of only 0.27x—most gains are unrealized even at that age.(ontheflyingbridge.wordpress.com) Silicon Valley Bank analysis cited in the same piece notes that top‑quartile funds typically don’t finish returning capital until roughly year 10–15.(ontheflyingbridge.wordpress.com) That makes it inherently too early to know the final MOIC for 2020–2022 funds, which are much younger.

3. Current data do show underperformance, but not final loss levels.
   Carta’s VC fund performance work (summarized in multiple analyses) finds that recent vintages (2021–2023) have median TVPIs hovering around 0.92–0.99x and median IRRs below zero—clearly weaker than earlier vintages but still close to 1x TVPI, not already at 0.5x–0.7x.(nehaldesai.com)(carta.com) A Wall Street Journal summary of Carta’s report notes that more than 90% of 2021 U.S. VC funds have made no distributions three years in, with median net IRR for 2021 funds around –1.5% and for 2022 funds around –5.8%, again indicating stress but not a realized permanent loss yet.(wsj.com)

4. Mid‑tier segment specifically is not observable at the needed granularity.
   Public benchmark datasets (Carta, AngelList, various LP/consultant reports) provide distributions and TVPIs by vintage, fund size, or percentile, but they do not cleanly isolate Friedberg’s "mid‑tier that historically would have been ~1.8x gross" as a separate category. We can see that many smaller and newer managers are struggling and that fundraising has consolidated toward top‑tier firms,(ft.com) but we cannot empirically confirm that this entire mid‑tier cohort is on track to finish at 0.5x–0.7x.

5. Because venture outcomes are heavily back‑loaded, it’s premature to declare net capital‑losing funds across the board.
   The same Carta data emphasize the J‑curve nature of fund performance: early‑life TVPI and IRR are often low or negative and only improve later as exits occur.(carta.com) Given that the funds in question have most of their life (and potential exit cycles) ahead of them, any claim about their ultimate MOIC is still speculative.

Putting this together: available evidence supports the idea that many recent‑vintage, non‑top‑tier VC funds are currently under pressure and may end up with below‑historical returns. But we are far too early in the fund life cycle—and lack a clean, observable dataset for Friedberg’s precise “mid‑tier” cohort—to say whether they will indeed realize only 0.5x–0.7x over their fund life. Therefore the prediction cannot yet be judged as right or wrong.

Evidence shows that both parts of Chamath’s prediction have not played out as stated.

1. Stripe has still not gone public as of late 2025. Multiple up‑to‑date sources note that Stripe remains a privately held company, with recent liquidity provided via tender offers (e.g., at ~$91.5B valuation in February 2025 and around ~$100B+ in later 2025), and explicitly state that it has no IPO completed or on the calendar yet.(fintechweekly.com) This means the hypothesized trigger event (a Stripe IPO) has not occurred.

2. Nonetheless, a broad reset of venture and late‑stage private tech valuations has already occurred without a Stripe IPO. Analyses of the private markets describe a “valuation reset” across venture capital in 2022–2024, with valuations falling back toward long‑term historical averages and a sharp rise in down rounds, particularly at later stages.(wellington.com) For example, Carta’s data conclude that venture valuations “have undergone a reset in the last two years,” with down rounds roughly doubling versus the pre‑2022 period, while other reports characterize 2023 as a year of market‑wide valuation reset and 2024–2025 as a period of gradual recovery from that reset.(edwardconard.com)

Because (a) Stripe has not IPO’d, and (b) the broad reset in private‑company valuations that Chamath was talking about is widely assessed to have already happened through 2022–2024—driven by macro conditions and public‑market repricing, not by any Stripe IPO pricing—his claim that “the reset can’t happen until Stripe goes public” and that Stripe’s IPO would set the cascading valuation framework is contradicted by observed market behavior. Therefore, the prediction is best judged as wrong.

The prediction is explicitly about multi‑decade trends (“over time,” “several trillion dollars”) in:

1. Climate/physical trend: rising ocean heat content and sea‑surface temperatures driving more frequent, rapidly intensifying storms like Hurricane Otis.
2. Economic impact: cumulative global losses on the order of trillions of dollars via insurance, reinsurance, and coastal real estate devaluation.

As of late 2025, only about two years have passed since the prediction. The relevant questions are:

• Do we already know the multi‑decade change in frequency of extreme, fast‑intensifying cyclones? → No; we only have short‑term indications. IPCC and recent studies do indicate warming oceans and an observed increase in the proportion of intense tropical cyclones and rapid intensification events, but full “coming decades” trajectories can’t yet be confirmed.
• Have we already observed “several trillion dollars” in additional, climate‑driven cumulative losses attributable specifically to this mechanism? → Global climate‑related catastrophe and coastal real‑estate risk are mounting, and insured catastrophe losses have risen sharply in recent decades, but attributing multi‑trillion losses specifically to the projected future increase in Otis‑like events requires decades of realized data, not a couple of years.

Because the claim’s time horizon is decades, and key elements (long‑run frequency shift and total losses) cannot yet be empirically verified or falsified, the correct status as of November 30, 2025 is:

• We have early evidence consistent with the mechanism (warmer oceans, high ocean heat content, concerns about insurability and coastal property risk),
• but insufficient elapsed time to say whether the predicted multi‑decade outcome (sustained higher frequency of storms like Otis plus multi‑trillion cumulative losses from that specific dynamic) is right or wrong.

Therefore the prediction is too early to call and must be labeled inconclusive rather than right, wrong, or permanently ambiguous.

As of November 30, 2025, nothing close to a $1 trillion federal bailout or backstop of coastal real‑estate–linked property insurance has occurred. Existing federal programs dealing with housing and catastrophe risk (e.g., TARP at $700 billion in 2008 for financial institutions, the Treasury’s commitments to Fannie Mae/Freddie Mac around $200 billion, and the long‑running National Flood Insurance Program with debts in the tens of billions) are far smaller in scale and predate Chamath’s 2023 prediction rather than being new responses to a Florida‑triggered insurance collapse. (en.wikipedia.org)

In Florida specifically, the state’s Reinsurance to Assist Policyholders (RAP) backstop has not blown up; instead, it has been under‑used. Florida lawmakers actually reduced the RAP program’s authorized funding from $2 billion to $900 million and pulled back about $1.1–2.1 billion in unused reinsurance support, while also repealing the separate FORA program—moves that lessen, not socialize, the state’s exposure, and there has been no federal assumption of those obligations. (insurancebusinessmag.com)

Because Chamath framed this as happening "at some point" in the future, with no explicit time horizon, and the hypothesized trigger event (state reinsurance insolvency leading to a ~$1T federal rescue) has not yet occurred—but could in principle still happen over coming decades—the prediction cannot be called right or definitively wrong at this time. It remains inconclusive (too early).

Friedberg predicted that events like Hurricane Otis would be early catalysts for a multi‑year, cascading downward repricing of high‑risk coastal real estate, primarily via unaffordable or unavailable insurance forcing owners to sell at lower values.

Since late 2023, there is clear evidence that this mechanism has begun operating in multiple coastal regions:

• In southwest Florida (Lee County / Fort Myers, heavily hit by Hurricane Ian), insurance premiums spiked, insurers pulled back, and home values fell more than 10% year‑over‑year and ~16% below pre‑storm levels; local real‑estate and finance experts explicitly attribute this to high insurance costs shrinking the buyer pool and pressuring sellers to cut prices.(opb.org)
• In coastal Lafitte, Louisiana, where flood and storm risk is high, homeowners saw premiums more than double to over $8,000/year, and local home values have dropped about 38% since 2020, with many for‑sale signs and owners reporting they are “stuck” because buyers cannot obtain or afford insurance.(insurancenewsnet.com)
• A National Bureau of Economic Research–based study of 74 million policies finds rising insurance premiums are already cutting home price growth by about $40,000 and causing an 11% relative price decline in areas most exposed to disasters, i.e., high‑risk coastal and wildfire zones.(realestatenews.com)
• Florida data show coastal home values softening by roughly 5–10% in 2025 while inland/low‑risk homes are still appreciating, with analysts directly linking the divergence to soaring insurance costs and insurer retreat after recent hurricanes.(naplesed.com)
• Federal and industry reports document insurers withdrawing from or sharply limiting coverage in coastal and wildfire‑prone states like Florida, California, Louisiana, and North Carolina, leaving some communities effectively “uninsurable” and reinforcing downward pressure on prices.(heinrich.senate.gov)

Analysts and policymakers now describe this as a nascent but growing insurance‑driven housing crisis in high‑risk areas—exactly the kind of early, cascading repricing dynamic Friedberg anticipated. The full multi‑year adjustment is still unfolding, but by late 2025 the trend and mechanism he described are clearly visible across multiple coastal regions, so his prediction is broadly borne out.

The prediction’s time horizon is about 10 years from October 2023 (i.e., until roughly 2033). As of today (November 30, 2025), fewer than 10 years have elapsed, so it’s too early to determine whether the Middle East will become the world’s second‑largest VC region by that date.

Available data today show that:

• The U.S. remains the largest venture capital market by a wide margin.
• Other major regions—especially China and Europe—still account for substantially larger VC investment than the Middle East/North Africa (MENA), though MENA (particularly Saudi Arabia and the UAE) is growing quickly in terms of startup funding, sovereign wealth capital, and new funds.

However, current rankings don’t settle a 2033 outcome. Rapid policy changes, sovereign‑wealth‑fund deployment, and regional development initiatives could significantly change VC geography over the remainder of the decade, and those future dynamics cannot be reliably inferred now.

Because the specified deadline has not yet arrived, the correctness of the prediction cannot yet be judged and must be marked as inconclusive (too early).

The prediction’s core claim was that Mike Johnson would not remain Speaker "long term," specifically that he would be removed or replaced before serving multiple Congresses, even though he might survive the first few skirmishes.

In reality, Mike Johnson was elected Speaker on October 25, 2023 during the 118th Congress and is still serving as Speaker in the 119th Congress. His biography notes he has been Speaker since 2023, and the page for the 119th United States Congress lists him as the current Speaker, confirming he continued in that role into a second Congress. (en.wikipedia.org) A November 25, 2025 Congressional Record entry also shows him exercising the office by appointing a Speaker pro tempore, further confirming he remains Speaker at least through late 2025. (congress.gov) Recent news coverage likewise refers to him as Speaker in late November 2025. (nypost.com)

Because he has already served as Speaker across more than one Congress (118th and 119th), the prediction that he would be removed before serving multiple Congresses is falsified, even though the lesser claim that he would survive the initial skirmishes did come true. Overall, the prediction is therefore wrong.

Evidence from late 2023 onward shows a sustained pattern of U.S. outlets resurfacing Mike Johnson’s past writings, sermons, and religious activism and framing them as extreme, matching Sacks’s prediction.

• Shortly after Johnson became Speaker in October 2023, CNN’s KFile ran a widely echoed piece detailing his early‑2000s editorials in the Shreveport Times supporting criminalization of gay sex and calling homosexuality an “inherently unnatural” and “dangerous lifestyle” that could destroy “the entire democratic system.” This was explicitly framed as digging into his past rhetoric now that he was Speaker and was repeatedly cited by other outlets and partisan comms operations.

  Sources: CNN article and re‑use in other political communications. (amp.cnn.com)

• New York Magazine/Intelligencer’s “Mike Johnson’s Old‑Time Religion” and related coverage highlighted an earlier sermon where Johnson rejected evolution and suggested teaching evolution contributes to school shootings, again using an old sermon to illustrate his worldview as Speaker. (en.wikipedia.org)

• Rolling Stone and the New York Times resurfaced Johnson’s comments on a pre‑speakership Christian prayer call in which he described America as “dark and depraved” and questioned whether God would bring judgment; these were packaged as revelations about the new Speaker’s theology and rhetoric from before he was in leadership. (debatepolitics.com)

• NPR and other outlets in November–December 2023 ran profiles scrutinizing Johnson’s ties to far‑right Christian networks (e.g., New Apostolic Reformation, dominion theology) and events like the National Gathering for Prayer and Repentance, explicitly presenting these older relationships and appearances as evidence of a Christian‑nationalist, theocratic bent now relevant because he is Speaker. (wusf.org)

• A wide range of commentators and outlets repeatedly described Johnson with terms such as “hard‑core theocrat,” “extreme Christian fundamentalist,” and “Christian nationalist”; some high‑profile commentators compared his rhetoric to Osama bin Laden’s or likened his Christian nationalism to the Taliban, clearly aiming to portray him as radical or unhinged. (catholicleague.org)

• Coverage did not stop after the initial week: through late 2023 and into 2024, outlets continued to publish pieces revisiting his past legal work for the Ark Encounter/Creation Museum, his earlier statements blaming evolution for moral decay, and his longstanding ties to organizations like Answers in Genesis and the Family Research Council, all framed as indicators of an unusually extreme religious‑political worldview for a modern Speaker. (religionnews.com)

Putting this together, media behavior matches the forecasted “drip, drip, drip” of stories about Johnson’s backbench‑era sermons, writings, and affiliations, with a consistent emphasis on presenting those past remarks as extreme. That is substantively what Sacks predicted, so the prediction is best judged as right.

Key elements of Sacks’ prediction did not play out as described.

• Lower‑level defendants did flip, but they did have “something on Trump.” Jenna Ellis, Sidney Powell, Kenneth Chesebro, and Scott Hall all took plea deals in October 2023 that required them to provide information that could be used against the remaining defendants, including Trump.(fultongrandjury.com) Their recorded proffer sessions describe Trump’s inner‑circle saying he would not leave office, Powell’s direct communications with Trump about seizing voting machines, and Chesebro’s role in the fake‑elector scheme—material that legal analysts characterized as damaging evidence against Trump, not mere leverage to reach others.(abc11.com) This directly contradicts Sacks’ claim that these figures “don’t have anything on Trump.”
• Prosecutors did not, in practice, flip Giuliani or Eastman. Reporting on the Fulton County DA’s strategy indicated Trump, Mark Meadows, and Rudy Giuliani were not being offered plea deals, with prosecutors instead aiming to take them to trial.(theguardian.com) John Eastman maintained a not‑guilty plea and never entered into a cooperation deal.(en.wikipedia.org) So the envisioned chain—use Ellis to flip Giuliani, then Giuliani to flip on Trump—never occurred.
• The Georgia case ended without any trial of Trump, so the projected use of this testimony at his trial never happened. After DA Fani Willis was disqualified, new prosecutor Pete Skandalakis assumed control and, on November 26, 2025, moved to drop all remaining charges against Trump and his co‑defendants, formally ending the Georgia election‑interference prosecution.(reuters.com) As a result, there was no opportunity to test Sacks’ scenario that only higher‑level flips could provide incriminating evidence sufficient to convict Trump.

Overall, while Sacks correctly foresaw that lower‑level defendants would take plea deals, his core assertions—that they lacked incriminating information on Trump, that they were mainly being used to flip Giuliani and Eastman, and that this flip‑up strategy would define the case against Trump—were not borne out by subsequent developments. Hence, the prediction is best classified as wrong.