๐ World Watch
What the people who actually move things โ Elon, BlackRock, JPMorgan, Goldman, the Saudis, the Chinese state, the Rothschilds, the Mars and Walton dynasties โ are betting on. The substrate of the next economy. Updated as the picture sharpens.
01 The single thesis under everything
Strip the personalities. Every serious actor โ Elon, BlackRock, JPMorgan, Goldman, the Gulf kingdoms, the Chinese state, the Indian conglomerate families, the European old money โ is betting the same hand: own the scarce inputs of the next economy. Not the apps. The substrate.
| Scarce input | Who is stacking it |
|---|---|
| Compute | Microsoft, Meta, Amazon, Google (combined $300B+ capex/yr), Stargate JV, Saudi/UAE sovereign AI funds, NVIDIA itself |
| Energy | SMR nuclear (TerraPower, Oklo, Helion), grid/peakers (Vistra, Constellation now hyperscaler-locked), Aramco, China's coal + renewables wedge |
| Critical minerals | DRC cobalt, Indonesian nickel, Chilean lithium, China's processing monopoly, US IRA-backed reshoring |
| AI talent | $1โ10M comp ceilings, India's IIT pipeline flowing west, China retention pacts |
| Land + grid + permits | Hyperscaler land grabs in N. Virginia, Texas, Ireland, Singapore, Riyadh |
| Capital flow itself | BlackRock, Vanguard, State Street collectively own ~22% of every S&P 500 company. The new sovereigns aren't states. They're index funds. |
| Distribution / attention | Meta, TikTok, YouTube, X โ political base + labor market discovery + ad rent |
| Defense + dual-use | Anduril, Palantir, SpaceX, Shield AI โ the line between defense contractor and consumer tech is gone |
The product layer commoditizes. The substrate compounds.
02 What happens โ by horizon
1โ10 years (high confidence)
- AI eats knowledge work. Coding, drafting, claims, basic legal, customer service. The companies that absorb it win the margin; the ones that fight it lose customers.
- Energy becomes the binding constraint. Datacenter buildout exceeds grid capacity in 7 of the top 10 markets by 2027. Texas + UAE become AI capitals because they can build power.
- Robotics goes physical. Humanoids deploy in factories 2026โ2028 (Figure, Optimus, Unitree, Apptronik). Ag + warehouse autonomy already scaled.
- Drones become routine. Ukraine showed a $500 quad can kill a $5M tank. Every military and police force is buying in. Civilian last-mile delivery scales 2027โ2029.
- Biotech compresses. Drug discovery 10y โ 3-5y. GLP-1s reshape pharma. CRISPR Casgevy approved; gene-edit cures for sickle-cell + thalassemia routine by 2030.
- Asia coalesces. India hits demographic peak; China starts shrinking. Both build sovereign AI stacks. USโChina decoupling continues.
- Crypto financializes everything. Real-world asset tokenization, stablecoin trade settlement, CBDCs in 15+ countries by 2030.
- Climate shows up as supply shocks โ wheat, coffee, semiconductors, water โ before it shows up as apocalypse narrative.
10โ30 years (medium confidence)
- AGI-ish systems are baseline infrastructure. Like cloud is today.
- The labor question becomes existential. UBI, sovereign dividend models, work-as-purpose-not-income. Most rich countries try variants. Most fail the first 2-3 attempts.
- Climate adaptation is the largest industry on earth. Bigger than oil was. Carbon removal, water infra, climate-resilient ag, geoengineering R&D.
- Africa is 25% of humanity by 2050. Nigeria passes the US in population. The AIโAfrica interaction defines the second half of the century.
- Longevity stratifies. Upper-middle-class lifespan creeps to 90โ100. The new class divide is years lived, not money.
- Space economy is small but real โ Starship cadence makes Mars/lunar industry plausibly profitable around 2035โ2040.
- BCI for niche medical uses by 2035, mainstream consumer ~2045.
30โ100 years (low confidence โ but the shape is visible)
- Either AGI restructures civilization, or one of the tail risks hits. Informed people put "transformative discontinuity by 2070" at 30โ60%.
- If the good path: most people don't have "jobs" in the 20th-century sense. They have purpose-economies (creative, communal, spiritual, status).
- The dominant companies in 2126 are not today's giants. They're things being founded in 2026โ2050 by people who saw the substrate game early.
03 Top 10 countries โ by 2050
| Country | Best case | Worst case |
|---|---|---|
| USA | AI + biotech superpower, military dominant | Civil cold war, debt crisis, populist drift |
| China | Compute + manufacturing co-superpower, demographic adjusted | Stagnation, real-estate collapse, Taiwan miscalc |
| India | Demographic + AI services + manufacturing rising power | Failed urban/institutional scale-up |
| EU | Regulated stability, scientific output | Stagnation, energy + security dependent on US |
| Japan | Robotics leader, managed decline | Acute demographic collapse |
| UK | Financial + AI research hub | Slow drift |
| Germany | Industrial AI, late but solid | Energy crisis, deindustrialization |
| Brazil | Ag + green energy giant, ESG arbitrage | Political volatility |
| Indonesia | Under-rated. 280M people, nickel, AI-curious | Resource curse |
| Saudi / UAE | Energy โ sovereign AI fund โ global capital allocator | Oil-revenue cliff if transition fails |
Wild cards: Russia (energy/military), Israel (cyber/biotech), Korea (semis), Vietnam (manufacturing pivot), Mexico (US nearshoring).
04 Top 100 companies โ seed list
By category โ every entry here is a node in the substrate. List grows over time as we watch them.
AI infrastructure
NVIDIA ยท Microsoft ยท Alphabet ยท Meta ยท Amazon (AWS) ยท OpenAI (private) ยท Anthropic (private) ยท xAI ยท TSMC ยท ASML ยท AMD ยท Broadcom ยท Cerebras ยท SambaNova ยท Groq ยท Tenstorrent ยท CoreWeave ยท Lambda ยท Together AI ยท Mistral ยท Cohere ยท Databricks ยท Snowflake ยท Hugging Face
Energy
Saudi Aramco ยท ExxonMobil ยท Chevron ยท Shell ยท BP ยท TotalEnergies ยท NextEra ยท Vistra ยท Constellation ยท Iberdrola ยท Enel ยท CNPC ยท Sinopec ยท Reliance ยท Adani Green ยท TerraPower ยท Oklo ยท Helion ยท Cameco
Finance + capital allocators
BlackRock ($11T AUM) ยท Vanguard ($9.5T) ยท State Street ($4.3T) ยท Fidelity ($5T) ยท JPMorgan Chase ยท Goldman Sachs ยท Morgan Stanley ยท Bank of America ยท Citi ยท HSBC ยท UBS ยท Bridgewater ยท Citadel ยท Renaissance Technologies ยท Apollo ยท KKR ยท Blackstone ยท Carlyle ยท Berkshire Hathaway
Defense + dual-use
Lockheed Martin ยท RTX ยท Northrop Grumman ยท General Dynamics ยท BAE Systems ยท Anduril (private) ยท Palantir ยท SpaceX (private) ยท Shield AI (private) ยท Saab ยท Leonardo ยท L3Harris ยท Boeing Defense ยท Hanwha ยท HAL (India)
Asia conglomerates
Tencent ยท Alibaba ยท ByteDance (private) ยท Baidu ยท Samsung ยท SK Hynix ยท SoftBank ยท Reliance Industries ยท Tata Sons ยท Adani Group ยท Toyota ยท Hon Hai/Foxconn ยท BYD ยท CATL ยท MediaTek
Pharma + biotech
Eli Lilly ยท Novo Nordisk ยท Pfizer ยท Roche ยท Johnson & Johnson ยท Merck ยท AstraZeneca ยท GSK ยท Sanofi ยท Novartis ยท Moderna ยท Vertex ยท Regeneron ยท Recursion ยท Insitro ยท Isomorphic Labs
Auto + mobility
Tesla ยท Toyota ยท BYD ยท Mercedes-Benz ยท VW ยท GM ยท Ford ยท Stellantis ยท Honda ยท Hyundai ยท Geely ยท Waymo (Alphabet) ยท Aurora
Healthcare delivery + payors
UnitedHealth Group ยท CVS Health ยท Cigna ยท Humana ยท Anthem (Elevance) ยท Walgreens ยท HCA ยท Tenet ยท DaVita
Critical minerals + materials
BHP ยท Rio Tinto ยท Glencore ยท Vale ยท Anglo American ยท MP Materials ยท Albemarle ยท SQM ยท Pilbara Minerals ยท Tianqi Lithium
Distribution + attention
Meta Platforms ยท Alphabet (YouTube) ยท TikTok/ByteDance ยท X Corp ยท Snap ยท Reddit ยท Spotify ยท Disney ยท Netflix ยท Comcast ยท Warner Bros Discovery
Consumer/luxury concentration
LVMH ยท Hermรจs ยท Kering ยท Richemont ยท L'Orรฉal ยท Estรฉe Lauder ยท Inditex (Zara) ยท Nestlรฉ ยท Unilever ยท Procter & Gamble ยท Coca-Cola ยท PepsiCo ยท Mars (private) ยท Walmart ยท Costco ยท Amazon Retail
05 Ultra-wealth โ the named ones
Approximate net worth bands as of mid-2026. No true trillionaire exists yet; Musk is the closest. Money compounds; attention follows it.
| Name | Approx | Source |
|---|---|---|
| Elon Musk | $400โ500B | Tesla, SpaceX, xAI, X, Neuralink, Boring Co. |
| Bernard Arnault & family | $200B | LVMH (Louis Vuitton, Dior, Tiffany, โฆ) |
| Jeff Bezos | $220B | Amazon, Blue Origin |
| Mark Zuckerberg | $200B | Meta Platforms |
| Larry Ellison | $200B | Oracle, real estate, biotech |
| Larry Page | $150B | Alphabet |
| Sergey Brin | $140B | Alphabet |
| Warren Buffett | $140B | Berkshire Hathaway |
| Bill Gates | $130B | Microsoft + diversified + philanthropy |
| Steve Ballmer | $120B | Microsoft heritage |
| Michael Bloomberg | $100B | Bloomberg LP (terminal monopoly) |
| Mukesh Ambani | $110B | Reliance Industries (oil/telecom/retail) |
| Gautam Adani | $80B | Adani Group (ports/energy/infra) |
| Carlos Slim | $80B | Amรฉrica Mรณvil + diversified |
| Franรงoise Bettencourt-Meyers | $100B | L'Orรฉal heiress |
| Mohammed bin Salman (estimated) | state-linked, ~$1T proxied via PIF | Saudi sovereign control |
06 Dynasties โ the ones most people don't see
Old money + low publicity โ low influence. Many of these clans run multi-generational family offices that allocate quietly across PE, real estate, and emerging tech.
| Family | Approx total | Base | What they control |
|---|---|---|---|
| Walton family | $430B | USA | Walmart (50%+), diversified |
| House of Saud | $1โ2T (state-linked) | Saudi Arabia | Aramco, PIF, oil revenues, Mecca |
| Mars family | $140B | USA | Mars Inc., Wrigley (private) |
| Koch family | $130B | USA | Koch Industries (private) |
| CargillโMacMillan family | $60B+ | USA | Cargill (world's largest private co.) |
| Hermรจs family | $150B | France | Hermรจs International |
| Albrecht family | $80B | Germany | Aldi Nord + Sรผd + Trader Joe's |
| Quandt family | $40B | Germany | BMW (~46% of voting shares) |
| Agnelli family | $15โ20B | Italy | Exor (Ferrari, Stellantis, Juventus, The Economist) |
| Wertheimer family | $80B | France | Chanel (private) |
| Al Nahyan (Abu Dhabi) | $300B+ proxied | UAE | ADIA, Mubadala, ADNOC |
| Al Maktoum (Dubai) | $25B+ proxied | UAE | Dubai Holding, Emirates, DP World |
| Al Thani (Qatar) | $500B+ proxied | Qatar | QIA โ Harrods, Heathrow, VW stakes, US/UK real estate |
| Rothschild family | $5โ30B visible ยท private banking network | UK/France/CH | Rothschild & Co, Edmond de Rothschild, multiple branches |
| Rockefeller family | $11B+ visible | USA | Rockefeller Capital, philanthropy, energy/finance heritage |
| DuPont family | $14B+ | USA | DuPont, chemicals, real estate |
| Mellon family | $11B | USA | BNY Mellon, Gulf Oil heritage |
| Pritzker family | $40B | USA | Hyatt, TransUnion, Marmon Group |
| Newhouse family | $30B | USA | Advance/Condรฉ Nast/Reddit stake |
| Cox family | $30B | USA | Cox Communications + automotive |
| Hearst family | $28B | USA | Hearst Communications |
| Murdoch family | $25B | USA/UK/AU | News Corp, Fox |
| Ambani family | $110B | India | Reliance โ oil, telecom (Jio), retail |
| Adani family | $80B | India | Ports, power, airports, green |
| Tata family + Trusts | $30B (+$165B group revenue) | India | Tata Sons โ 30+ public cos; Trusts own 66% of Sons |
| Birla family | $20B | India | Aditya Birla Group (cement/metals/financial) |
| Kwok family | $15B | Hong Kong | Sun Hung Kai Properties |
| Li Ka-shing & family | $35B | Hong Kong | CK Hutchison, ports, telecoms |
| Chearavanont family | $35B | Thailand | Charoen Pokphand โ feed, retail, telecom |
| Suharto / Salim / Hartono families | various tens of $B | Indonesia | BCA, Djarum, agribusiness |
| Sy family | $15B | Philippines | SM Investments (retail + property + finance) |
07 Meta-rules โ what binds the world together
The actual operating system of global commerce and security. Most of it is one Wikipedia tab away; most people never open the tab.
Money flow
- Bretton Woods (1944) โ USD as global reserve. Weakening but not replaced. BRICS+ pushes for alternative settlement.
- SWIFT โ interbank messaging. US/EU control. Russia kicked off it in 2022. Alternatives: CIPS (China), SPFS (Russia), INSTEX (failed EU attempt).
- Basel III โ global banking capital reserve rules. Enforced via national regulators.
- FATF โ money-laundering watchdog. Grey/black lists kill capital access for "non-compliant" states.
- BIS (Bank for International Settlements) โ central banks' central bank. Most powerful institution most people have never heard of.
- IMF + World Bank โ development financing with strings (Washington Consensus).
- Big Three rating agencies โ Moody's, S&P, Fitch. They decide what sovereign + corporate debt costs to issue.
- Big Four auditors โ Deloitte, PwC, EY, KPMG. The fingerprint on every public company's financials.
Trade + security
- WTO โ trade dispute framework. Weakening since 2018.
- NATO โ 32 countries, Article 5 mutual defense.
- AUKUS โ US/UK/AU Pacific defense pact.
- QUAD โ US/Japan/India/AU informal alignment.
- BRICS+ โ alternative bloc (now 10+ members).
- OPEC+ โ oil supply cartel including Russia.
- UN Security Council P5 โ US/UK/France/Russia/China veto.
- CHIPS Act + IRA + EU Chips Act โ semiconductor reshoring, $200B+ subsidy floor across US/EU.
- Export controls โ US controls advanced-node chips to China (EAR 7nm cap, FDPR).
Tech + AI governance
- ICANN โ internet name + address authority.
- EU AI Act (2024) โ first major AI regulation; risk-tiered.
- US AI Executive Orders โ chip controls, model-reporting thresholds.
- UK AI Safety Institute ยท US AISI ยท Singapore AI Verify โ evals + red-teaming.
- Bletchley + Seoul + Paris Summits โ multilateral AI safety alignment (loose).
Climate + health
- Paris Agreement (2015) โ 1.5ยฐC target. Failed in practice; the framework persists for offsets + ESG.
- IPCC โ scientific consensus body.
- WHO โ health, pandemic preparedness (weakened post-COVID by US/China tensions).
- IAEA โ nuclear safeguards.
Quiet rooms
- World Economic Forum / Davos โ annual elite alignment ritual.
- Bilderberg โ invite-only Atlantic elite. Less powerful than legend; more powerful than they admit.
- Trilateral Commission ยท Council on Foreign Relations ยท Chatham House โ long-cycle policy formation.
- Aspen Institute ยท Brookings ยท CSIS ยท RAND โ DC's mind-share machinery.
- YPO ยท TED ยท Sun Valley ยท Allen & Co. โ where deals get done before the SEC sees them.
08 Sectors โ winning vs losing
- AI infrastructure (compute, model serving, agent platforms)
- Energy: nuclear (SMR), grid, storage
- Robotics & autonomy (drones, humanoids, ag, defense)
- Healthcare AI (denial coding, prior auth, diagnostics)
- Biotech, longevity, gene editing
- Defense + dual-use tech
- Cybersecurity
- Sovereign AI / on-prem inference
- Critical minerals + processing
- Climate adaptation infrastructure
- Tokenized real-world assets
- Creator economy / owned-audience media
- BPO / call centers / generic RCM (margins compress fast)
- Traditional retail
- Print + broadcast media
- Insurance back-office
- Entry-level legal, translation, transcription
- Generic SaaS (commodified by AI builders)
- Brick-and-mortar banks
- Cable, broadcast TV
- Single-stack ERP / CRM monoliths
09 Where we (Sriram) sit
Already on the right side of the curve. The day-to-day grind hides it.
- Getmax = AI-native, healthcare AI is winning. Generic RCM is losing โ but the version that uses AI to do the work is on the winning list. That's why we exist and why DFS firing us is survivable.
- FlyMax = drones, autonomy, dual-use. Winning column. Public-ready MIT repo sits at github.com/jakesparow/flymax-orchestrator awaiting brand decision.
- Voice OSS migration = moving from price-taker (ElevenLabs floor) to substrate-owner (own RunPod-hosted stack). BlackRock pattern at small scale.
- NVIDIA Inception = aligning to the compute kingmaker before they pick winners.
- Sarvam credits = India-language voice. Demographic-dividend bet.
- Embedded / home AI lab = personal sovereignty over compute. Bezos building rocket factories logic, in miniature.
- War-room visualizer = agentic orchestration substrate. Next-decade SaaS gets built on it.
The single sharpest move in the next 90 days: make the dual-use case airtight. The same orchestrator brain runs healthcare RCM agents AND the drone fleet. That's the Series B narrative nobody else has โ and it's worth a 5-10x valuation premium.
The bigger risk for the next 30 years isn't lifespan โ longevity tech compresses that ceiling further. The risk is whether the next decade feels meaningful. That's the only question that matters and it's in your hands.