WHO CONTROLS
THE CAPITAL?

SECTION 00 — THE CAPITAL ARCHITECTURE

MONEY MOVES THROUGH STRUCTURES

Every technological empire is built on a financing architecture. Before there is a model, a data center, or a product, there is a decision — made somewhere, by someone, with someone else's money or with public money — about which ambition deserves to be resourced. That decision is never visible in the technology itself. It is visible only in the structures through which capital moves.

The original sources of capital are the places money begins: household savings, bank deposits, pooled institutional funds, corporate balance sheets, government budgets, philanthropic endowments, and state-held wealth. Most people participate in this layer directly — through pensions, bank accounts, insurance policies, and taxes — usually without knowing that their money is, somewhere downstream, underwriting the construction of artificial intelligence.

Above the sources stand the allocators — the people and institutions that decide which ventures receive capital and on what terms. And connecting sources to ventures stand the instruments: equity, debt, guarantees, contracts, credits, and structured vehicles. Each layer carries different terms, different disclosure rules, and a different balance of public benefit and private gain.

THE ALLOCATORS

THE INSTRUMENTS

SECTION 01 — THE NINE QUESTIONS

NINE QUESTIONS, ONE INVESTIGATION

This investigation is organized around nine questions. Each can be asked of any layer of the capital architecture — and each leads to the same underlying issue: whether the financial system that is building artificial intelligence is one that a self-governing public can see, enter, and hold accountable.

The questions are sequential in logic but simultaneous in practice. Who provides the capital shapes who allocates it; who allocates it shapes what the capital costs; what it costs shapes who can afford to build; and who can afford to build shapes whose technology the future runs on.

SECTION 02 — QUESTION ONE

WHO PROVIDES THE CAPITAL?

The deepest layer of the capital architecture is occupied by savers, not investors. Pension funds, insurance portfolios, bank deposits, endowments, and tax revenues are, in accounting terms, other people's money — accumulated from wages, premiums, tuition, and taxes. When that money is committed to a technology fund or a data-center bond, the commitment is made by professionals on behalf of people who will never see the term sheet.

This distance is not inherently illegitimate — it is how modern finance works, and professional stewardship can protect savers. But distance has consequences. A household whose pension is invested in a fund that backs a frontier laboratory has no vote on the laboratory's conduct, no visibility into its governance, and no practical means of exit short of leaving the pension system itself. Beneficial ownership without voice is the quiet foundation of concentrated technological power.

Public capital is a distinct source with distinct rules. Government research funding, procurement, loans, guarantees, and tax incentives have built substantial parts of the technological commons — and TECHNO-STATE CAPTURE and TECHNO-WELFARE ROYALTY document in detail how public money becomes private value, and what category discipline that documentation requires. This investigation builds on that work rather than repeating it.

Finally, there is the capital almost everyone holds and almost no frontier venture ever receives: household savings, community savings, and small-dollar investment. It is the most broadly held money in the architecture and the least able to reach frontier scale. The distance between those two facts is one of the central subjects of this investigation.

SECTION 03 — QUESTION TWO

WHO ALLOCATES THE CAPITAL?

Between the sources and the ventures stands a comparatively small population of allocators: general partners and their investment committees, corporate venture arms, program officers, foundation boards, and sovereign fund governance. The proportion of the public's money that passes through rooms most of the public will never enter is extraordinary — and the composition of those rooms determines, more than any algorithm, which futures get financed.

Allocation is discretion plus pattern. An allocator's choices are shaped by mandate, incentive structure, career history, professional networks, and the sheer mechanics of evaluating thousands of opportunities with finite attention. None of this requires bad intent to produce concentration: familiarity, proximity, and shared background are efficient screening heuristics, and heuristics compound into pattern.

The discipline this investigation applies is to separate documented patterns from assumed motivations. Where a pattern — in who gets meetings, who gets term sheets, who gets follow-on funding — can be established from records, it will be documented with its source. Where a motivation is only inferable, it is stated as a question, not a finding. Allocators are entitled to the same evidentiary discipline the investigation demands elsewhere.

Limited partners sit one layer up, and their mandates matter: a pension fund that directs its managers toward a strategy, an endowment that sets ESG constraints, or a sovereign directive that channels national wealth into domestic champions are all allocation decisions made before any deal sheet exists. Who controls the capital begins with who writes the mandates.

SECTION 04 — QUESTION THREE

WHAT DOES THE CAPITAL COST?

Capital has a headline price and a real price. The headline price is the valuation printed in a funding announcement. The real price is everything a founder gives up to receive the money: ownership percentage, control rights, information rights, board composition, veto rights, liquidation preferences, option-pool refills, and the acceleration of expectations that comes with each successive round.

The distinction between ownership and control is the foundation of this section. A founder can retain a majority of shares and still lose effective control — through a board the investors appoint, protective provisions that veto fundamental decisions, or a liquidation preference that makes later common equity worth less than the cap table suggests. Legal control and economic outcome diverge in ways that are routinely invisible at signing and decisive at exit.

The costs compound across rounds. Each new financing dilutes existing holders; each preference stack ranks earlier money ahead of later common; each down round reprices everything that came before it. Founders can be removed from companies they founded, missions can drift toward whatever the next round requires, and exit pressure can push a company toward acquisition or monetization strategies its original purpose never anticipated.

For companies building on public contracts or regulated infrastructure, another cost appears: the obligations that come with government customers, and the scrutiny that follows them. And for founders whose communities have historically been excluded from these rooms, the cost of capital includes the cost of access itself — the networks, track records, and introductions that gate the first meeting. WHO GETS FUNDED? WHO GETS EXCLUDED? takes that up directly in Section 05.

The calculator below is an educational illustration: a simplified, single-class model that shows how dilution, preferences, and exit mathematics interact. It is not a simulation of any actual company, and it is not legal or financial advice. Its assumptions are displayed so they can be examined — which is precisely the transparency the real term sheet rarely offers at signing.

SECTION 05 — QUESTION FOUR

WHO GETS FUNDED? WHO GETS EXCLUDED?

This section is foundational to the entire investigation, because the pattern of who receives capital is the pattern of who gets to build. It examines racial, ethnic, gender, disability, socioeconomic, and geographic disparities in technology financing — and it does so with a discipline the subject demands: specific attention to specific communities, without collapsing distinct experiences into broad demographic categories.

That specificity matters. The experiences of Black founders, Latino founders, Asian founders — whose communities are not a single homogeneous category and whose outcomes differ across national, ethnic, and immigrant lines — Native American, Alaska Native, and other Indigenous founders, women founders, and founders with disabilities are distinct, and the evidence about each must be examined on its own terms. For Indigenous communities specifically, questions of tribal sovereignty and Indigenous data sovereignty are substantive research requirements, not demographic footnotes: who owns, governs, and interprets data about Indigenous peoples is itself a capital-adjacent question.

Measurement discipline comes first, because this field is full of numbers that do not mean what they appear to mean. A funding share is not an approval rate. A company count is not a dollars total. An early-stage statistic is not a late-stage one. Investor demographics are not founder demographics. Follow-on rates, ownership retention, and exit outcomes are separate measurements that often move in different directions. This investigation will state which measure is being reported — or will not report the number at all.

The underlying datasets carry their own caveats, and this investigation discloses them rather than laundering them: much venture demographic data is self-reported, coverage is partial, definitions of race, ethnicity, and gender vary across sources and over time, and intersectional categories are frequently too small for stable statistical claims. Where a published dataset will support a documented comparison, it will be cited with its methodology and retrieval date. Where it will not, the comparison is marked PENDING RESEARCH rather than estimated.

What can be stated now, as framework rather than figure: the financing of frontier technology is concentrated in a narrow band of the population and a short list of metropolitan geographies; the communities least represented among funded founders are, in many cases, the communities least represented in the datasets of the systems being built; and the consequence compounds — because Section 08 shows how financing exclusion propagates backward into the entire opportunity pipeline.

SECTION 06 — QUESTION FIVE

WHO MAKES THE INVESTMENT DECISIONS?

Allocation looks like an industry and operates like a set of rooms. Inside a venture firm, the decision path runs from associate screens to partner meetings to consensus or conviction votes; inside a corporation, from business unit to development committee to board; inside a foundation, from program staff to board resolution; inside a sovereign fund, through governance whose final authority is political. Each path has its own thresholds, its own veto points, and its own culture of who speaks and who decides.

The documented mechanics matter because they are the difference between plural capital in the aggregate and concentrated discretion in practice. A thousand limited partners can fund a fund; a handful of partners still decide every deal. A university endowment can span hundreds of managers; an investment office still signs the commitments. The architecture aggregates many sources into few decisions — and the composition of the deciding group is therefore the single most consequential variable in the entire pipeline.

Networks are the visible surface of this discretion. Sourced deals arrive through introductions; pattern recognition runs on familiar profiles; reference checks travel the same professional circuits. These mechanisms are efficient and human, and they are also how the demographics of one decade become the demographics of the next. The investigation documents network effects where records support them, and resists the temptation to infer motive where only pattern is visible.

The governance question follows directly: are investment decisions reviewable — by limited partners through advisory rights and side letters, by regulators through disclosures, by the public through anything at all? For public capital the answer is at least statutorily defined. For private capital it is mostly a matter of contract, which means it is mostly a matter of leverage.

SECTION 07 — QUESTION SIX

WHO GETS TO BUILD THE INTELLIGENCE THAT SHAPES OUR FUTURE?

Frontier artificial intelligence is built by a small set of institutions, and public discussion routinely blurs the different things that can be controlled inside them. This investigation insists on a six-way distinction, because the six dimensions do not move together — an institution can be founder-owned and investor-governed, technically led and infrastructure-dependent, or any combination in between.

The named laboratories — OpenAI, Anthropic, Google DeepMind, Meta AI, xAI, Microsoft AI, Cohere, World Labs — and the international public and academic institutions entering this field will each be examined case by case against these six dimensions, using their own disclosed structures: charters, filings, published governance documents, and stated control arrangements. Ownership structures are never presumed equivalent to governance structures, and governance structures are never presumed equivalent to operational control.

The stakes of the distinction are concrete. Whoever holds technical leadership shapes the research agenda. Whoever holds board control can replace the executives, redirect the mission, or approve deployment the researchers oppose. Whoever owns the compute can make independence expensive. A public that cannot distinguish these dimensions cannot evaluate any claim about who is responsible for what a frontier system does.

SECTION 08 — QUESTION SEVEN

THE DIGITAL OPPORTUNITY PIPELINE

This is the central narrative of the investigation, and it reframes the title question. Who controls the capital that built today's artificial intelligence is one question. The other is who has the opportunity to acquire capital and build competing technological institutions — because concentration is not only about who owns what exists, but about who is positioned to own what comes next.

The pipeline runs from broadband access through education, technical expertise, professional networks, and access to capital and computing resources, to founder ownership and finally to a seat in the governance of frontier systems. Each stage compounds the one before it: a household without broadband produces a student without digital fluency, a graduate without networks, an entrepreneur without introductions, and a community absent from the rooms where the next decade's institutions are financed and governed.

Breakage at any stage propagates downstream, which is why access policy is capital policy. The documented record of recent years includes: the Affordable Connectivity Program — the federal benefit that helped millions of low-income households pay for internet service — exhausting its funding in 2024, with no continuation appropriation; the cancellation of Digital Equity Act programs in 2025, followed by legal challenges whose outcomes remained pending as of this investigation's retrieval date; and restructuring actions at the Department of Education affecting the administration of programs that feed the pipeline's education stage. Each of these is a documented action with stated objectives; measured consequences are still accruing, and claims about motivation that go beyond the stated record are treated as unproven, not adopted.

The pipeline framework also explains why capital exclusion and educational exclusion are the same investigation at different depths. WHO CONTROLS OUR EDUCATION? documents the systems now mediating what children learn; WHO CONTROLS OUR WORK? documents the labor systems the pipeline feeds into. What this investigation adds is the financing stage: the moment when accumulated opportunity either converts into ownership and governance — or does not.

SECTION 09 — QUESTION EIGHT

WHOSE TECHNOLOGY GETS BUILT?

Capital allocation is technological design conducted in advance. Long before any product decision is made, the pattern of funding has already shaped which problems are considered worth solving, which teams exist to solve them, and which deployment contexts the resulting systems will and will not understand. This section investigates that connection explicitly.

The inquiry proceeds along six threads, and it proceeds carefully: nothing here implies that demographic identity alone determines the quality or safety of a technology. Systems built by homogeneous teams are not automatically worse, and systems built by diverse teams are not automatically better. The question is structural, not essentialist: who is present when the decisions are made, and what does the absence of entire communities from those rooms predict about the systems that result?

The threads are examined case by case in the case files and the map: which technological problems receive investment; which communities participate in product development; who determines research and product priorities; who controls training data and model deployment; whether underserved communities can finance independent alternatives; and who ultimately owns and governs the technologies that result. Where the evidence for a connection is documented, it is documented; where it is only suggestive, it is marked as an investigative question rather than a finding.

SECTION 10 — QUESTION NINE

WHO PROFITS? WHO PAYS WHEN THE SYSTEM FAILS?

Every capital architecture has an upside ledger and a downside ledger, and they are rarely held by the same hands. The upside of the AI buildout accrues to equity holders, fund investors, and in the largest outcomes, to a very small number of individuals. The downside — if the buildout's economics do not close — lands wherever the obligations were signed.

The financing structures now assembling around AI infrastructure deserve specific scrutiny on this point: debt raised against data centers, special-purpose vehicles holding compute assets, GPU-backed financing, and long-term power commitments. These are ordinary tools of infrastructure finance, and they are also the mechanisms by which concentrated speculation can become distributed exposure. Direct exposure — investors and lenders who chose the risk — must be kept distinct from indirect exposure: pension beneficiaries whose funds hold the paper, utility ratepayers who absorb power commitments, taxpayers who stand behind guarantees, and communities that host the physical infrastructure.

The accountability question completes the investigation's nine: when a system fails — financially or functionally — who is answerable? For public capital, statutes and inspectors general at least define an answer. For private capital structured through layers of vehicles, the answer is whatever the contracts say it is. WHO PROFITS? WHO PAYS? is therefore not a rhetorical question but an audit trail that this investigation will build record by record — and, where the records are not yet assembled, will mark PENDING RESEARCH rather than estimate.

SECTION 11 — CASE FILES

NINE FILES, NINE DISCIPLINED INQUIRIES

The case files organize the investigation's specific inquiries. Each file is bounded — a subject, a scope, an approach stating which records would establish or refute its questions, and a set of open questions. Each documented item, as the evidence base grows, will carry its classification, its source, and its date. Nothing enters a case file without that discipline.

The roster is deliberately structural rather than adversarial: a supplier that invests in its own customers; a firm whose founder's capital, board seats, and political activity require separate ledgers; a doctrine-driven fund; the builder-investor roster; the infrastructure investors; a hedge fund and its affiliated market-maker; a sovereign-adjacent fund; the sovereign wealth funds of the Gulf and beyond; and the hyperscalers' dual role as operators and financiers of the laboratories they supply. Each is examined with the same standard: facts only, dated, sourced.

Where public reporting raises questions about leadership conduct or institutional controversies, that material is confined to the case files, held to the documentation standard, and never rendered as established fact without records behind it. Ordinary commercial relationships are not misconduct, and this investigation does not insinuate otherwise. What it does insist on is that the public can see the structure — because structure, unlike motive, is verifiable.

SECTION 12 — THE ORIGINAL CONTRIBUTION

THE NETWORK OF POWER AND THE NETWORK OF EXCLUSION

DSC's original contribution to this subject is the insistence that two networks must be mapped together. The Network of Power is the documented web of relationships among capital allocators, technology companies, and governing institutions — who funds whom, who sits on whose board, whose money moves through whose vehicles. The Network of Exclusion is the demographic and geographic record of who is absent from those relationships — who has never been in the room, across every stage of the opportunity pipeline.

Mapped alone, the Network of Power reads as a story about a few actors. Mapped alone, the Network of Exclusion reads as a story about missing statistics. Mapped together, they are one finding: the same architecture that concentrates decision-making also reproduces the conditions of its own exclusivity — because capital follows networks, networks follow opportunity, and opportunity follows capital. Breaking that cycle requires intervening at pipeline stages, not merely at deal stages.

This framework is DSC's own. It is stated here as an interpretive framework — a structure for organizing evidence, not a measurement. Where other published work examines adjacent terrain, this investigation credits it explicitly and verifies independently. That discipline is set out in EVIDENCE AND SOURCING below.

SECTION 13 — THE MAP

THE CAPITAL, OPPORTUNITY & CONTROL MAP

The map renders the two networks as a single navigable structure across six layers: MONEY (the flows of public and private capital), DECISION-MAKERS (the allocators and their mandates), EXCLUSION (the demographic and geographic record), INFRASTRUCTURE (the physical substrate — compute, power, and data centers — and who finances it), CONSEQUENCES (concentration, profit, exposure, and failure), and ALTERNATIVES (the models that can widen the circle).

The map is built as a static, evidence-based structure first. Every record on it carries its subject, its relationship, its evidence classification, its source, and the date the source was recorded. Records enter the map only when they meet the documentation standard; layers whose records are not yet assembled carry their research commitments openly rather than decoration.

Interactive exploration — selecting an entity, filtering by relationship type, viewing verified relationships only — is technically straightforward and deliberately deferred: it will be added when the verified dataset beneath it is substantial enough to be worth navigating. A map that renders guesses is worse than a map that admits what it does not yet know.

SECTION 14 — THE RESPONSE

ALTERNATIVES TO CONCENTRATED CAPITAL

If the pipeline can concentrate opportunity, it can also widen it — and the instruments for widening it already exist. Community development financial institutions, public banks, credit unions, employee ownership, revenue-based financing, evergreen patient capital, public venture programs, mission-aligned philanthropy, community investment vehicles, and shared open infrastructure are not theoretical proposals; they are operating models with real track records, real governance tradeoffs, and real constraints on scale.

The investigation examines each model honestly: how it is capitalized, how it is governed, whom it has reached, and what limits its growth — because an alternatives section that oversells its subjects serves concentration as surely as one that ignores them. None of these models is a complete substitute for the private venture architecture, and none needs to be. The claim is pluralist: a technology civilization financed through many channels is more resilient, more accountable, and more representative than one financed through a handful.

This is where the capital investigation joins DSC's constructive program. The Plurality of Digital Sovereignty — digital power not the exclusive domain of a handful of corporations, governments, or centralized systems — is a capital claim as much as a governance claim. The infrastructure alternatives examined in WHO CONTROLS OUR INFRASTRUCTURE? Section 4, the cooperative commons of the Civic Commons, and the civic intelligence capability proposed in the Computational Mosaic all presume financing structures that a broader public can actually enter. The pipeline is where that presumption either becomes real or remains rhetoric.

SECTION 15 — THE PEOPLE WHO FINANCE THE FUTURE

THE MONEY THAT CHOOSES WHAT GETS BUILT

The preceding sections traced where the capital for artificial intelligence comes from and where it goes. This section opens the next layer of the investigation: the people and institutions who decide where it lands. Venture investors are not merely a source of startup financing. Through investment selection, financial terms, board representation, governance arrangements, strategic guidance, executive networks, commercial introductions, government relationships, follow-on financing, exit strategies, policy advocacy, and access to computing infrastructure, they participate in deciding which founders are financed, which technologies are developed, and which business models survive.

CENTRAL INVESTIGATIVE QUESTION: WHO GETS TO FINANCE, BUILD, OWN, AND GOVERN THE INTELLIGENCE THAT WILL SHAPE HUMAN SOCIETY?

The investigation must consider the relationship between financial power, demographic opportunity, institutional access, technological development, and democratic accountability. It examines a relatively concentrated network of financial decision-makers and the structures through which their choices reach the frontier of artificial intelligence — and it does so with the same discipline that governs every other file in this investigation.

STANDING CAUTION: Different investors exercise different degrees of influence. This investigation does not assume that an investment automatically establishes corporate control, that a board seat establishes day-to-day authority, or that a public forecast establishes a future economic outcome. For each case study, the specific mechanisms of influence are identified separately, and each is supported by evidence before it is treated as fact.

THE TWELVE CHANNELS OF INVESTOR INFLUENCE — DOCUMENTED MECHANISMS THIS INVESTIGATION EXAMINES

SECTION 16 — INVESTOR CASE FILES

THE FINANCIERS, ON THE RECORD

Each featured case file below follows a fixed presentation standard: identity and institution; sources of capital; investment strategy; key investments; ownership and control; policy and government relationships; demographic funding questions; technological consequences; public benefits and risks; and evidence and sourcing. The structure is deliberately identical from file to file, because the discipline is the point — the same questions asked of every financier, and no file's importance resting on how it is formatted.

These records are staged as part of an ongoing verification process. Founding facts and corporate structures that are matters of extensive public record are labeled accordingly. Reported investment details are labeled REPORTED CLAIM until they are confirmed against primary records. Where the evidence for a complete profile is not yet assembled, the file is staged as awaiting verification rather than completed — an honest ledger is worth more than a confident one.

Each file expands in place. Nothing here is a judgment about the individuals profiled; it is a map of documented financial relationships held to the documentation standard.

THE RESEARCH ROSTER — ADDITIONAL FRONTIER AI INVESTORS

SECTION 17 — WHO GETS TO MAKE THE DECISIONS?

THE DEMOGRAPHICS OF VENTURE CAPITAL POWER

The question this section examines is structural: who occupies the seats from which capital-allocation decisions are made? The investigation covers venture fund founders, general partners, managing partners, investment committees, limited-partner decision-makers, institutional investment executives, corporate venture leadership, and sovereign-wealth investment leadership — the full decision stack between a founder's pitch and a funded company.

The distribution of decision-making authority is examined across racial, ethnic, gender, disability, educational, geographic, and socioeconomic dimensions wherever reliable evidence permits. The investigation also examines whether underrepresentation at the capital-allocation level is associated with disparities in investment outcomes — a question that requires documented evidence of decision-making structures and investment patterns, not inference.

MEASUREMENT DISCIPLINE: funding share is not approval rate, is not company count, is not stage mix, and is not ownership. These measures answer different questions and are never conflated. Communities are examined distinctly — Black, Latino, white, East Asian, Southeast Asian, South Asian, Native American, Alaska Native, and other Indigenous founders, together with relevant intersectional categories — and are never collapsed into a generic minority classification. No demographic figure enters this page without a named, dated dataset behind it.

STATUS: the demographic data pull for this expansion is staged, not complete. Widely cited industry reporting has placed the share of United States venture dollars going to Black-founded companies at roughly one percent in recent years, and the share going to women-founded teams at a small fraction of the total (REPORTED CLAIM — these figures require confirmation against the named primary datasets below before they are treated as established). The investigation does not present unavailable demographic data as zero representation; it identifies the gaps explicitly.

DECISION LAYERS UNDER EXAMINATION

NAMED DATASETS CONSULTED FOR VERIFICATION (PULL PENDING)

SECTION 18 — WHO GETS TO OWN THE INTELLIGENCE?

THE FRONTIER AI OWNERSHIP QUESTION

For each frontier laboratory, this investigation distinguishes six things that are routinely blurred together: FOUNDERSHIP (who founded it), INVESTMENT (who financed it), ECONOMIC OWNERSHIP (who holds its equity), BOARD AUTHORITY (who governs it), TECHNICAL LEADERSHIP (who directs its research), and COMPUTE DEPENDENCE (whose infrastructure it runs on). A laboratory's story is incomplete — and often misleadingly told — when any of these is presented as another.

The economics matter as much as the entities. Establishing a competitive frontier AI operation requires capital and computing infrastructure at a scale that excludes otherwise qualified founders who lack access to institutional finance or advanced compute. The register below records the structural facts that are matters of public record; the investment and governance detail for each organization is staged for verification against primary sources.

The register does not assume that every frontier laboratory is independently incorporated, venture-funded, or governed in the same manner. Several of the most consequential AI research organizations are subsidiaries or divisions of the largest companies in the investigation's other files — a structural fact that changes what ownership means.

FRONTIER LABORATORY REGISTER — STRUCTURAL FACTS; INVESTMENT AND GOVERNANCE DETAIL PENDING VERIFICATION

SECTION 19 — WHO DECIDES WHAT GETS BUILT?

CAPITAL ALLOCATION AND TECHNOLOGICAL DESIGN

Financing decisions are design decisions made years before a product ships. What receives venture capital tends to get built; what does not tends not to. This section examines whether funding priorities influence the direction of entire domains — and, symmetrically, whether technologies addressing underserved communities encounter distinct financing barriers.

The investigation holds two cautions in tension. It does not presume that venture capital investment inherently produces harmful outcomes, and it does not treat absence as proof of conspiracy. What it documents is the pattern: which domains attract capital at scale, which struggle for it, and what the people allocating it say about why.

A final separation: the demographics of founders are one question; the design, governance, performance, and social consequences of their technologies are another. A founder's identity alone neither determines nor excuses what a technology does. Both questions are examined, and neither is reduced to the other.

DOMAINS UNDER EXAMINATION

SECTION 20 — FOLLOW THE NETWORKS

INVESTOR TO GOVERNMENT TO CONSEQUENCE

The relationship ledger below links INVESTOR → INVESTMENT FIRM → PORTFOLIO COMPANY → BOARD / GOVERNANCE → TECHNOLOGY → GOVERNMENT / MARKET → PUBLIC CONSEQUENCES. Every record identifies its relationship type — investment, founding role, board appointment, commercial partnership, government procurement, public policy advocacy, political contribution, employment relationship, or infrastructure dependency. Distinct relationships are never merged: a personal holding is not a fund position, a founding role is not a governance right, and a contribution is not influence.

The records below open the financiers' ledger. Founding facts and matters of extensive public record are classified DOCUMENTED FACT with the caveat that every record in this expansion is awaiting primary-source verification pass; reported transactions are classified REPORTED CLAIM. Records enter the main Capital, Opportunity & Control Map only when they meet the full documentation standard — these are staged here first, precisely so that their verification status travels with them.

Where public reporting raises questions about conduct, that material is confined to case files and held to the documentation standard. Ordinary commercial relationships are not misconduct, and this investigation does not insinuate otherwise.

THE CHAIN EVERY RECORD MUST SURVIVE

RELATIONSHIP TYPES — NEVER MERGED, ALWAYS IDENTIFIED

THE FINANCIERS' LEDGER — OPENING RECORDS

SECTION 21 — THE ORIGINAL SOURCES OF VENTURE CAPITAL

WHERE THE MONEY COMES FROM BEFORE THE CHECK

The trail does not end at the venture firm's name. Venture funds are vehicles; the capital inside them belongs to their limited partners. This section investigates where that capital originates: public pension funds, university endowments, family offices, foundations, sovereign wealth funds, institutional investors, corporate investment vehicles, wealth management firms, and high-net-worth individuals. Specific limited partners are identified only where verified disclosures permit — and undisclosed investor identities are never invented.

CENTRAL INVESTIGATIVE QUESTION: CAN THE RETIREMENT SAVINGS OR INSTITUTIONAL CAPITAL OF ORDINARY CITIZENS INDIRECTLY FINANCE TECHNOLOGIES THAT TRANSFORM THEIR EMPLOYMENT, PRIVACY, AND CIVIL RIGHTS?

This is a structural question, not an insinuation. Public pensions are fiduciaries with duties to their beneficiaries; endowments fund scholarships and research; sovereign funds answer to their states. The investigation examines the actual financial structures and their associated risks — concentration, transparency, and the distance between the saver and the decision — and leaves the alarm to the evidence.

LIMITED-PARTNER CATEGORIES UNDER EXAMINATION

SECTION 22 — THE COST OF CAPITAL AND FOUNDER INDEPENDENCE

WHAT FINANCING STRUCTURES DECIDE BEFORE ANY PRODUCT SHIPS

The terms of a financing round reach further than its bank account. This section connects the investor case files to the investigation's approved Founder Cost of Capital Calculator (Section 04) and examines how financing structures may affect founder ownership, board control, voting rights, liquidation preferences, financing flexibility, mission independence, commercial strategy, exit expectations, and employee compensation.

Where possible, documented financing examples are used. Confidential investment terms are never invented, and reported terms are labeled as reported. The purpose is structural literacy: a founder — or a citizen evaluating who will own the next frontier laboratory — should be able to read a term sheet's consequences before signing it.

WHAT THE STRUCTURE OF A ROUND DETERMINES

RETURN TO THE FOUNDER COST OF CAPITAL CALCULATOR (SECTION 04) ↑

SECTION 23 — WHO IS FUNDING THE FOUNDERS OTHERS OVERLOOK?

THE ALTERNATIVE INVESTMENT ECOSYSTEM

The investigation does not portray underrepresented founders solely as excluded populations. Alongside the barriers it documents structural alternatives: investment institutions focused on historically underfunded entrepreneurs, and the founders, investors, and financing models creating alternative pathways to ownership.

The register below is a research roster, not a verdict. Current operating status, leadership, investment strategy, and actual financing activity are verified before publication — several institutions in this space have changed scale or structure in recent years, and nothing is asserted here that has not been confirmed. What the section establishes now is the investigative frame and the questions it asks.

THE QUESTIONS THIS SECTION ASKS OF EVERY ALTERNATIVE INSTITUTION

RESEARCH ROSTER — OPERATING STATUS OF EACH SUBJECT VERIFIED BEFORE ANY PUBLICATION

SECTION 24 — INVESTOR IDEOLOGY AND PUBLIC POLICY

WHAT THE FINANCIERS SAY ABOUT THE RULES

Venture investors do not only finance companies; some of them advocate for the rules those companies will live under. This section examines publicly documented views on artificial intelligence regulation, technology competition, defense innovation, government procurement, labor automation, data privacy, public education, economic redistribution, open-source AI, national security, corporate concentration, and democratic accountability.

Three separations govern everything here. An investor's statements are distinguished from their firm's official position, and both are distinguished from the actions of portfolio companies. Ideological affiliations are never assigned without supporting evidence. And the purpose is analytical, not adversarial: to understand how financial incentives and publicly advocated policies may intersect — not to portray all venture capitalists as sharing one political agenda.

Direct, verifiable sources are used wherever possible. Where an investor's position is documented in their own writing, speeches, or filings, it is cited as such; where it is characterized by press coverage, it is labeled as reported. The line between an argument and its financial context is drawn in every entry.

POLICY DOMAINS UNDER EXAMINATION

EVIDENCE AND SOURCING

HOW THIS INVESTIGATION KNOWS WHAT IT KNOWS

This investigation holds itself to the documentation standard DSC applies across the WHO CONTROLS WHOM? initiative. Every factual claim carries its category, its source, and its date. The classification vocabulary is explicit: DOCUMENTED FACT for records verified against primary or official sources; DOCUMENTED SOURCE RECORD for secondary estimates with named methodology; ATTRIBUTED STATEMENT for positions attributed to a named speaker; INVESTIGATIVE QUESTION for an inquiry stated openly as unresolved; and PENDING RESEARCH for a comparison or figure the record does not yet support — which is marked rather than estimated, always.

The financial-records discipline inherited from TECHNO-WELFARE ROYALTY applies throughout: procurement records are not disbursements, disbursements are not profits, valuations are market estimates with named methodologies, and contract ceilings are not obligations. Where this investigation draws on the same public records examined elsewhere — the USAspending award records, the Federal Register's published documents, the FEC's political-finance filings — it links to the documented treatment rather than duplicating it.

The demographic-analysis discipline is stated in Section 05 and repeated here because it governs the most sensitive part of this investigation: measures are named, datasets are disclosed with their self-reported and definitional limitations, intersectional categories are handled with statistical honesty, and Indigenous data sovereignty is treated as a research requirement. No demographic figure enters this page without a named, dated source.

SOURCE REGISTRY