EA Shareholder Diversification from Anthropic

Quick Assessment

Overview

The effective altruism ecosystem faces an extraordinary concentration risk: an estimated $27-76B in risk-adjusted EA-aligned capital is tied to a single private company at the $380B Series G valuation — scaling to $42-119B at secondary market pricing (≈$595B, March 2026) — while total annual EA funding is approximately $120-150M/year. EA Forum This may represent the largest concentration of expected future philanthropic capital in a single illiquid asset in history, and it is almost entirely inaccessible today.

At AnthropicOrganizationAnthropicComprehensive reference page on Anthropic covering financials ($380B valuation, $14B ARR at Series G growing to $19B by March 2026), safety research (Constitutional AI, mechanistic interpretability...Quality: 74/100's $380B valuation (Series G, February 2026), the EA-aligned capital breaks down as follows: Anthropic

The secondary market column uses a 1.57x multiplier (= $595B / $380B). Secondary market prices are volatile and reflect derivatives/prediction market sentiment rather than institutional primary round pricing. Nevertheless, if these prices approximate the eventual IPO valuation, the total EA-aligned capital could significantly exceed the Series G-based estimates.

No prudent financial strategy would hold 95%+ of expected future capital in a single illiquid position. Yet this is effectively the EA ecosystem's situation with respect to its largest pending funding source.

The Concentration Problem

Why This Is Dangerous

Single-point-of-failure risk. If Anthropic's valuation drops 80%—comparable to historical tech corrections like WeWork, Snap's 2022 decline, or the 2000 dot-com bust—$20-60B in expected EA capital evaporates. The remaining $5-15B would still be significant but represents a fundamentally different philanthropic future than the $27-76B base case.

Liquidity risk. All this capital is illiquid until an IPO, which prediction markets place at a median of June 2027. Manifold Markets If the IPO is delayed to 2029-2030, capital remains locked for 3-4 additional years during a critical window for AI governance.

Correlation risk. AI safety funding is concentrated in a company whose value correlates with AI industry conditions. The scenarios where funding is most needed (rapid, uncontrolled AI progress) may coincide with either very high Anthropic valuations (good for capital) or competitive displacement (bad for capital), creating unpredictable dynamics.

Value drift risk. Two distinct dynamics erode EA alignment over time. First, compositional shift: as Anthropic (Funder)AnalysisAnthropic (Funder)Comprehensive model of EA-aligned philanthropic capital at Anthropic. At $380B valuation (Series G, Feb 2026, $30B raised): $27-76B risk-adjusted EA capital expected. Total funding raised exceeds $...Quality: 65/100 documents, the fraction of employees with "EA-ish perspectives" is decreasing among newer hires, diluting the EA share of total employee equity. Second, individual drift: delayed liquidity means delayed giving, and the longer donors wait, the more likely their cause priorities shift away from current EA priorities. EA Forum

Historical Precedents for Concentration Risk

Pre-IPO Diversification Strategies

1. Private Secondary Market Sales

EA-aligned shareholders can sell Anthropic shares to buyers on private secondary markets. This is the most direct path to diversification before IPO.

Secondary market data (February–March 2026):

Secondary market pricing in March 2026 shows extreme dispersion — from $186/share (Premier Alternatives, 28% discount to Series G) to $596/share (Ventuals, 57% premium). The wide range reflects differences between traditional OTC brokers (often showing stale or discounted pricing) and derivatives/prediction markets (which can reflect forward-looking sentiment). The Hiive indicative price of $474 has zero live orders, suggesting it may be an algorithmic estimate rather than a transaction price. Nevertheless, the overall trend is upward: the highest-liquidity platforms (Ventuals, UpMarket) show substantial premiums over the Series G, implying that the market consensus may already be pricing in continued revenue growth toward the $20-26B 2026 guidance.

Key precedent: Dustin MoskovitzPersonDustin MoskovitzDustin Moskovitz and Cari Tuna have given $4B+ since 2011, with ~$336M (12% of total) directed to AI safety through Coefficient Giving (formerly Open Philanthropy), making them the largest individu...Quality: 49/100 moved a $500 million Anthropic stake into a nonprofit vehicle to reinvest returns. Fortune This demonstrates that large EA-aligned holders can and do execute private transactions at scale.

Constraints:

• Right of First Refusal (ROFR): Most private company shares require company approval for transfers
• Volume limitations: Secondary markets have limited depth; blocks above $1B are difficult to place without material price impact
• Price dispersion: March 2026 secondary prices range from 28% discount (Premier Alternatives) to 57% premium (Ventuals). Actual transaction prices likely fall in the middle. The employee tender offer at $350B pre-money suggests Anthropic itself prices internal liquidity below the Series G.
• Tax: Capital gains triggered on sale

Realistic capacity: $500M-$3B could move through traditional secondary markets pre-IPO. Additionally, the new $5-6B employee tender offer provides a direct, company-sanctioned liquidity channel — the combined capacity is significantly higher than previously estimated.

2. Expanded Company Buyback Programs

Anthropic has already conducted one employee buyback. Advocating for expanded programs is a high-leverage intervention.

March 2025 buyback: Maginative

• Price: $56.09/share ($61.5B valuation)
• Eligible: Employees with 2+ years tenure
• Limit: Up to 20% of equity, capped at $2M per person

February 2026 employee tender offer: Anthropic launched a much larger employee share sale program of up to $5-6 billion at a $350B pre-money valuation — notably below the $380B post-money Series G. Bloomberg This is a significant expansion of employee liquidity and represents a major step toward the recommendations below. At $5-6B, this is 2,500-3,000x larger than the March 2025 buyback cap.

Recommendations for further expansion:

• Allow transfers to 501(c)(3) organizations as an alternative to cash buyback
• Offer quarterly tender windows rather than ad hoc
• Consider pricing at or above Series G ($380B) rather than below, given secondary market premiums

3. Pre-IPO DAF Transfers

Shareholders can transfer private shares directly to Donor-Advised Funds, which can sell when liquidity becomes available.

Advantages:

• Immediate tax deduction at fair market value
• Avoids capital gains tax on appreciated shares (saving ~37% in California) EA Forum
• Legally binding commitment to charitable giving
• Some DAF providers (Fidelity Charitable, Schwab Charitable) accept private company stock

Scale already achieved: An estimated $20-40B is already in DAFs through Anthropic's employee matching program—historically 3:1 at 50% of equity, now reduced to 1:1 at 25% for new hires. See Anthropic Pre-IPO DAF TransfersAnalysisAnthropic Pre-IPO DAF TransfersAnalyzes Anthropic charitable giving mechanisms from both the equity holder's and philanthropic community's perspective. The employee matching program (3:1 at 50% historically, 1:1 at 25% currently...Quality: 58/100 for detailed analysis, including time-sensitive employee tax considerations (ISOs, AMT, 83(b) elections, QSBS eligibility) that create urgency around the transfer timing.

Limitation: DAF donors retain full advisory control over which charities receive grants. Capital in a DAF is legally committed to some 501(c)(3), but not necessarily to EA causes.

4. Private Placement to EA-Aligned Institutional Buyers

Large EA-aligned foundations or purpose-built investment vehicles could purchase Anthropic shares from existing holders, keeping the capital within the EA ecosystem while providing liquidity.

Potential buyers:

• Good Ventures / Coefficient GivingOrganizationCoefficient GivingCoefficient Giving (formerly Open Philanthropy) has directed $4B+ in grants since 2014, including $336M to AI safety (~60% of external funding). The organization spent ~$50M on AI safety in 2024, w...Quality: 55/100 (Moskovitz's vehicle)
• Purpose-built EA investment vehicles
• Longview PhilanthropyOrganizationLongview PhilanthropyLongview Philanthropy is a philanthropic advisory organization founded in 2018 that has directed $140M+ to longtermist causes ($89M+ to AI risk), primarily through UHNW donor advising and managed f...Quality: 45/100 or similar advisors facilitating transfers

Constraints: ROFR provisions, accredited investor requirements, and concentration risk for the buying institution.

Post-IPO Diversification Strategies

5. Systematic 10b5-1 Selling Plans

Pre-arranged selling plans that execute automatically, compliant with insider trading regulations. Available after the lock-up period (typically 6-12 months post-IPO).

Best practice: Sell 5-10% of holdings per quarter, diversifying into a broad market portfolio. This provides certainty of execution regardless of market conditions or psychological biases.

6. Direct Charitable Stock Transfers

Transfer large blocks of public stock directly to EA organizations or DAFs. This is the primary mechanism for deploying the bulk of EA-aligned capital.

Tax advantage: Avoids capital gains tax on appreciation (potentially saving 20-37% depending on jurisdiction and holding period).

7. Charitable Remainder Trusts (CRTs)

Transfer appreciated shares to a CRT that sells tax-free, invests proceeds, pays income to the donor for life, and transfers the remainder to charity.

Trade-off: Provides diversified income stream and tax benefits, but delays full capital deployment to EA causes.

Priority Framework

Why At Least 20% Should Be Diversified Pre-IPO

20% of $27-76B is approximately $5-15B. Even achieving the lower end would represent a transformative injection into the EA ecosystem—roughly 33-125x current annual AI safety funding.

Decision Framework: Should You Sell?

The right recommendation depends on who you are — exposure size, role type, and liquidity constraints. The following decision tree applies specifically to EA-aligned holders evaluating whether to sell or hold Anthropic equity.

flowchart TD
  START["Do you hold Anthropic equity?"] -->|Yes| SIZE{"Exposure size?"}
  SIZE -->|">$10M + grantmaking role"| SENIOR["Senior Grantmaker / Major Funder"]
  SIZE -->|"$1-10M, active in EA"| ANGEL["EA-Adjacent Investor / Early Employee"]
  SIZE -->|"Employee equity"| EMP{"Role type?"}
  SIZE -->|"<$500K"| JUNIOR["Junior Community Member"]
  SIZE -->|"Institutional"| INST["Foundation / Institutional Investor"]

  EMP -->|"Safety-critical research"| SAFETY["Safety Researcher Employee"]
  EMP -->|"Other roles"| TYPICAL["Typical Employee"]

  SENIOR --> SELL_FAST["SELL QUICKLY via DAF\nBias risk outweighs upside"]
  ANGEL --> SELL_MOST["SELL 70-80%\nKeep small residual position"]
  SAFETY --> SELL_EARLY["SELL EARLY\nConflict of interest on safety judgments"]
  TYPICAL --> HOLD_THEN_SELL["HOLD → DIVERSIFY POST-IPO\n18-24 month schedule after lockup"]
  JUNIOR --> DIVERSIFY["STANDARD DIVERSIFICATION\nPersonal finance, not EA strategy"]
  INST --> HEDGE["HEDGE CORRELATION\nComplex; fiduciary + signal concerns"]

  style SELL_FAST fill:#ff9999
  style SELL_MOST fill:#ffcc99
  style SELL_EARLY fill:#ff9999
  style HOLD_THEN_SELL fill:#ccffcc
  style DIVERSIFY fill:#ccccff
  style HEDGE fill:#ffffcc

Stakeholder-Specific Analysis

1. Senior Grantmakers / Major Funders (>$10M exposure)

Recommendation: Sell quickly — confidence 80%.

This is where the bias argument is most acute. Someone at a major foundation making grants about AI safety while holding $20M+ in Anthropic equity has a structural conflict that's hard to manage through willpower alone. The expected reputational and epistemic damage to the field from compromised grant decisions likely exceeds the expected financial upside from holding.

Optimal execution: Donate to DAF immediately, liquidate from within the DAF over 12-24 months to avoid both capital gains and secondary market price depression from bulk selling.

Counterweight: These actors may have informal governance influence worth preserving through shareholding. Probably not sufficient to outweigh the bias concern at this scale.

2. EA-Adjacent Investors / Early Employees ($1-10M exposure)

Recommendation: Sell 70-80%, keep a small position — confidence 70%.

This cohort has the most interesting tradeoffs. They're liquid enough to act, the financial upside from holding is real, but the conflict of interest is meaningful if they're active in community discourse.

• Liquidate enough to remove financial anxiety about the position (probably 70-80%)
• Retain a small residual position — the psychological and informational value of being a shareholder (following the company closely, having skin in the game on AI outcomes) has some positive value
• The retained position should be small enough that it doesn't influence grant or research decisions

3. Typical Anthropic Employees

Recommendation: Hold until IPO lockup expiry, then diversify aggressively over 18-24 months — confidence 65%.

The situation is different for employees:

• Liquidity is constrained — secondary sales require company approval (though the $5-6B tender offer helps)
• Their labor income is already maximally concentrated in Anthropic's success, so the marginal bias from equity on top of employment is smaller than it sounds
• Financial diversification is the dominant argument, not the bias argument — they're already epistemically compromised by employment
• Post-IPO, a disciplined 18-24 month diversification schedule is appropriate

4. Safety-Critical Research Employees

Recommendation: Sell early, even at financial cost — confidence 75%.

The exception to the general employee guidance. Employees in safety-critical research roles making judgments about Anthropic's own safety practices face a direct conflict: equity appreciation depends on capability progress, but their role requires honest assessment of whether safety is keeping pace. Faster liquidation via the tender offer or DAF transfers makes sense even at the $350B pre-money discount.

5. Junior Community Members (<$500K exposure)

Recommendation: Standard financial diversification — confidence 75%.

At this scale the bias argument is largely noise — a researcher with $200K in Anthropic options isn't going to swing major community decisions. The relevant analysis is standard personal finance: concentrated single-stock exposure in a high-variance private company is bad portfolio construction regardless of the EA framing. Exercise options when financially feasible, diversify, and don't let ideological framing overcomplicate a personal finance decision.

6. EA-Aligned Institutional Investors

Recommendation: Hedge the portfolio correlation — confidence 55%.

The most complicated case. Institutions face fiduciary constraints, signal risks (a large institutional sale affects Anthropic's secondary price), and potential information rights creating additional complexity.

The strongest argument for selling is the portfolio correlation problem: if the EA funding ecosystem is heavily correlated with Anthropic's success, a bad outcome for Anthropic (regulatory action, capability plateau, safety incident) simultaneously destroys the funding base and represents a bad AI outcome. That's a catastrophic correlation worth hedging against.

Cross-Cutting Considerations

Timing urgency varies enormously. "Sell ASAP" means different things for different positions:

• For a $50M position: 18 months of careful execution via secondary markets
• For a $2M employee position: participate in the tender offer now, or wait for IPO lockup expiry in early-mid 2027
• For a $200K junior position: exercise options when feasible, no rush

Near-term capital has outsized leverage. If there's a 2-3 year critical window for AI governance and safety infrastructure spending, capital available in 2026-2027 may genuinely have 3-5x the leverage of capital available in 2029 when AI outcomes are clearer. This is the most empirically contestable but potentially most important variable in the entire analysis.

Devil's advocate: The entire sell-now framing assumes EA/longtermist capital is well-deployed and faces diminishing returns. If the field is still severely talent- and opportunity-constrained rather than capital-constrained, the urgency of converting equity to cash drops substantially. See EA Funding Absorption CapacityConceptEA Funding Absorption CapacityThe EA ecosystem's ability to absorb large capital inflows is limited by talent pipelines, management capacity, and the challenge of maintaining quality at scale. Current AI safety funding is $120-...Quality: 3/100.

Recommended Actions by Timeline

Tier 1 — Immediate (0-6 months):

Tier 2 — Pre-IPO (6-18 months):

Tier 3 — Post-IPO (18-36 months):

Risk Scenarios

Scenario: Anthropic Valuation Drops 80%

If Anthropic falls from $380B to $76B:

Note: "Before" column shows gross figures for rows 1-3; the total applies the same risk-adjustment (cause allocation, pledge fulfillment) as the overview table. See Anthropic (Funder)AnalysisAnthropic (Funder)Comprehensive model of EA-aligned philanthropic capital at Anthropic. At $380B valuation (Series G, Feb 2026, $30B raised): $27-76B risk-adjusted EA capital expected. Total funding raised exceeds $...Quality: 65/100 for methodology.

If 20% had been diversified pre-crash, $5-15B would be preserved regardless of valuation changes.

Scenario: Successful Early Diversification (20-30%)

If $5-15B is diversified pre-IPO:

• Deployed to EA causes 2-3 years earlier than full IPO timeline
• Remaining 70-80% still participates in potential upside
• Funds available for time-sensitive AI governance interventions
• Reduced systemic risk for EA funding ecosystem

Scenario: Extended IPO Delay (to 2029-2030)

Without pre-IPO diversification:

• 3-4 additional years of total illiquidity
• Missing critical AI governance windows during 2026-2028
• Increased value drift among younger employees
• EA organizations remain resource-constrained during a pivotal period

Coordination Considerations

Who Should Coordinate

Several organizations are already positioned to facilitate EA shareholder diversification:

Risks of Coordination

• Securities law: Shareholders "acting in concert" face regulatory scrutiny under SEC rules
• Information asymmetry: Coordinated selling by insiders could raise legal concerns
• Reputational risk: Perception of "cashing out" before IPO
• Conflict with Anthropic interests: Company may prefer shareholders retain stakes for alignment

These risks argue for individual action guided by shared analysis, rather than formal coordination.

Limitations

Estimates depend on uncertain inputs. The $27-76B range relies on stake estimates, valuation assumptions, and cause allocation predictions that could all be wrong. Only 2 of 7 founders have documented strong EA connections, and 71% of founder equity may go to non-EA causes.

Secondary market capacity is limited. Moving $5-15B through private markets is orders of magnitude beyond normal secondary volume for any single company. Realistic pre-IPO diversification may be limited to $2-5B.

Tax and legal complexity. Each diversification strategy involves significant tax, legal, and regulatory considerations that vary by individual circumstances. Shareholders need qualified professional advice.

DAF cause allocation is not guaranteed. Even capital in DAFs—the most "locked in" form—can be directed to any 501(c)(3), not just EA causes.

This analysis assumes current valuation holds. If Anthropic's valuation has already peaked, the urgency calculus changes but the argument for diversification strengthens.

Key Uncertainties