The unvalidated 51% majority we tracked in the Better.com feud last week has officially materialized, returning Vishal Garg to control. Also in today's briefing: a Pennsylvania bill funding employee ownership transitions, and a Chinese chipmaker dispute where a departed CTO's escalating claim tests whether buyback clauses hold at peak valuations.
The unvalidated shareholder majority we tracked last week has been confirmed: Vishal Garg regained control of Better Home & Finance on Monday after shareholders representing 52.02% of voting power backed his consent solicitation to remove five directors — including Chairman Harit Talwar and interim CEO Daniel Lewis. The company's Special Committee withdrew its challenge to the election inspector's preliminary determination, citing dispute costs. Garg's reconstituted five-seat board includes himself, Electronic Arts co-founder Bing Gordon, Activant Capital founder Steven Sarracino, and two unnamed Better employees. The new board is removing Better's poison pill, dismissing pending litigation against Garg, and targeting $60 million in annual cost savings with a $30 million stock buyback.
Why it matters
The resolution timeline matters as much as the outcome: Garg was removed as CEO in early August, launched a consent solicitation, and had the board reconstituted by early October — roughly two months from removal to reinstatement — with the company declining to challenge based on litigation cost rather than the merits. That speed confirms that a founder with a meaningful shareholder coalition can recover board control faster than most governance disputes resolve through litigation. What this does not resolve is the underlying strategic disagreement — AI automation versus broker partnerships — that triggered the removal. Reconstituting a board aligned with the founder addresses the governance structure but leaves the strategic bet unproven; the $60 million savings target and Tinman AI expansion are performance commitments that will test whether founder reinstatement actually creates shareholder value or simply restores a contested status quo.
Anthropic president and co-founder Daniela Amodei advised Stanford GSB students this week to test co-founder compatibility by taking a vacation together — and sharing a hotel room — before starting a company, framing it as a low-cost way to surface interpersonal friction before equity splits, board seats, and governance calcify misalignment into structural conflicts. She offered a related thought experiment: ask each founder separately to describe the company; if one draws a unicorn and the other a platypus, the partnership is in trouble. Amodei co-founded Anthropic with her brother Dario and five others who had known each other for 15+ years and worked together at OpenAI.
Why it matters
The advice is deliberately pre-legal: it addresses the human layer before the structural one, which is the correct sequencing. Most equity disputes that land in Delaware — or in the court rulings tracked across this briefing — originate in misaligned founder visions that were never surfaced before ownership percentages were assigned. Amodei's Anthropic example is instructive not because seven co-founders are typical, but because 15+ years of pre-existing relationship created conflict-resolution channels that most founding teams skip entirely. For founders negotiating early splits without VC backing, the practical implication is that no vesting schedule or buyback clause substitutes for knowing whether you and your co-founder actually agree on what you're building — and testing that before the paperwork makes disagreement expensive.
Former Groq engineers Benjamin Serebrin and Joshua Rubin filed suit in Delaware — around October 3–4 — challenging the board's December 2025 licensing arrangement with Nvidia, valued at roughly $20 billion. The complaint alleges the structure channeled approximately $3 billion in Nvidia stock to ~200 engineers (including founder Jonathan Ross and president Sunny Madra) who moved to Nvidia, while common stockholders were left with undervalued equity and a diminished independent company. The plaintiffs argue the $17 billion in licensing fees were taxed at the Groq level rather than distributed as acquisition proceeds, bypassing traditional merger-style fiduciary review. Four investors — BlackRock, Social Capital, Infinitum, and Disruptive — remained affiliated with surviving Groq despite alleged conflicts. The DOJ separately opened an antitrust inquiry in September 2026 examining whether the arrangement evaded Hart-Scott-Rodino premerger notification requirements.
Why it matters
The structural question before Delaware is whether a licensing-and-equity-grant exit can be designed to look like a commercial deal while functioning economically as an acquisition — and if so, whether courts will impose merger-level fiduciary scrutiny regardless of the label. If Delaware extends that scrutiny, it will substantially constrain how startup boards structure talent-and-IP deals with large acquirers going forward, making the insider/common-holder payout split a reviewable governance decision rather than a negotiated commercial term. The dual pressure — Delaware civil litigation plus DOJ Hart-Scott-Rodino review — means Groq's board faces compounding liability on two tracks simultaneously, which increases settlement incentive and accelerates the timeline for a precedent-setting ruling.
Liang Jun, former deputy general manager and CTO of Chinese AI chipmaker Cambricon, escalated his labor dispute claim in September 2026 from 4.29 billion yuan to approximately 27.83 billion yuan (roughly $3.9 billion), recalculating losses using Cambricon's peak stock price of 1,620 yuan per share (reached July 3, 2026) and factoring in shares held indirectly through employee platforms. Liang departed Cambricon in early 2022, triggering a buyback clause in the equity incentive plan that he disputes. Cambricon has won six prior proceedings related to the plan, with rulings confirmed on appeal in August 2026. The new labor dispute claim is pending a hearing with no court ruling yet issued.
Why it matters
Cambricon's prior court victories suggest the buyback clause itself is contractually valid — the legal question Liang is pressing is whether the clause's valuation methodology (which price, which date, which share class) was sufficiently explicit at drafting time to withstand a 6x escalation argument based on post-departure market performance. For founders writing departure buyback provisions today, the case is a concrete stress test: if a company's stock grows 10x after a co-founder or executive departs, an ambiguously worded buyback clause becomes an open invitation to relitigate on peak-price terms. Specifying valuation methodology, price-reference dates, and eligible share classes in the original grant documents is no longer a drafting nicety — the Cambricon trajectory makes it a financial exposure question.
The Pennsylvania Senate passed Senate Bill 478 on Monday on a 48-1 vote, establishing an Office of Employee Ownership within the Department of Community and Economic Development and creating a grant-and-loan program covering up to 50% of feasibility study costs — capped at $35,000 — for companies exploring conversion to ESOPs, worker cooperatives, employee ownership trusts, or direct employee ownership. The loan converts to a grant if the company does not convert, removing financial risk from exploration. The bill also mandates an Employee Ownership Advocate within DCED and directs funding toward a Pennsylvania nonprofit to run outreach programs. The bill moves to the House.
Why it matters
The near-unanimous 48-1 vote removes any ambiguity about political viability; the real test is now House passage and funding appropriation. What distinguishes SB 478 from earlier state-level efforts is the asymmetric risk design: the feasibility study cost is a loan, not a grant, until the company confirms it will not convert — meaning the state absorbs downside risk from false starts rather than requiring owners to gamble on whether a study will be useful. That mechanic directly addresses the most common barrier practitioners cite: small-business owners will not pay $30,000–$70,000 for a feasibility study on an outcome they cannot predict. The mandate for sustained institutional capacity (an Advocate, a dedicated nonprofit) suggests Pennsylvania is building an ecosystem rather than a one-cycle program, which is the condition that made states like Colorado more effective at converting ownership interest into actual transitions.
Austin-based burger chain P. Terry's announced a transition to employee ownership through an employee ownership trust affecting 1,800 workers across 38 locations. The accompanying profit-sharing program — open to employees with at least two years of service — begins distributing 5% of operating income and scales to as much as 20% over time. The founders framed the move as an extension of the company's people-centric operating philosophy rather than a founder exit vehicle. This follows our earlier coverage of P. Terry's trust announcement; the new reporting adds the graduated profit-sharing structure and the specific two-year eligibility threshold.
Why it matters
The graduated structure — 5% rising to 20% — is the operationally significant detail here. It gives the company a test period to validate that profit-sharing is financially sustainable before committing to a higher payout rate, which matters in quick-service restaurants where margins typically run 5–10% and any fixed distribution commitment can become a constraint during volume downturns. The two-year eligibility threshold is a retention mechanism that reduces the cost of high turnover by limiting which employees participate in distributions — effectively a vesting cliff applied to profit-sharing rather than equity. Combined with the ownership trust structure, P. Terry's is demonstrating a layered model: formal ownership through the trust provides long-term alignment, while the graduated profit-sharing provides near-term income signal that hourly workers can actually observe and connect to daily decisions.
A Northern Trust analysis published this week documents a recurring RSU liquidity failure: 10,000 RSUs vesting at $150 per share creates $1.5 million in taxable ordinary income due at vesting, but trading windows, blackout periods, insider-trading restrictions, and company stock-ownership mandates can prevent executives from selling shares to cover the bill. If the stock subsequently falls 30%, the executive has recognized a $1 million tax liability on a position that is now worth $300,000 less — and cannot claim that decline as an ordinary income offset. The piece recommends sell-to-cover elections and net-share withholding as structural solutions built into the grant itself.
Why it matters
The failure mode is not theoretical: it is the default outcome when RSU grants are designed without explicit liquidity planning for tax withholding. For early-stage teams granting equity without establishing sell-to-cover mechanics upfront, the Cambricon and Groq disputes covered in this edition illustrate what happens at scale — but the RSU trap can hit a single senior hire whose grant was structurally defective from day one. The design fix is not complicated: specifying withholding method (net settlement or sell-to-cover), vesting schedule alignment with trading windows, and Section 83(b) eligibility in the grant document prevents the problem entirely. The cost of omitting it is a disgruntled early team member with an unexpected six-figure tax bill and no liquid shares to cover it.
Laura Pace and David Heron launched The Curiosity Collective on Tuesday, combining B2B events agency Yellow Fish, brand strategy shop Blue Panda, and new experiential agency Untold into a multi-agency group that preserves specialist identities and eliminates competing internal P&Ls. Since Pace acquired Yellow Fish in 2021, the group has grown 525% in revenue — from £1.84 million to £11.5 million in 2025 — scaling to 42 employees and tracking toward £13 million in 2026. The model was explicitly designed to reject PE-backed holding-company structures, with senior leadership retaining direct client access and operational autonomy across agencies.
Why it matters
The 525% revenue growth figure under founder control provides a direct counter-data point to the standard agency acquisition pitch — that PE infrastructure unlocks growth impossible under independent ownership. The Curiosity Collective's growth occurred precisely by preserving what PE structures typically eliminate: specialist identity, senior-level client relationships, and the absence of competing P&L pressures. For agency founders evaluating acquisition or partnership offers, the more useful question this case surfaces is not whether to sell but whether the acquirer's structure will preserve or destroy the operational characteristics that produced the growth in the first place.
Ali Partovi's Neo Residency accelerator, which we tracked in October 3 coverage, has published additional detail this week on its SAFE structure: the $750,000 investment converts to 5% equity if the company is valued at $15 million, but only 0.75% if valued at $100 million — an inverse relationship between company success and accelerator ownership stake. Y Combinator takes 7% for $125,000; Andreessen Horowitz's Speedrun takes 10% for $500,000. Neo's model means founders who build successfully pay a smaller accelerator toll, while less successful companies pay a higher percentage of a smaller base.
Why it matters
The inverse-dilution design removes a perverse incentive embedded in fixed-percentage accelerator deals: under standard models, a founder who raises a strong Series A at high valuation delivers a larger dollar return to the accelerator without negotiating more help — the accelerator's upside scales with founder success regardless of what it contributed. Neo's model re-ties accelerator ownership to the company's early-stage value rather than its exit trajectory, which is a closer match to contribution timing. The practical question for founders evaluating accelerator programs before incorporation or first institutional raise: Neo's structure protects cap table headroom at the moment it is most valuable — before priced rounds establish the option pool and Series A investors negotiate their percentages.
A Crowley Law analysis published Monday maps the IP assignment chain required when U.S. startups hire internationally through employer-of-record (EOR) arrangements or direct contractors. The two-link chain — worker assigns to EOR under local employment law, EOR assigns to startup under the services agreement — frequently breaks because one or both agreements lack present-tense assignment language, moral rights waivers, or further-assurances clauses. Germany prohibits copyright transfer by statute, China requires inventor compensation, and France protects moral rights — all of which override U.S.-style blanket assignment clauses unless adapted by local counsel. An H-1B fee uncertainty has pushed more early-stage founders toward these international hiring structures since late 2025.
Why it matters
For a contribution-based equity framework, this matters because equity grants to international contributors are only as valuable as the company's IP ownership chain is clean. A founder who grants a 3% dynamic equity stake to a Berlin-based developer — without a German-law-compliant work-for-hire arrangement — may be granting ownership in a company whose core technology is legally ambiguous. A 70% valuation haircut at due diligence from undiscovered IP gaps, cited in a separate piece this week on IP strategy, gives that failure a concrete cost. The practical guidance: before granting equity to any cross-border contributor, confirm that the IP assignment runs all the way back to the startup in that contributor's jurisdiction, not just in the U.S.-law-governed grant agreement.
A comprehensive Mondaq analysis published Monday maps Indian venture-capital governance architecture, showing that true founder control depends less on equity percentage than on contractual governance rights that accumulate across rounds: board seats, reserved matters, anti-dilution provisions, and vesting terms can collectively shift decision-making power from founders to investors without changing cap-table percentages. The piece traces four critical mechanisms: founder vesting as a governance tool; good-leaver and bad-leaver forfeiture provisions (India is moving toward stronger clawback rights including vested-share forfeiture); ESOP pool placement on a pre-money basis (shifting down-round dilution onto founders and employees); and pay-to-play clauses that force continued investor participation or accept dilution. It also covers SEBI's September 2025 amendment permitting founder-promoters holding options granted at least one year before IPO filing to retain and exercise them.
Why it matters
The reserved-matters analysis is the piece that travels beyond India: a founder with 40% equity who has ceded veto rights over hiring, financing, and strategy to investors holds economic exposure without operational authority — a mismatch that contribution-based equity frameworks cannot fix after the fact. Cross-border founders comparing term sheets across U.S. and Indian VC deals on percentage terms alone will systematically underestimate the governance cost of Indian deal structures, which press harder on reserved-matter scope and clawback precision than U.S. counterparts typically do. The SEBI IPO amendment is a useful data point in the other direction: regulators are also actively restoring founder rights in specific contexts, confirming that governance rights are contestable through both contract negotiation and regulatory advocacy.
Writing this week, Lomit Patel argues that startup equity grants create upside participation for employees — not structural ownership or operational control. Drawing on Carta's 2024 State of Private Markets data showing compressed median employee equity at Series B while founder and investor ownership held stable, and on research indicating employees in 2024 acquisitions frequently received less than 20% of headline deal value after liquidation preferences cleared, Patel distinguishes between behavioral ownership (how employees are expected to act) and structural ownership (who actually controls exits, pivots, and capital allocation).
Why it matters
The less-than-20% figure is the load-bearing number in this piece: it means employees who accepted below-market salaries in exchange for equity grants were, in aggregate, compensating themselves with an instrument that converted to roughly a fifth of its stated value at the moment it was supposed to matter most. For founders designing contribution-based equity frameworks, this is evidence against the assumption that percentage ownership and financial outcome are proportional — they are not, because liquidation preferences sit above common stock regardless of contribution or tenure. The practical implication is that any equity framework claiming to be fair must model preference-adjusted returns, not headline percentages, and communicate that distinction to contributors before they make compensation trade-offs based on nominal ownership.
Founder Reinstatement Increasingly Runs Through Shareholder Consent, Not Governance Design The Better.com resolution — where Vishal Garg secured 52.02% of voting power to remove five directors and reclaim board control — illustrates a pattern where founders who lose operational authority recover it through shareholder coalition-building rather than pre-negotiated governance protections. The speed of resolution (the company declined to challenge based on litigation cost) and the immediate structural moves (poison pill removal, $30M buyback authorization) confirm that founder reinstatement now comes bundled with investor reconfiguration. Teams designing ownership structures before conflict should treat shareholder consent mechanics as a primary control variable, not a fallback.
Buyback Clause Interpretation Is Becoming a Nine-Figure Legal Battleground Cambricon's former CTO escalating his equity claim from 4.29 billion yuan to 27.83 billion yuan — using peak stock prices rather than departure-date valuations — signals that high-growth environments are turning departure buyback clauses into existential disputes. Cambricon has won six prior proceedings, suggesting the clause is contractually valid, but Liang's persistent escalation raises the question practitioners must answer at drafting time: were the terms clear about which price, which date, and which share class? For founders writing buyback provisions today, the Cambricon case is a concrete argument for specifying valuation methodology and price-reference dates explicitly rather than leaving them implied.
Employee Ownership Is Accumulating Legislative and Sector Infrastructure Simultaneously Pennsylvania's SB 478 — passed 48-1 and creating an Office of Employee Ownership with $35,000 feasibility study grants — lands in the same week as P. Terry's 1,800-employee profit-sharing launch and Keller Inc.'s 40-year ESOP retrospective. The three developments together show parallel tracks: policy reducing the financial barrier to exploring ownership conversion, an operator demonstrating a phased profit-sharing model (5% to 20% of operating income) accessible to thin-margin businesses, and a mature practitioner validating the long-run retention and community effects. The convergence makes 2026 a credible inflection point for small-business employee ownership adoption.
Exit Structure Innovation Is Outrunning Fiduciary-Duty Doctrine — and Courts Are Starting to Push Back The Groq lawsuit — alleging that a $20B Nvidia licensing-and-hire arrangement circumvented traditional merger review and shortchanged common stockholders in favor of ~200 insider engineers — tests whether Delaware will extend fiduciary scrutiny to deal architectures designed to look like licensing rather than acquisitions. The parallel DOJ antitrust inquiry into Hart-Scott-Rodino notification avoidance compounds the pressure. If Delaware courts apply merger-style scrutiny to licensing-and-equity-grant exits, the ruling will reshape how boards structure founder and employee payouts in any deal that blends IP transfer with talent acquisition.
IP Ownership Chains Are the Silent Dilution Event for Cross-Border Early-Stage Teams Two pieces published this week — one on international hiring compliance for U.S. startups, one on IP assignment in EOR and contractor structures — document the same structural gap: founders building cross-border teams frequently lack the two-link IP assignment chain (worker → EOR → startup) or fail to adapt U.S.-style assignment clauses to jurisdictions like Germany, China, and France that override them by statute. A 70% valuation haircut at due diligence from undiscovered IP gaps, cited in a third piece on IP strategy, gives the failure a concrete cost. The practical implication is that founders designing contribution-based equity for distributed teams must treat IP assignment as a precondition to granting equity, not an afterthought to incorporation.
What to Expect
2026-11-30—Pulley customer opt-in deadline for Carta migration ahead of December 8 platform shutdown — cap table audit window closes.
2026-12-08—Pulley platform shuts down; remaining cap table data must be migrated to Carta or an alternative platform.
2027-03-01—South Korea's Venture Investment Act amendment banning early redemption, refixing, and shareholder-inducement VC terms takes effect.
2028-07-01—Australia's proposed 30% minimum tax on discretionary trust distributions takes effect, eliminating the default founder share-holding structure.
2026-11-10—Ireland's Statutory Instrument 406/2026 takes effect, opening the Central Register of Beneficial Ownership to NGOs and journalists under 'legitimate interest' rules.
How We Built This Briefing
Every story, researched.
Every story verified across multiple sources before publication.
🔍
Scanned
Across multiple search engines and news databases
859
📖
Read in full
Every article opened, read, and evaluated
182
⭐
Published today
Ranked by importance and verified across sources
12
— The Fair Share
🎙 Listen as a podcast
Subscribe in your favorite podcast app to get each new briefing delivered automatically as audio.
Apple Podcasts
Library tab → ••• menu → Follow a Show by URL → paste