The code does not lie; only the founders do. This is a principle I have repeated countless times during my years auditing smart contracts and dissecting protocol failures. But what happens when the failure is not in the code, but in the judgment of those who claim to use evidence and rationality to maximize good in the world?
Manifund, a charitable funding platform affiliated with the Effective Altruism movement, hired Caroline Ellison. Ellison, the former CEO of Alameda Research, walked out of federal prison in late 2024 after serving time for her role in the FTX collapse. The move, announced by Manifund CEO Austin Chen, was framed internally as a strategic acquisition of talent. The external reaction was immediate, severe, and ultimately devastating to the organization.
The Centre for Effective Altruism, the community hub that functions as the movement's primary platform for coordination and fundraising, moved to ban Manifund from its events and promotional channels. The justification was blunt: Ellison had not done enough to earn back the community's trust. Chen, caught between a failing institution and a furious donor base, publicly acknowledged the decision might "blow up." Three weeks later, Manifund is fighting for survival on multiple fronts simultaneously.
The situation exposes something the crypto industry rarely discusses with honest language: the governance of off-chain institutions attached to this ecosystem is often worse than the governance of the protocols themselves. At least with smart contracts, the logic is transparent and the execution is deterministic. Human organizations make decisions in secret, reverse them publicly, and then wonder why trust evaporates.
Context: The Effective Altruism and FTX Connection
To understand the current crisis, you need to understand the relationship between Effective Altruism and the crypto industry. EA is a philosophical movement that advocates using evidence and reason to maximize positive impact. The movement gained significant traction in crypto circles during the 2021-2022 bull market, particularly after Sam Bankman-Fried positioned himself as the embodiment of EA principles: earning to give, long-term thinking, and cause prioritization.
SBF donated tens of millions to EA-aligned causes, including research institutions, individual grantees, and the infrastructure that held the community together. When FTX collapsed in November 2022, the reputational damage to the entire EA ecosystem was immediate and severe. The movement had accepted hundreds of millions from a man who was simultaneously running a fraudulent exchange and a secretly insolvent trading arm.
Caroline Ellison served as CEO of Alameda Research, the trading firm that SBF used to siphon customer funds. She pleaded guilty to federal charges, cooperated with prosecutors, and received a 24-month sentence. Gary Wang, Alameda's co-founder, received a similar cooperation credit. Both agreed to five-year trading bans as part of their plea agreements. The question of whether either should ever work in the crypto industry again, let alone in organizations that claim to prioritize doing good, has divided the community since.
Manifund operated as a funding intermediary within the EA ecosystem. The platform connected donors with researchers, projects, and causes deemed high-impact under EA criteria. In 2022, Manifund received approximately $1.5 million from FTX or associated entities. This fact alone would have made the Ellison hire controversial. Combined with the ongoing legal exposure from FTX's bankruptcy estate, the decision borders on organizational self-destruction.
The FTX bankruptcy proceedings have entered the recovery phase, where liquidators systematically pursue assets and transfers made before the collapse. Charitable donations are not immune. Under U.S. bankruptcy law, trustees can claw back transfers made during the "lookback period" if they can demonstrate constructive fraud or insolvency at the time of transfer. Effective Ventures, another EA-affiliated organization, has already agreed to return FTX donations. Manifund is fighting a $1.5 million claim while simultaneously burning bridges with its primary community infrastructure.
Core: The Governance Failure Behind the Decision
During my audit work, I have encountered projects where the single point of failure was a critical vulnerability in the smart contract code. The audit reveals the flaw, the team patches it, and life continues. What I rarely see, and what is far more dangerous, is an organization where the single point of failure is concentrated in a single decision-maker with no structural checks.
Manifund's crisis has this exact structure. Chen acknowledged that the decision to hire Ellison was made with almost no internal consultation. "It was pretty much just me," he said in an internal post that later became public. This is not humility. This is an admission of governance failure. An organization distributing charitable capital, operating in a community already reeling from association with the largest crypto fraud in history, made a unilateral decision to employ the most visible symbol of that fraud's internal operations.
The decision-making process lacked three fundamental elements that any serious institutional framework should contain: a reputational risk assessment, a legal exposure review, and a stakeholder consultation mechanism. None of these require sophisticated infrastructure. They require basic organizational discipline that a platform managing donor money should have implemented years ago.
The CEA's response crystallized what Manifund failed to understand about the community it operated within. The EA ecosystem functions as a reputation network. Membership is not determined by legal incorporation or financial standing. It is determined by community acceptance. CEA, as the primary coordinating body, serves as a de facto gatekeeper. When it decided that Manifund's hire demonstrated insufficient commitment to trustworthy operations, it acted with the swift finality of a smart contract executing a liquidity sweep.
The platform access revocation was not a legal sanction. CEA has no regulatory authority. What it has is something far more powerful in close-knit communities: the ability to make an organization's existence irrelevant. Without access to CEA's event platform, promotional channels, and donor coordination infrastructure, Manifund loses its primary mechanism for attracting funding and distributing grants. The soft power turned hard in hours.
The communication failures compounded the damage. Chen initially posted about the hire in the EA forum, framing it as a positive development that would bring experienced operational talent to the platform. When the backlash materialized, the same forum saw him reverse course entirely, acknowledging the decision might be disastrous. The whiplash told the community something important: the leadership either lacked judgment on the initial decision or lacked conviction in defending it afterward. Neither interpretation inspires confidence.
The legal dimension adds financial existential risk to the reputational catastrophe. FTX's bankruptcy estate is pursuing recovery aggressively. The $1.5 million claim against Manifund is not a rounding error for a charitable platform. Legal defense costs, potential settlement exposure, and the reputational damage that makes future fundraising nearly impossible create a convergence of threats that could force shutdown within six to twelve months. Chen's reference to "lights out" suggests he is already modeling scenarios where the organization cannot continue.
Contrarian: The Bulls Got Something Right
Here is the uncomfortable angle that the righteous critics will not acknowledge: Ellison's actual capabilities were never in question. She ran a complex trading operation, managed a large team, and navigated a business environment that most crypto executives could not handle. The skills that made her effective at Alameda are transferable to legitimate operations. The argument that she brings no value to a charitable platform is intellectually dishonest.
The EA community's absolutist stance on Ellison also reveals something uncomfortable about reputation networks: they are excellent at punishment but poor at rehabilitation. A five-year trading ban means she cannot touch securities. It does not mean she lacks the operational capacity to manage grant disbursements, coordinate with researchers, or optimize back-office functions. The community decided, based on moral grounds, that her talents were worthless regardless of context. This is ideological purity over pragmatic effectiveness.
There is a legitimate case that EA, as a movement, needs to grapple with its own complicity in the FTX story. Organizations accepted millions from SBF, promoted his version of long-termism, and built career pipelines into his ecosystem. When the collapse came, the moral accountability was distributed unevenly. Ellison went to prison. The movements that amplified and accepted SBF's generosity largely escaped structural consequences beyond reputational damage. Now those same movements are positioning themselves as arbiters of who can and cannot participate in rebuilding.
The CEA's gatekeeping power is real and probably necessary. Communities need mechanisms to exclude actors who pose reputational or ethical risks. But the absence of transparent standards, appeal processes, or written criteria transforms this power from governance into arbitrary authority. A decentralized protocol that could ban wallets without explanation would face immediate criticism. CEA operates the same mechanism without any of the transparency obligations.
Takeaway: The Structural Vulnerability of Off-Chain Governance
This episode reveals the fundamental weakness of organizations that exist at the intersection of crypto wealth and real-world causes: they inherit the governance sophistication of neither world. Manifund operated with the risk management rigor of a Discord server and the transparency standards of a private equity fund. When the crisis hit, it had no structural defenses.
The implications extend beyond Manifund. The EA ecosystem's handling of the FTX fallout, including the ongoing clawback lawsuits against charitable recipients, suggests that the financial reckoning from the collapse is far from complete. Other organizations that accepted FTX-linked funding should be modeling their legal exposure now, not waiting for the summons. The bankruptcy estate has demonstrated it will pursue recovery aggressively, and charitable status provides no immunity.
For the broader crypto industry, the lesson is about governance, not technology. The protocols we audit have upgrade mechanisms, timelocks, and multi-sig requirements because we learned that concentrated power creates catastrophic failure points. Human organizations need the same discipline. When a single person can make a decision that puts an entire institution at existential risk, that institution is not well-governed. It is a liability waiting to crystallize.
Manifund may survive. Stranger things have happened in crypto. But the structural conditions that produced this crisis remain intact: no independent oversight, no reputational risk framework, and a community that learned the wrong lessons from the FTX collapse. The Ellison hire was the symptom. The disease is the belief that good intentions substitute for institutional rigor.
They are not the same thing. The code does not lie about that either.