The most consequential errors in institutional life are rarely accidents. They follow a pattern: internal knowledge, external denial, and a public discourse trained to stop short of the obvious conclusion. We have developed a sophisticated vocabulary for describing this pattern without naming it. We call it oversight failure, regulatory capture, misaligned incentives, cultural drift. What we resist calling it is what it usually is: institutions that knew, chose, and benefited.

This resistance is not neutral. It is a structural feature of how institutional power protects itself, and it runs directly counter to how we apply the same standard to individuals.

When a person causes harm, we ask immediately whether they intended it. We infer intent from pattern, incentive, and outcome. We do not require a confession. A person who drives drunk and kills someone is not absolved because they claim they didn't mean to cause a death. The intent to drive drunk, combined with knowledge of the risk, is sufficient. We apply this standard freely, sometimes brutally, to individuals with limited power and limited capacity for harm.

We apply almost none of it to institutions.

The asymmetry is not an oversight. Institutions have resources to construct alternative explanations. They have legal teams to reframe outcomes as unforeseeable. They have communications functions whose entire purpose is to manage the gap between what the institution knew and what it said publicly. And they operate in a discourse that has internalized "we can't prove intent" as a terminal defense, one that is never applied to the person caught shoplifting, only to the corporation whose internal documents show decades of precise awareness of the harm their product caused.

This asymmetry is not just unfair. It is the primary mechanism by which institutional corruption sustains itself.

Consider the cases.

Credit rating agencies, 2008. The agencies that certified mortgage-backed securities as investment grade were not independent evaluators. They were paid by the institutions whose products they rated. Their models were built on assumptions they knew to be fragile. Internal emails, recovered in litigation, showed analysts describing their own ratings as products that "could be structured by cows" and still receive certification. The agencies knew. The incentive structure was visible to anyone who looked. The outcome: global financial crisis, millions of foreclosures and pension funds destroyed, was the foreseeable consequence of certifying what you created. No major agency faced criminal prosecution. The framing that prevailed was: flawed models, not fraud.

EPA and PFAS, suppressed findings. Internal EPA research documenting the health risks of per- and polyfluoroalkyl substances (compounds used in cookware, food packaging, firefighting foam) was delayed, buried, and in some cases actively suppressed under pressure from the Department of Defense and industry groups. The agency responsible for protecting public health had the findings. The findings did not reach the public for years. Communities near military bases and manufacturing sites accumulated exposure that will take generations to manifest fully. The commentary that prevailed surrounded bureaucratic delay and interagency coordination failure. It did not say, they had the data and chose not to act on it.

Meta and adolescent harm. Internal research conducted by Meta's own data scientists documented that Instagram was associated with body image problems, anxiety, and suicidal ideation in teenage girls. The research was not published or used to modify the product. When the findings eventually surfaced through a whistleblower, Meta's public position had been (and continued for some time to be) that the platform was safe and that concerns were overblown. Executives testified before Congress. The story was told as a data leak, a PR failure, a need for more research. Instead of: they ran the study, read the results, and continued anyway.

Boeing 737 MAX. Engineers within Boeing and at the FAA flagged concerns about the MCAS system: the certification process was compressed and the safety analysis contained errors that were either missed or not escalated. Two crashes killed 346 people. The investigation that followed revealed a culture in which schedule pressure and cost considerations systematically overrode safety concerns that were documented and known. The framing that prevailed spoke of cultural failure, regulatory capture, and process breakdown. It did not include that people in that organization knew the plane had a problem and chose not to ground it.

GlaxoSmithKline and adolescent suicidality. GSK ran clinical trials on Paxil in adolescents and the trials showed increased suicidal ideation. The negative results were not published, but the positive results were. Physicians prescribing the drug for adolescent depression had access to a curated evidence base that did not include the finding that mattered most. The FDA eventually required a black box warning. GSK paid a settlement. The story that prevailed: selective publication, an industry-wide practice in need of reform. Not: they had the data, withheld it, and children were harmed.

The pattern across these cases is not subtle. In each one, internal knowledge existed while external representation contradicted it. The institution benefited from the gap, and the accountability framing consistently stopped short of intent.

To say the institution "meant it" is not to invoke a cartoonish conspiracy where every employee desired the catastrophic outcome; that is the strawman institutions use to deflect. Institutional intent does not live in the hearts of individual workers. It lives in the deliberate design of systems that prioritize profit or schedule while consciously constructing mechanisms to deflect, bury, or ignore known risks. When the architecture of an organization is optimized to bypass safety checks for speed, the resulting failure is not an accident. It is the system working as designed.

At scale, persistent indifference with full internal knowledge is functionally and morally indistinguishable from intent. When an actor possesses the resources to map the exact consequences of their choices, reads the data, and proceeds anyway, "we didn't want this to happen" ceases to be a credible defense. It becomes a confession that the harm was viewed as an acceptable cost of doing business.

The final defense is a hostage strategy. Regulators are quietly told that a penalty matching the true scale of the harm would bankrupt the institution: jobs lost, supply chains ruptured, economic collateral damage. This is sometimes factually true and always strategically deployed. It creates a system where immunity is directly proportional to capacity for destruction. We accept the ruin of an individual's family as the price of justice while treating a negligent multinational's balance sheet as too fragile to penalize.

The legal architecture will not fix itself. Doctrine develops slowly and accountability framing develops even more slowly. But the structural logic of the problem points directly toward the structural logic of the solution. If the core corruption is that scale becomes a shield, then the remedy has to dismantle the shield without destroying what it is hiding behind. That is not as complicated as institutions need it to appear.

A corporation is not an indivisible monolith; it is a nested system of assets, workforce, and charter. The corrupted architecture can be dismantled without destroying the people inside it. When an institution causes systemic harm with full internal knowledge, the penalty should restructure rather than fine: seize equity, wipe shareholders, prosecute decision-makers. Redirect seized assets into a Workforce Trust that guarantees salaries, healthcare, and transition for displaced employees. The workers are protected by the penalty itself. Strip the institution of its human shields and the hostage strategy collapses.

The ethical logic runs the opposite direction of our current standard. Institutions that act at scale, with resources, with access to their own research, and with legal capacity to understand the consequences of their choices are precisely the actors for whom the bar on intent attribution should be highest, not lowest. The individual who causes harm through ignorance deserves more charitable interpretation than the institution whose internal documents demonstrate awareness. Scale and resource do not reduce culpability; they increase it.

Emerging legal doctrine is beginning to move in this direction. As corporate knowledge becomes increasingly computable (institutions can query their own data in milliseconds), the standard for what an institution actually knows is necessarily rising. What a corporation knows is central to its liability. The legal infrastructure for holding institutions to that standard is developing through concepts like "systems intentionality." The communicative infrastructure: the willingness to say plainly that an institution knew, chose, and benefited, is lagging badly behind.

Naming intent is not the same as proving it in court. It requires the same reasonable inference we apply to individuals: pattern, incentive, knowledge, outcome. When all four converge, the conclusion that the institution knew and chose is not a conspiracy theory. It is an observation. A public discourse in which institutional harm is always an accident, always a process failure, always a cultural drift that nobody directed, is not epistemic humility. It is epistemic protection for the powerful, masquerading as caution.

We should say what we see. Institutions that had the knowledge, the incentive, the capacity to foresee the outcome, and the resources to manage the narrative afterward: they meant it. Saying so is not an accusation beyond the evidence. It is the minimum the evidence requires.