Answers are grounded in this paper and cite their sources.
Post-mortem analysis identifies control system failures in reactors, suggesting improvements for future safety.
«Единственный вариант, при котором СКАЛА могла повлиять на ход событий, это скорость выполнения вычислений — если бы данные по реактору обрабатывались быстрее, то возможно последствия были бы иными.»“The only option in which SKALA could affect the course of events is the speed of the calculations — if the data on the reactor were processed faster, then the consequences would have been different.” — Habr.com retrospective on engineers SKALA, April 27, 2014Every commercial computer ever deployed in nuclear safety I&C, from SKALA in 1986 to MELTAC in 2026, belongs to the byte-based architectural class. That class has an upper bound on how fast it can complete a sense-process-respond loop, and the bound sits on the wrong side of the doubling time of a reactor in prompt criticality. Forty years of engineering improvement narrowed the gap by 250,000 times per joule but the class did not change. An information physics post-mortem of the Chernobyl SKALA system shows the reactor doubled faster than the control system could see it: a 22.6 millisecond doubling time against a control loop measured in tens of seconds. The same analysis applied to the MELTAC safety system deployed in reactors today shows the same failure in a narrower margin: a 22.6 millisecond doubling time against a 30 millisecond control loop. The Chernobyl reactor outran its control system in 1986. A reactor in 2026 still doubles faster than its control system can respond, and if it enters a prompt-criticality transient, a byte-based control system cannot track it in time.A second Chernobyl has not happened, but the absence of meltdowns shows only that reactors have not entered the prompt-criticality transient, not that their control systems could survive it if they did. The industry's defense rests on the reactor's own physics preventing that transient, not on the control system being fast enough to catch it once it begins. That defense holds inside the design basis, which the industry has steadily improved. An earthquake larger than the design basis dissolves the assumptions the design was built on, and it can introduce reactivity through fuel rearrangement, coolant voiding, and control-channel distortion on a timescale the protection systems were never required to catch, because the physics was assumed to prevent it.Whether a given reactor at a given site can be driven into a prompt-criticality transient by a beyond-design-basis earthquake is a question that has never been answered, because the analysis has never been performed. What this framework establishes is the other half: if that transient occurs, the byte-based control system that is the last line of defense is too slow to track it. The safety record covers the reactor that stays inside its envelope. It is silent on the reactor an earthquake pushes out. At Brigham City, on the Wasatch Fault, that earthquake is not a remote tail probability. It is better than one in three over the operating license, and certain over the 24,110-year product lifecycle (the Pu-239 commitment). The framework supplies the half of the analysis that can be done from physics alone. The other half, the site-specific seismic-to-reactivity analysis, is what no deployment has performed and what must be performed before any reactor is built in that basin. The alternative is to run the test in production, in an earthquake-prone basin where the reactor is already standing and hope for the best— a test the Planetary Information Security Officer is adamantly against running. Where the mathematics is unambiguous, demanding a live test is not rigor, it is choosing to cause the failure in order to measure it. The operators of Chernobyl have long carried blame for the failure of systems that were never theirs. This cybersecurity post-mortem incident report asks a question no prior investigation has asked: what role did the information system play in the failure of the safety systems, and what can be learned from it that applies to the information systems running reactors today? The link between information and energy that Landauer's principle describes is a young and underexamined area of physics, and the findings here are new. They are not presented as anti-nuclear but as areas of opportunity to improve existing technology by identifying where the information systems failed. This report identifies ten outstanding items that must be addressed through the PISO before any further nuclear deployment is advanced.
No takes yet. Share an insight, caveat, or question.
January Walker (2026) studied this question.