Each published addition stays linked to its sources. A separate assessment explains what it changes, what remains uncertain and what to investigate next.
The programme asks whether reality could be a simulation created by AI. The work assessed here does not establish that it is. A useful contribution can refine an assumption, challenge a calculation or propose a better test without supplying new evidence of simulation.
The foundation remains version 1.3. Assessments below document developments separately; “not yet assessed” means no effect has been recorded.
Clarifies why finite resources alone do not predict a visible glitch. Under the stated assumptions, slower external execution leaves internal records unchanged. A testable version of the theory therefore needs an explicit resource policy that changes observations. This is a conditional argument, not evidence that an AI created our universe.
Adds a resource-cost question, but the numerical range is not established by the listed source. The first AI review cycle disagreed about the arithmetic. For all-pairs spatial work over successive time steps, the time count is multiplied once; squaring all spacetime updates describes a different computation. The estimates remain under review and do not strengthen the simulation claim.
Proposes an invariance catalogue: list what changes under a specified implementation and what stays the same. This gives the programme a clearer test-design method. No catalogue experiment has been completed, and the proposal supplies no new empirical evidence of a simulation or an AI creator.