The Moura Fundamental Principle of Intelligence (FPI) as a Paradigm Shift in the Search for Extraterrestrial Intelligence

This work introduces the Moura Fundamental Principle of Intelligence (FPI) as a structural and falsifiable framework for the analysis and detection of intelligence-compatible signatures across different domains. Rather than defining intelligence ontologically, the FPI establishes a minimal structural condition: systems exhibiting intelligence-compatible behaviour must involve the recursive interaction of construction (𝒞) and decomposition (𝒟) across one or more levels of abstraction. This formulation is substrate-independent and does not rely on behavioural or anthropocentric criteria. The framework is operationalised through three structural axes: recursion depth (𝓡d), abstraction levels (𝓐ₗ), and decompositional integrity (𝓘D), combined into a logarithmic structural indicator adjusted by an observability factor (𝓞ₒbs). A necessary threshold condition (𝓡d ≥ 1 and 𝓐ₗ ≥ 1) is proposed as a falsifiable criterion for filtering candidate signals. The work further introduces a structural detection pipeline and applies the framework to the Search for Extraterrestrial Intelligence (SETI), proposing a shift from technology-specific detection (e. g. , radio signals) to the identification of structural invariants in observed data. The approach is complementary to traditional SETI methodologies and expands the search space to include signals and artefacts that may not produce conventional electromagnetic signatures. A synthetic demonstration illustrates the behaviour of the proposed metrics under controlled conditions, contrasting stochastic and recursively structured signals. Preliminary empirical tests on large language models suggest structural asymmetry (𝒞 ≫ 𝒟), supporting the diagnostic potential of the framework. , although broader validation remains an open research direction. The framework also explicitly addresses its limitations, including measurement challenges, observability constraints, and the inability to detect intelligence-compatible signatures in structurally impoverished data. This work is presented as a preprint and constitutes an initial formalisation of the FPI, intended to support further empirical testing, refinement, and cross-domain application.