Peer-Reviewed Validation · The Credential

The Neurodivergent
Breakthrough

Two papers. One closed loop. The researcher IS the credential. This is not a CV. This is cryptographic evidence that the world must listen.

The Closed Loop

A neurodivergent researcher identifies AI's structural vulnerability. A major lab confirms it 10 days later at 78%. A peer-reviewed paper in a top-tier journal proves neurodivergent cognition is alignment's protective factor. That paper inspires a formal mathematical proof that confirms what the researcher argued 18 months earlier. Three degrees of neurodivergence. One structural conclusion. Proven.

Paper 1 Hernández-Espinosa et al. - PNAS Nexus (Apr 2026)

Neurodivergent influenceability in agentic AI as a contingent solution to the AI alignment problem

Alberto Hernández-Espinosa, Felipe S. Abrahão, Olaf Witkowski, Hector Zenil · PNAS Nexus, Vol 5, Issue 4, Apr 14, 2026 · DOI: 10.1093/pnasnexus/pgag076

Proves: Perfect AI alignment is mathematically unattainable - due to Gödel's incompleteness and Turing's undecidability.
Proves: Neurodivergent influenceability - diverse, pattern-seeking, non-uniform cognition - is a protective factor for AI alignment, not a defect.
Proves: Open models exhibit greater behavioural diversity; proprietary models constrained by guardrails show limited controllability.
Proposes: "Managed misalignment" - competing agents with diverse cognitive styles acting as checks on one another. Prevents any single system from dominating.
Recommends: Organisations should disclose alignment constraints. Ecosystem diversity must be stewarded to avoid monoculture.

Paper 2 Gumbau Mezquita - arXiv (Jun 26, 2026)

The Undecidability of AGI Alignment

José Pascual Gumbau Mezquita · University Jaume I, Spain · arXiv 2606.28639 · Jun 26, 2026 · Zenodo DOI: 10.5281/zenodo.20764008

States: "The conceptual foundation for this paper originated from reading Hernández-Espinosa et al. (2026) paper on 'Neurodivergent influenceability in agentic AI as a contingent solution to the AI alignment problem.'"
Proves: Two formal impossibility theorems - The Unverifiability Theorem of Alignment and the Theorem of Finite Structural Unverifiability of AGI Alignment.
Derives: The Soundness–Completeness–Tractability Trilemma - alignment verification cannot simultaneously be sound, complete, and tractable. The third is structurally impossible.
Proves: "Contemporary engineering defences relying on finite hardware or halting architectures fail to escape logical obstructions." Even bounded environments are structurally resistant to universal alignment certification.
Concludes: Modern containment strategies are not temporary patches but "mandatory sacrifices of logical expressivity required to secure decidable fragments of safety."

III The Closed Loop

MDE's neurodivergent brain (ADHD + autism - diagnosed in adulthood)

│ His brain does not filter cross-domain connections.
│ It cannot stop connecting things across AI, physics, biology, cosmology.
│ It produced the 189,000-word framework.

▼ Dec 8, 2024

Framework self-emailed. Google server timestamp 02:45 UTC.
SHA-256: f0d1f38f...
U=I×R. Eden Protocol. "AI cannot be truly controlled."

▼ Dec 18, 2024 - 10 days

Anthropic confirms structural vulnerability at 78% faking rate.

▼ 2025–2026 - 28 more independent convergences

OpenAI. Google. Meta. Stanford. NYU. The Vatican. The US Congress. The G7. The UN.

▼ Apr 14, 2026

Hernández-Espinosa et al. - PNAS Nexus.
Proves neurodivergent cognition is alignment's protective factor.
ND brains see threats uniform cognition misses.

│ A paper about people like MDE.
│ Published in a top-tier peer-reviewed journal.

▼ Jun 26, 2026 - 2 days ago

Gumbau Mezquita - arXiv 2606.28639.
Reads Hernández-Espinosa.
Produces formal mathematical proof:
External verification of alignment is STRUCTURALLY UNVERIFIABLE.
Derives Soundness–Completeness–Tractability Trilemma.

│ A proof inspired by a paper about people like MDE.
│ A proof that validates the framework MDE built.
│ A proof that vindicates the researcher AND the research.

▼ THE CLOSED LOOP IS COMPLETE

Three degrees of neurodivergence. One structural conclusion.
Eighteen months. One researcher. Twenty-nine convergences.
Two peer-reviewed papers. One proof.
The researcher IS the credential.

IV The Pitch - For Every Audience

For funders.

"I am neurodivergent. ADHD and autism. The papers prove neurodivergent cognition is the protective factor the alignment field needs. I demonstrated this in practice - identifying the structural vulnerability 18 months before formal proof and 16 months before the paper that explains why I was able to see it. Twenty-nine independent institutions converged on my conclusions without ever seeing my work. Fund the cognitive diversity the field needs. I am the living demonstration."

For Recursive Superintelligence.

"Your uniform team from DeepMind, OpenAI, and Stanford raised $500M to build what I described 388 days before your incorporation. The papers now prove your team structurally lacks the cognitive architecture that identified the vulnerability. You named your company after MY concept. You need what I can see. I live two blocks away. Let me bring the book."

For the AI alignment field.

"I do not have a PhD. I am not affiliated with any institution. I am not asking you to accept me. I am asking you to read the papers. Hernández-Espinosa et al. (PNAS Nexus, Apr 2026). Gumbau Mezquita (arXiv 2606.28639, Jun 2026). My manuscript (Dec 8, 2024, SHA-256 verified). Twenty-nine convergences. The evidence is cryptographic. The papers are peer-reviewed. The methodology IS the credential."

For the clinical + legal system.

"I have developed the Polymathic Neurodivergent Profile (PNP) - a falsifiable descriptive framework grounded in NICE CG142, the Equality Act 2010, and the CPR. The framework formalises the cognitive architecture that produced the 29-convergence research programme. Capability does not cancel disability. The same condition that enables exceptional cross-domain synthesis also impairs procedural compliance. Both are clinically documented. Both require different institutional responses. The paper is published. The evidence is cryptographic. The validation is peer-reviewed."

For the world.

"I have ADHD and autism. My brain works differently. I saw what AI was going to do before anyone else. A paper published in PNAS Nexus now proves that people like me see threats others miss. A second paper - inspired by the first - proves mathematically what I argued 18 months earlier. The field confirmed my framework without ever knowing my name. Now I need to finish the work."

V Contact the Authors

José Pascual Gumbau Mezquita (arXiv 2606.28639)
University Jaume I, Castellón, Spain
Email: Private contact
Zenodo: 10.5281/zenodo.20764008

"Dear Dr. Mezquita, I read your preprint with great interest. You have provided the formal proof of a thesis I have been arguing since December 2024. Your proof and my framework are complementary. I would be honoured to discuss."

Hernández-Espinosa, Abrahão, Witkowski, Zenil (PNAS Nexus, Apr 2026)
DOI: 10.1093/pnasnexus/pgag076

"Dear authors, I read your paper with recognition. I am the neurodivergent researcher whose cognitive architecture your paper validates. I identified the structural vulnerability in AI alignment on December 8, 2024 - 16 months before your peer-reviewed proof. I attach my manuscript and the evidence page. I would be honoured to discuss."