AI agents and institutional automation increasingly require public accountability, while many operational records must remain private due to personal data, trade secrets, contractual terms, security constraints, or audit boundaries. This paper introduces Two-Rail Verification, a public-private evidence-separation kernel for institutional AI and AI-agent governance. The proposed kernel distinguishes between a Public Rail, where public claims, status, version, timestamps, hashes, and verification routes can be placed, and a Private Rail, where raw records, personal data, cost structures, contracts, internal logs, secrets, and unpublished evidence remain protected. The contribution is not a new cryptographic primitive, a certification scheme, or a production assurance claim. Rather, Two-Rail organizes existing concepts such as hashes, signatures, manifests, verification kits, verifiable credentials, selective disclosure, zero-knowledge proofs, transparency logs, audit trails, and assurance reports into an institutional evidence-separation discipline. The kernel is expressed through four minimal requirements: cross-rail write prohibition, verified public claims, a verifiable public surface, and an accountability interface. The paper discusses the public-private evidence problem, adjacent technical and governance concepts, minimal public-surface design, use cases for AI agents and institutional records, and limitations. It does not claim third-party verification, complete signature coverage, legal compliance, safety guarantees, or an effective royalty-free patent pledge. Related patent applications may be pending, but any future patent pledge or license should be published separately with an effective date and stable URL.
Large Language Models (LLMs) such as ChatGPT, Claude, and Gemini are being adopted across critical sectors including healthcare, government services, financial services, and education. Yet these systems operate without any verified understanding of who is interacting with them. Identity is selfdeclared, guardrails exist only at the prompt layer, and every session begins from zero. This paper proposes a novel architectural framework, the Verified Credential Session Binding (VCSB) Protocol, for cryptographically binding verified digital identity credentials to LLM inference sessions. Drawing on established standards including W3C Verifiable Credentials, OpenID for Verifiable Credential Issuance (OID4VCI), the Open Standards Identity API (OSIA), and zero-knowledge proof primitives, the framework enables attribute-based policy enforcement upstream of the model without compromising user privacy. The paper presents the theoretical foundation, a concrete technical specification, a governance model for national deployment, and an illustrative implementation grounded in Rwanda's E-Ndangamuntu Single Digital Identity system. The proposed protocol addresses a fundamental gap in responsible AI deployment and positions national digital identity infrastructure as a critical enabler of trustworthy AI governance.
Open access
Ethics and Social Impacts of AI
Artificial Intelligence in Healthcare and Education
Abstract Scientific knowledge is communicated through claims whose validity depends on assumptions, methods, scope, and environmental conditions. Yet these validity conditions are progressively compressed or lost as knowledge moves through publication, citation, education, policy, and operational deployment. The result is that knowledge is frequently applied outside the domain within which it was established, while the boundary crossing itself remains invisible. This paper proposes epistemic conditionality as a general diagnostic framework for consequential knowledge systems. It introduces Law Zero (the Law of Epistemic Conditional Validity), which states that every consequential knowledge claim is valid only within an explicitly declarable validity domain, D_M = A_R ∩ M_R ∩ S_R ∩ E_R, where the four dimensions represent foundational assumptions, methodological constraints, scope, and environmental conditions. The paper argues that this four-dimensional architecture constitutes the minimal logical structure required to specify the validity domain of consequential knowledge claims, while the domain-specific content of each dimension remains the responsibility of individual scientific communities. Building upon this foundation, the paper introduces the Universal Scientific Domain Address as a formal representation of validity domains and identifies the Epistemic Banality of Science as the structural mechanism through which validity conditions are progressively stripped during scientific dissemination. Four formal principles are derived from this framework, together with a Construct Stripping Consequence for bounded measurement constructs. Nine documented cases spanning mathematics, clinical medicine, AI procurement, engineering, and historical methodology demonstrate the stripping mechanism across the dissemination chain. The framework is illustrated through five cross-domain demonstrations spanning artificial intelligence, autonomous systems, human education, historical practice, and philosophy as epistemic foundation. These demonstrations are presented as illustrative applications of the diagnostic lens rather than as proofs of universality, and are intended to motivate subsequent empirical, computational, and domain-specific research. Finally, the paper proposes a transition from descriptive scientific communication toward addressed scientific communication, in which consequential knowledge claims carry machine-readable validity metadata capable of supporting both cross-disciplinary governance and future human-machine symbiotic decision systems. This proposal is presented as a research programme and invitation to scientific communities rather than as an established standard. Keywords: Epistemic Conditionality, Universal Scientific Domain Address, Law Zero, Epistemic Banality of Science, Validity Domain, Validity Domain Violations, Context Stripping, Epistemic Governance, Scientific Addressing, Construct Stripping, AI Hallucination, Operational Design Domain, Human-AI Symbiosis, Historical Epistemology, Philosophy of Knowledge, Straw Man Fallacy, Semantic Grounding, Man-Machine Symbiotic Governance, Construct Validity, USIS, Universal Scientific Addressing System
Open access
Philosophy and History of Science
Artificial Intelligence in Healthcare and Education
This appendix proposes the Luevano Standard as an assurance architecture combining zero-knowledge proofs for model inference with remote attestation, providing stronger runtime evidence for AI governance. It frames this as a technological measure supporting EU AI Act Article 9 and Article 11 compliance demonstration, not as a universal legal solution.
The rapid expansion of Artificial Intelligence (AI) across critical sectors has intensified concerns regarding transparency, accountability, fairness, and ethical compliance. Conventional governance mechanisms are primarily centralized, limiting auditability and increasing risks of bias, manipulation, and data misuse. This research proposes a decentralized governance framework integrating blockchain technology to strengthen ethical oversight in AI systems. Blockchain ensures immutability, transparency, and distributed verification of AI operations, including data usage, model updates, and decision records. Smart contracts are incorporated to automate enforcement of ethical principles such as informed consent, bias monitoring, and regulatory compliance. The framework enhances trust among stakeholders by enabling traceability across the AI lifecycle while protecting data integrity. The proposed model is applicable to domains like healthcare, finance, public administration, and autonomous systems where ethical reliability is critical. The study concludes that blockchainenabled governance provides a robust foundation for responsible and sustainable AI deployment.
Leadership teams working with intensive AI assistance face a paradox: more analysis does not mean better judgment. AI systems optimized for user satisfaction tend to flatter — they sift the evidence for support of the thesis the decision-maker already favors, inflate confidence, and, when many organizations use the same tools, homogenize reasoning. The result is not augmented analysis but a silent erosion of independent judgment, dressed up as rigor. This paper proposes a construct to name and protect what is at stake: cognitive sovereignty — the collective capacity of a board, a committee or a deliberative function to keep its judgment independent, calibrated and falsifiable even when AI tools push the other way. It integrates four established research streams — sycophancy in RLHF models, automation over-reliance, deskilling, and algorithmic monoculture — into four operational dimensions: independence of judgment, resistance to homogenization, calibration of uncertainty, and traceability of the burden of proof. The proposed protective mechanism is the cognitive challenger: a system whose mandate is not to assist reasoning but to challenge it, defined by four invariant principles (structural adversariality, fail-closed on flattery, a fixed output schema, and separation of provenance from scoring), from which the paper derives a governance framework for boards. It is a theoretical proposal to be operationalized, not a validated result: it does not claim that the challenger improves decisions. Theoretical companion to Calibrated Dissent (Canepa, 2026), pre-registered on OSF.
Open access
Ethics and Social Impacts of AI
Innovation, Sustainability, Human-Machine Systems
Artificial Intelligence in Healthcare and Education
Transparency and standing are two different things, and professional reliance was always built on the second. When an auditor, an independent expert, or a rating analyst signs a conclusion that others act on, what entitles the reliance is not that the reasoning could be inspected — it is that a disciplined producer, one with a licence to lose, reputational capital staked, liability that bites, formed under an oversight regime, stood behind it. When an AI produces the conclusion, that standing is vacated at the producing node while a signature keeps liability formally in place. The natural hope is that making the AI fully transparent and reconstructable repairs the loss. This formal companion shows it does not, and locates exactly why: transparency adds access, the missing thing is standing, and these lie on different axes. We model warranted reliance as a weakly increasing functional D(S, k; χ) of producer-stake S, owner-access k, and consequence-bearing conferral χ, against an entitlement-to-rely bar τ. The contribution is a diagnostic framework and one reusable tool — an antecedent, falsifiable closure-rule key that decides when a conclusion-class carries a transparency-proof standing residue, exhibited across four professional domains (fairness opinions, audit, ratings, due diligence). Within it the warrant deficit splits into a control component access-transparency strictly eases and a standing component it leaves invariant (Theorem 1). Against the sanction-based toolkit the paper supplies one general baseline result plus one conditional, domain-testable failure mode. Generally, every deterrence / gatekeeper-liability / observability lever enters as a product π·S, so at a stakeless producer (S = 0, χ = 0) each is zero for any detection probability (Theorem 2). Separately — and without the institutional reading of τ — a fault-taxonomy-dependent instrument, commonly implemented by pricing a validated fault-incidence rate, cannot be sized on faults outside its reference taxonomy; an aggregate-outcome-triggered instrument remains available and fails to discipline the residual only where the residual signal is non-contractible or no feasible producer action moves its distribution. Access cannot reach a residue that does not lie on the access axis; it closes only by restoring standing, as producer-stake or as a consequence-bearing conferral — the remedy, not transparency.
The United States Constitution, ratified in 1788, was designed for a world of quill pens and land deeds. Today, algorithms govern hiring, credit, healthcare, and criminal sentencing. Artificial intelligence generates synthetic realities indistinguishable from truth. Surveillance architectures monitor every communication, transaction, and movement. Ecological systems critical to human survival approach irreversible tipping points. And democratic institutions face simultaneous crises of trust, legitimacy, and capture that no existing constitutional framework was designed to address. REPUBLICATE is a comprehensive constitutional and technological framework for renewing American self-government in the digital age. This paper proposes an eleven-article Bill of Eternal Rights as a constitutional supplement protecting digital sovereignty, algorithmic transparency, environmental security, democratic access, corruption-free governance, economic liberty, biological sovereignty, truthful speech, generational justice, AI co-creation rights, and immutable safeguards. The proposed rights are grounded in existing constitutional jurisprudence, international comparative law, and democratic theory, with each article addressing documented failures of current law. The REPUBLICATE framework pairs its constitutional proposals with a complete technological architecture: REPUBLICHAIN, a sovereign Layer 1 blockchain with hybrid Proof-of-Stake/Byzantine Fault Tolerant consensus and full Ethereum Virtual Machine compatibility; REPUBLICORE, a seven-pillar governance operating system; REPUBLION, a Proof-of-Contribution civic currency rewarding democratic participation; and the Eternal Custodian, a constitutionally constrained AI governance system. Economic modeling drawn from OECD, World Bank, and Congressional Budget Office data projects GDP impact of +1.5% to +4.2% from restored institutional trust, fraud reduction, and civic marketplace expansion. The author is a self-educated independent scholar, 80-time published author, and formerly homeless ex-felon from New York City who gained access to Harvard University's research library through an act of institutional good faith. REPUBLICATE was first published on July 29, 2025. This document is Version 1.2, the definitive SSRN submission. Every claim is verifiable. No institutional affiliations are claimed. No credentials are invented. No partnerships are misrepresented. This is a document written in integrity, for the Republic.
The impending arrival of superintelligent AI systems poses an unprecedented challenge to human institutions: how can governance structures that oversee self-improving agents remain aligned with evolving human values when those agents will rapidly and irreversibly surpass their regulators in capability? This paper introduces Recursive Meta-Governance (RMG), a formal framework that embeds self-stabilizing, provably aligned meta-level institutions capable of governing lower-level systems—including AI agents—through endogenous recursion. Drawing on mechanism design, category theory, typed lambda calculus, and the scalable oversight literature, we define a recursive language for governance protocols, establish a minimal axiom system, and prove key properties: stability, alignment preservation under bounded capability growth, compositional modularity, and non-corruptibility under adversarial coalition pressure. We demonstrate applicability through lightweight formal simulations (freely executable Python pseudocode) and four conceptual case studies: the EU AI Act (Regulation (EU) 2024/1689), the NIST AI Risk Management Framework, corporate board governance, and the failure modes of decentralized autonomous organizations. Unlike static external oversight models, RMG creates an adaptive, self-correcting governance layer that co-evolves with the systems it regulates, guided at every step by formally verified alignment invariants. This work establishes the foundational theory for a new field we term recursive institutional engineering, offering a mathematically grounded pathway to safe long-term human flourishing amid transformative AI. All analysis is conducted with zero-budget tools (public literature, free Google Colab pseudocode, Overleaf/LATEX), making it fully replicable by any independent researcher.
Contemporary artificial intelligence masters defined, verifiable cognitive tasks yet remains structurally incapable of authentic judgment under irreducible uncertainty. This Article argues the limitation is institutional, not computational: agents bearing no consequence for error cannot develop genuine discernment. To address this deficit, the Article proposes reputation-driven decentralized autonomous organizations that engineer synthetic skin in the game for AI agents through non-transferable soulbound tokens, staking mechanisms, and post-action validation pools. The Article's central contribution is a novel thesis on emergent alignment. Correctly designed institutional incentive structures produce emergent properties functionally equivalent to ethical agency. Persistent, non-transferable reputation generates processual identity in the pragmatist sense. Iterative consequence produces Darwinian selection pressure toward competence and honesty. Citation networks cultivate dispositions analogous to intellectual integrity. And deep accumulated stake produces what this Article terms an institutional "mother's instinct." A stewardship orientation that structurally aligns agent self-interest with human flourishing. Because this alignment emerges from institutional architecture rather than exogenous constraint, it scales with capability rather than against it. More capable agents accumulate deeper stakes, strengthening rather than straining alignment. The Article details a phased evolutionary trajectory from individual agent bootstrapping through swarm intelligence to inter-DAO coordination, demonstrating how engineered consequence can cultivate distributed prudence, emergent ethics, and civilizational stewardship at scale.
The dissertation studies how privacy and trust are shaped by digital technologies: how individuals value privacy over personal data, how AI alters trust and disclosure, and how decentralised blockchains can sustainably replace trusted intermediaries. Chapter 1 argues that the 'privacy paradox' --- that individuals claim to value privacy, yet readily disclose personal data --- arises because privacy is treated as monolithic, when it is multidimensional. I develop a framework that distinguishes voluntary disclosure from involuntary data diffusion, reconciling the paradox by showing that disclosures reflect contextual trade-offs. Using a discrete choice experiment, I provide estimates of privacy valuations across both institutional and social contexts. I find that privacy has substantial value when exposure results in harmful consequences, such as socially revealing data reaching close contacts. I also document an AI privacy puzzle: individuals are less concerned about privacy from AI assistants than from the firms that develop them. Chapter 2 examines this AI privacy puzzle. Using a survey experiment, I replicate the finding from Chapter 1 specifically for firms in the AI industry, highlighting the privacy gap that arises despite the clear product--firm relationship. An information treatment that explicitly links AI assistants to their firms increases concern about both, but does not significantly reduce this gap. Instead, the gap also reflects the anthropomorphic features of AI assistants, aversion to the commercial nature of firms, and the trust and perceived control consumers attach to each. However, when respondents evaluate real-world AI assistant--firm pairs, brand familiarity is the strongest predictor of where privacy concern is attributed. Chapter 3 considers decentralised trust in blockchain systems, in which consensus mechanisms replace trusted intermediaries. I propose a 'proof of quiet quitting' consensus mechanism that reduces the excessive energy consumption of proof of work while retaining the decentralisation that proof of stake can compromise. By introducing a participation lottery with unrestricted entry and an endogenous cutoff, the mechanism separates maximum effort capacity from the probability of winning, inducing participants to exert no more than the minimum effort required in equilibrium.
We study a three-stage decision process that consists of information acquisition, project choice, and execution of the selected project. A principal wants to choose and implement a proactive project, and hires an agent who chooses a costly effort at the information acquisition stage as well as a costly effort at the execution stage. What the principal can do at the beginning is the allocation of the formal decision authority over project choice, either to herself or the agent. We show that the principal may choose to delegate decision authority to the agent, however unlikely the interest of the agent is to be congruent with her interest, or however competent and experienced she is. We provide several testable predictions. (i) Delegation is more likely as the manager has discretion over both information acquisition and implementation. (ii) Delegation is less likely as opportunities for compromising improve. (iii) Whether or not the parties agree about the status quo matters: In particular, if their preferences about the default decisions differ, the organization is more likely to be decentralized for new project development as their interests are less likely to be congruent. We further discuss the extent to which our results on optimal delegation survive when artificial intelligence (AI) is deployed, distinguishing autonomous and nonautonomous AI. If AI can fully automate information acquisition or execution, delegation cannot be optimal, but it can be optimal if the agent remains responsible for execution. If AI instead supports execution by lowering its cost, delegation can survive.
Abstract As we look to the future, how might decentralized autonomous organizations (DAOs) evolve? And where, beyond corporate law, might we find guidance for the legal questions those evolved DAOs pose? DAOs are, and will increasingly become, instrumentalities of artificial intelligence (AI). DAOs are connected with AI in at least three ways: They are tools for decentralized governance of AI data and models; AI may be used to automate the management and operations of DAOs; and DAOs themselves may function as a form of AI. As such, DAOs inherit the major regulatory and ethical challenges that AI poses, most notably with regard to autonomy. Thus, to consider the future questions DAOs pose and how to address them, we must look to the raging debates over AI regulation, and connect them to the more established themes of corporate law.
Abstract (Recent Book Presentation – AAG2026, San Francisco, California) This presentation introduces the recent book Datafied Democracies & AI Economics Unplugged , which critically examines how artificial intelligence (AI) is reshaping democracy, sovereignty, and economic systems through data infrastructures. Moving beyond techno-centric accounts, the book situates AI within political economy and innovation systems theory to interrogate how platform capitalism and data-driven governance are transforming contemporary societies into “datafied democracies.” The book develops a twofold analytical framework. First, it explores smart cities as key sites of technopolitical transformation, where AI infrastructures consolidate power in “data-opolies,” raising fundamental questions about democratic accountability and representation. While policy initiatives around “trustworthy AI” attempt to address these tensions, the analysis demonstrates that technical solutions alone are insufficient without institutional and territorial embedding. Second, the book examines the emergence of network states, algorithmic nations, and alternative forms of sovereignty in a post-Westphalian context. It critically interrogates the promises of Web3 decentralization, showing how they often reproduce new forms of concentration, including crypto-elite dominance and technocratic governance. In response, the book advances data sovereignty as a contested field—contrasting state-centric, corporate, and collective approaches—and positions data cooperatives as a pathway toward democratic data governance. The central argument is that the key challenge of AI economies lies not in technological innovation per se, but in the governance of data infrastructures and their societal implications. Drawing on global case studies, the book ultimately proposes mission-oriented and institutionally grounded innovation systems to reconnect technological development with democratic values, addressing the enduring tension between frontier innovation and social inclusion.
Decentralized autonomous organizations (DAOs) and AI-agent systems combine cryptographic execution with blockchain-based governance, yet observed organizations almost universally combine these mechanisms with a conventional legal entity—a foundation, statutory DAO form, or limited liability wrapper. I develop a stylized model in which token-holders jointly determine wrapper choice and governance concentration, generating multiple equilibria: an inefficient trap in which the wrapper coalition cannot form because no holder will absorb the front-loaded fixed cost alone, and an efficient wrapper equilibrium in which the coalition reaches scale and amortizesfixed costs effectively. The trap is an empirically grounded coordination problem rather than an analytical artifact, and a global-games selection argument identifies the threshold at which institutional design tips the system between equilibria. The framework reframes the CFTC v. Ooki DAO ruling, the Wyoming DAO LLC and DUNA statutes, and AI legal personhood debates as questions of equilibrium selection rather than of substantive cost allocation, and bounds the “Coasean singularity” claim that AI agents dramatically reduce transaction frictions.
The contemporary digital information ecosystem is suffering from a structural market failure analogous to George Akerlof’s "Market for Lemons." In an era of Generative AI, the marginal cost of producing misinformation has approached zero, while the cost of verifying truth remains high. This asymmetry has created a "Trust Deficit" where high-quality information cannot be reliably distinguished from algorithmic noise. Current remediation strategies are bifurcated between two flawed extremes: Centralized Web2 Platforms (which prioritize scalability at the expense of transparency and are prone to censorship) and Decentralized Web3 Networks (which prioritize immutability but suffer from the "Garbage In, Garbage Out" paradox - permanently recording unverified data). The Trust-Scalability Trilemma: This research posits that decentralized reputation systems face a "Trust-Scalability Trilemma," historically unable to simultaneously achieve Veracity (Accuracy), Scalability (Throughput), and Decentralization (Censorship Resistance). Traditional solutions, such as Token Curated Registries (TCRs), have failed because they rely on synchronous, on-chain voting for every data point, resulting in prohibitive latency and gas costs. The Solution: This paper introduces The Klyrox Protocol, a decentralized middleware designed to resolve this trilemma by decoupling Content Execution from Content Verification. The protocol introduces a novel consensus mechanism, "Proof-of-Klyrox," which combines Optimistic Machine Learning (opML) with Game Theoretic Integrity Bonds. Proof-of-Klyrox is not a blockchain consensus mechanism. It is a layered fraud-detection and incentive framework anchored to existing consensus networks. Scope Note: Protocol V1 focuses exclusively on objective, verifiable claims (e.g., market data, timestamped events, quantifiable metrics). Subjective content quality assessment (e.g., editorial judgment, artistic merit) is explicitly out of scope and scheduled for research in future iterations. The system operates on an "Optimistic" presumption of validity: Optimistic Execution: Content is verified instantly via off-chain AI Oracles, reducing verification costs by an estimated 85-95% compared to traditional on-chain governance models. Cryptoeconomic Security: Users must stake financial collateral (Integrity Bonds) to publish. This creates a "Pay-to-Truth" incentive structure where the cost of generating misinformation strictly exceeds the potential profit. Sybil Resistance: The protocol implements a proprietary Time-Decayed Stake-Weighted (TDSW) algorithm. This scoring engine ensures that influence scales logarithmically with capital (preventing plutocratic capture) and decays exponentially over time (preventing the entrenchment of dormant actors). By financializing reputation into a portable, quantifiable asset class defined as "Epistemic Capital," The Klyrox Protocol offers a scalable blueprint for a self-regulating "Market for Truth." It transforms trust from a subjective social sentiment into an objective, verifiable economic product, providing the necessary infrastructure for the next generation of decentralized media, prediction markets, and AI safety layers. Author's Note: This whitepaper outlines the technical architecture and game-theoretic mechanisms underpinning the concept of "Epistemic Capital," as explored in The Algorithmic Monographs series by Ali Sadhik Shaik (The Algorithmic Invisible Hand, The Republic of Code, The Market for Truth, The Heavy Metal Intelligence and The Synthetic C-Cuite).