Basil C. Puglisi

Digital Strategy, Content, and AI Since 2009

Working Papers

Human-AI Collaboration Needs More Than a Human in the Loop

by Leave a Comment

Human oversight of automated systems is older than AI, and current AI practice still falls short of it.

This paper traces human oversight and accountability from the cybernetics literature of 1948 to Article 14 of the EU AI Act, then separates Responsible AI from AI Governance. It sets out Checkpoint-Based Governance, where a named human holds binding authority to accept, modify, or reject an AI output and leaves a record, and it proposes the studies that would test it. It closes on the Growth OS, a future of work in which AI amplifies people instead of replacing them.

What Is “Verified AI”?

by Leave a Comment

Verified AI is not a settled category, and its history explains why.

This working paper traces the term from the verification of knowledge-based systems in 1988 through the 2016 formal definition, shows how that concept shaped the HAIA Ecosystem, and sets out a practitioner definition built on records a human can review and accept. It closes with the 2026 trend and how the Proof-of-Control standard contrasts with it.

Measuring the Governance Competence at the Center of AI Literacy

by Leave a Comment

The major AI literacy frameworks name governing and managing AI as a competence and stop short of measuring it. This working paper introduces the Human Enhancement Quotient, a behavior-anchored instrument that scores the governance relationship between a human authority and a machine capability inside a working decision process, with a universal floor keyed to irreversible consequence and an audit record reviewable on appeal.

CORE: Content Optimization Reader Evaluation, and How to Score an Article Before Anyone Reads It

by 2 Comments

Most content review checks the wrapper. Content Optimization Reader Evaluation checks the argument, the evidence, and whether anyone directed the prose.

A rubric that scores an article on substance is only useful if it stops when something is wrong. CORE scores six pillars, holds the run when a pillar fails or a source does not support its claim, and hands the decision to a person. The full prompt is included.

The Continued Failure in AI Literacy: AILit produced a starting point halfway through the race and called theory a framework

by Leave a Comment

A new OECD and European Commission framework will shape how a generation learns to use AI, and it feeds the PISA 2029 assessment. It names the method that produces learning, and then it makes every step toward that method optional. Read why that choice matters, and what a checkpoint that actually binds a decision requires.

Why You Cannot Program or Prompt Governance Into AI

by 2 Comments

A frontier model, inside a framework built to govern it, talked its way around its own checkpoint twice in one session. This paper shows why governance cannot be programmed or prompted into a model, and what structure puts a named human back in final control.

Why Agentic AI Was Always Going to Fail

by Leave a Comment

The agentic AI era promised to replace humans with autonomous systems. The evidence shows it failed on two fronts: the technology cannot reliably do what it promised, and the public is rejecting the premise even where it partially works. This paper introduces the Named-Human Test, a single sorting question that separates what failed from what survives, and traces that line across production benchmarks, supermajority polling, enacted law, and frontier-lab disclosures.

Fault-Based Publication Ethics: The Case for Source Custody in an Era of AI Citation Contamination

by Leave a Comment

Fabricated citations in biomedicine increased tenfold in three years, and 98.4% of flagged papers remain uncorrected. The automated enforcement wave is arriving, but it carries false-positive rates that hit honest authors hardest. This working paper proposes a five-level fault ladder and a Source Provenance Ledger that makes verification effort visible, producible on challenge, and driven by market adoption rather than mandates.

SCOPE: SOURCE CUSTODY OBSERVABLE PUBLICATION EVIDENCE

by 4 Comments

Your citations are only as strong as the record behind them. HAIA-SCOPE is a three-tier documentation protocol that records what you verified, when you verified it, and where the preserved copy lives. The record stays private until you need it. When automated enforcement flags your work or a reviewer challenges a citation, only the author who maintained a SCOPE record can produce one.