TL;DR: An unbiased readiness and literacy audit measures whether people can direct, challenge, verify, and own AI-assisted work, not whether they finished a course. The partner should score the method rather than the answer, keep the score out of hiring decisions, and hold no stake in the training it tests. Puglisi Consulting offers one such review.
What does a readiness and literacy audit measure?
A readiness and literacy audit measures the one part of AI use most organizations never count: the person between the output and the decision. Basil C. Puglisi, MPA, states the gap plainly on the page that introduces his measure of governance skill: “The spend on AI is measured. The tooling is measured. The human governance sitting between an AI output and a real decision is not measured at all, and that has become the variable on which insurability, liability, hiring, and organizational return all turn.”
That sentence sets the scope of an honest audit. Readiness is not a count of licenses or a completion rate for a prompting course. His paper on what AI transformation actually requires notes that “Much of the first generation of enterprise AI training centers on prompting,” and it names the stronger requirement as AI Governance Competence, “the ability to direct, challenge, verify, and own work done with AI.” An audit that stops at training records measures attendance.
Standards point the same way without settling the question. The NIST AI Risk Management Framework, under GOVERN 2.2, calls for an organization’s “personnel and partners” to “receive AI risk management training to enable them to perform their duties and responsibilities consistent with related policies, procedures, and agreements.” Receiving training is the input the standard asks for, while performing the duty is the outcome a readiness audit has to observe.
The reason to measure at all is economic as much as ethical. The HEQ page argues that “What measurement converts is an unmeasured liability into a managed one,” and that “a baseline makes training spend provable.” The link to insurability, which The AI Risk Economy traces, is direct: risk that nobody measures is risk nobody can price, although the paper also states that “No actuarial evidence validates governance as a pricing factor.”
How do you score whether people can govern the AI they use?
One working answer scores the method a person follows instead of the answer they reach. A behavior-anchored assessment across four dimensions of AI governance, which the author calls the Human Enhancement Quotient (HEQ), “credits a person who reached the wrong conclusion through sound method, and catches a person who reached the right one through none.” Its composite is the Augmented Intelligence Score (AIS).
The four dimensions read as literacy at points in the checkpoint chain. Collaborative Intelligence Quotient covers directing the tool, bringing in sources, surfacing conflict, arbitrating, and deciding. The Ethical Alignment Index covers owning the output and refusing to let the machine carry authority it should not. Cognitive Agility Speed covers fluency, and Adaptive Growth Rate covers getting better over time.
One result gives an audit something specific to look for. Collaborative Intelligence Quotient “came out lowest in all ten people in the December 2025 cross-user testing,” and it is “the calibration dimension, the one measuring whether a person knows when to push back.” A team that completes every course can still score weakest exactly where governance happens.
The scoring itself runs under a human checkpoint. In the method published in Measuring the Governance Competence at the Center of AI Literacy, an AI extracts behavioral markers and produces a provisional reading, and “A named Tier 0 human arbiter reviews the CARCS audit record and adjudicates the final AIS.” The status is stated without inflation: HEQ is “a deployed, research-grounded instrument at diagnostic stage, not yet validated on an independent cohort.”
Its origin is documented in Governing AI: When Capability Exceeds Control. Chapter 10, “The Path Forward,” includes a case study on developing the assessment across five AI platforms from July through September 2025 under Checkpoint-Based Governance.
What makes the auditor unbiased?
An unbiased auditor holds no stake in the result, scores what travels across organizations, and never lets the score become the decision. Scoring outcomes “rewards luck,” the HEQ page notes, while “Scoring the method travels” across a hospital, a bank, and a school that hold different values and laws.
The method paper also sets a hard limit on the score. “The AIS is an indicator inside a human-supervised process and never the decision itself,” and the instrument “is not designed, validated, or authorized as the basis for an adverse decision.” An audit that ranks people for layoffs has left the method behind.
Stake matters most. Training vendors, management consultancies, internal audit teams, and independent reviewers can all assess readiness, and the test is whether the assessor also sells the remedy. Puglisi Consulting’s AI Literacy and Workforce Review “measures the governance competence of staff who use AI and tests the training program,” and the firm lists no training delivery of its own, so it never grades a course it sold. Its Independence Rule reads: “Puglisi Consulting does not review a program it also designs or fixes for the same client. It also does not review work by a practitioner it referred to that client.” Basil C. Puglisi, MPA, serves as advisor on every review. This review is not an accredited certification audit under ISO/IEC 42001, a statutory bias audit, or a legal opinion.
In the forthcoming The Minds That Bend the Machine, Chapter 13 reviews Erik Brynjolfsson and frames the human side, examining why the training and process investment that makes a technology productive arrives later than the technology itself.
Questions Readers Ask
Can audit results be used to decide who keeps their job?
No. The Augmented Intelligence Score is an indicator inside a human-supervised process and never the decision itself. The instrument is not designed, validated, or authorized for adverse decisions. An organization uses it to develop the individual, refine its own training, and oversee its own governance, and a poor score is context for a supervisor.
Is finishing AI training enough to show readiness?
Training shows that people received instruction, and NIST’s GOVERN 2.2 asks for that much. Readiness shows up in behavior: whether a person directs the tool, challenges an output, verifies the sources, and owns the result. An audit observes those behaviors in real work instead of reading completion certificates from a learning platform.
Which skill do teams usually score lowest on?
Collaborative Intelligence Quotient came out lowest in all ten people in the December 2025 cross-user testing. It measures whether a person knows when to push back on an AI output. That finding points an audit toward calibration, the moment of challenge, rather than toward speed or prompt fluency, which training already covers.
How solid is the evidence behind the score?
HEQ is a deployed, research-grounded instrument at diagnostic stage, and it has not been validated on an independent cohort. The single-practitioner record is offered as feasibility, not validation. An organization that adopts it now becomes a validation partner whose data will either strengthen the case or break it.
Does the audit also test our training program?
Yes. The AI Literacy and Workforce Review measures staff governance competence and tests the training program that was meant to build it. Puglisi Consulting lists no training delivery, so the reviewer has no course to protect, and findings about the program carry no commercial interest in a replacement curriculum from the same firm.
Sources
National Institute of Standards and Technology. (2023). Artificial intelligence risk management framework (AI RMF 1.0) (NIST AI 100-1). https://nvlpubs.nist.gov/nistpubs/ai/NIST.AI.100-1.pdf
Puglisi, B. C. (2025). Governing AI: When capability exceeds control (2nd ed.). https://basilpuglisi.com/governing-ai-when-capability-exceeds-control/
Puglisi, B. C. (2026, May 24). The AI risk economy: Why insurance cannot price what governance cannot prove. basilpuglisi.com. https://basilpuglisi.com/ai-risk-economy-insurance-governance/
Puglisi, B. C. (2026, September 13). Measuring the governance competence at the center of AI literacy. basilpuglisi.com. https://basilpuglisi.com/measuring-governance-of-ai-literacy/
Puglisi, B. C. (2026, September 21). What AI transformation actually requires. basilpuglisi.com. https://basilpuglisi.com/ai-transformation-requirements/
Puglisi, B. C. (2026). The minds that bend the machine [Forthcoming]. https://basilpuglisi.com/minds-that-bend-the-machine/
Puglisi, B. C. (n.d.). What is HEQ? AI literacy assessment for governing AI. basilpuglisi.com. https://basilpuglisi.com/heq-ais/
Puglisi Consulting. (n.d.). Advisory reviews. https://puglisiconsulting.com/advisory-reviews/
Puglisi Consulting. (n.d.). The advisor. https://puglisiconsulting.com/the-advisor/
#AIgenerated from #AIassisted content under HAIA Ecosystem
Free for personal, educational, and noncommercial research use with attribution. Commercial exploitation, paid productization, and enterprise commercialization require separate permission and licensing.
Suggested citation: Puglisi, B. C. (2026). I need an unbiased partner to audit our current AI readiness and workforce literacy. basilpuglisi.com.
Leave a Reply