The Platform
We're Not Chasing Trends.We Set the Standard.
Every design decision in LAI Workforce Readiness traces to a published, peer-reviewed source. Not to competitive benchmarking. Not to what other platforms are doing. Not to what's popular in L&D this quarter.
The reason this matters: trends in AI training move faster than most organizations can evaluate them. A methodology built on trend-chasing requires constant rebuilding. A methodology built on cognitive science doesn't — because how humans learn to think hasn't changed.
Program Architecture
Maximum Impact. Minimum Disruption.
The foundation program requires under four hours of total learner time — including the Tier 4 Practicum capstone — completed over four to six weeks at a pace that fits around existing work schedules. No classroom. No cohort scheduling. No implementation overhead.
Foundation
All three tiers. One session per tier, 30 to 45 minutes each, self-paced and asynchronous. No scheduling overhead for the institution. Each session produces a documented artifact that becomes part of the learner's permanent, portable portfolio.
Measurable critical thinking improvement begins in the first session.— Favero et al. (2024)
Quarterly Reinforcement
One abbreviated session per quarter. The learner runs a condensed 9±2 chain on a current work problem. Sentinel reads from the established behavioral baseline — the session doesn't start from zero. Skills developed through structured feedback-driven practice are maintained without daily application.
Skills persist beyond the training period with reinforcement.— Yang, Newby & Bill (2005)
Annual Recalibration
Once per year, all three tiers with updated sector-specific scenarios reflecting current AI tools and workplace demands. Produces a new artifact portfolio demonstrating current competency. Provides institutions with year-over-year capability data at the cohort level — not just completion rates.
Capability stays current as AI tools and roles evolve.
A workforce that completes the LAI Workforce Readiness foundation program demonstrates functional, transferable AI thinking skills — not tool-specific knowledge that expires when the platform changes. The three-hour time investment produces capability that compounds over time with reinforcement.
The LAI Framework™
Three Methodologies. One Destination.
Each tier is a distinct methodology grounded in published research. Each produces a concrete artifact. But the three tiers are not the finish line — they are the preparation. They build the thinking that makes the Practicum possible.
A structured diagnostic questioning chain — between seven and eleven questions, sequenced in three bands — that surfaces the root of a problem before any AI interaction begins. The learner produces a Reverse Brief: a two-sentence problem restatement that is specific, unhedged, and causal.
Research basis: Miller (1956) — cognitive load and chunking. Paul & Elder (2006) — question taxonomy. Favero et al. (2024) — Socratic AI tutor validated at p<0.001.
Reverse Brief
Two-sentence problem restatement. Specific. Unhedged. Causal.
30–45 min
Sentinel determines completion — not the clock.
A structured brief-generation process — named for the mathematical order of operations — that sequences thinking before engaging AI as a scenario-runner. The coach runs a five-to-seven question intake that assembles a Situation Brief. Bad input produces bad output at scale. PEMDAS() ensures the input is well-formed before anything executes.
Research basis: Katsara & De Witte (2019) — PBL and structured objectives. OECD (2023) — RPL and making prior knowledge visible.
Situation Brief
Knowns, problem, audience, success criteria, and what's been ruled out.
30–45 min
Sentinel determines completion — not the clock.
A generative exploration methodology — Amp (amplify) squared — where the learner ranges outward from what they know to find problems worth solving. Not defensive stress-testing, but forward-moving scenario pressure. Exploration ends when the learner can produce a Decision Brief: a refined direction, what was ruled out and why, and the next concrete action.
Research basis: Katsara & De Witte (2019) — self-directed adult learning. Paul & Elder (2006) — implications and consequences questioning.
Decision Brief
Refined direction. What was ruled out and why. Next concrete action.
30–45 min
Sentinel determines completion — not the clock.
Tier 4 · Where the Program Pays Off
The Practicum — Applied AI in Practice
The Practicum is not a methodology — it is the moment the methodology becomes irrelevant. By Tier 4, the learner has internalized the diagnostic thinking of 9±2, the structured input discipline of PEMDAS(), and the generative ranging of Amp². The Practicum puts all of it to work in a real session, with a real problem, without scaffolding. What was practiced becomes applied. What was coached becomes instinct.
Phase 1
AI Understanding
The three tiers teach how to think. Phase 1 of the Practicum teaches what you're thinking with. How AI models work functionally. How they differ in tendency and gap. How to establish the contextual relevance that makes outputs predictable. How to evaluate what comes back critically — and recognize when AI is filling gaps with assumptions you never authorized.
This is the moment the official documentation becomes readable at its intended depth. The learner isn't learning tactics anymore. They're recognizing principles they've already been living.
Living Reference Sources
Anthropic — Prompt Engineering GuideOfficial ↗Google — Gemini Prompting StrategiesOfficial ↗OpenAI — Prompt EngineeringOfficial ↗US Dept. of Labor — AI Literacy Framework (TEN 07-25)Federal ↗Links maintained by their respective organizations. Content updates automatically as models evolve — no static curriculum that goes stale.
Phase 2
Live Practicum
The learner selects a genuine work challenge and conducts a live AI session without guidance. The diagnostic depth from 9±2, the structured input discipline from PEMDAS(), and the generative ranging from Amp² are all in play — simultaneously, without prompting, because they are no longer techniques. They are how this person thinks.
Five questions emerge naturally from this session for anyone who has genuinely completed the three tiers:
- 1.What does the AI need to know to produce a predictable output?
- 2.How do you let the AI teach you what you don't yet know?
- 3.How does new understanding let you push the edge of what you thought was possible?
- 4.How does your model of the tool change as your understanding deepens?
- 5.Where does your unfair advantage emerge now that you know both the problem and the instrument?
Practitioner Record
A documented account of the live session: what the AI surfaced unexpectedly, where the learner pushed beyond it, and what advantage they now see that they couldn't see before. Sentinel evaluates reasoning quality against Paul and Elder's intellectual standards. The learner assesses their own thinking — not a chain they were guided through.
Under the Hood
Built Different. By Design.
The architecture behind LAI Workforce Readiness isn't borrowed from existing LMS platforms or content libraries. It was designed from first principles to produce a specific outcome: workers who think, not workers who comply.
Voice-First AI Coaching
Each learner selects one coach — Vivian or Jason — at the start of the program. That voice delivers the entire curriculum. Consistent. Uninterrupted. Sentinel runs silently underneath, assessing reasoning quality at every turn and instructing the coach on tone, direction, and pacing without the learner ever seeing it.
Sentinel — Silent Diagnostic Engine
Sentinel reads every learner response in real time. It assesses reasoning quality against Paul and Elder's intellectual standards, generates and validates sector-specific scenarios using a five-condition formula, determines when each band's threshold is met, and decides when a tier is complete. The learner never knows they are being assessed. They experience a conversation that somehow always knows where they are.
Artifact-Based Completion
Completion is not measured by time spent or questions answered. It is measured by artifact quality. Sentinel evaluates the Reverse Brief, Situation Brief, and Decision Brief against defined quality thresholds. A learner who produces a clean artifact in five high-quality turns advances. A learner who produces a hedged artifact in twelve surface-level turns does not. Quality of reasoning, not volume of interaction, determines progression.
Employer Dashboard — Aggregate Only
Institutions and employers receive cohort-level data: tier completion rates, band progression, where thinking skills are developing, and year-over-year capability trends. Individual learner profiles are owned by the learner. The employer portal is architecturally incapable of querying individual records — this is not a policy position, it is a technical constraint. Designed to comply with Connecticut's active AI employment legislation.
Recognition of Prior Learning
The platform does not assume every learner starts from zero. Sentinel builds a behavioral profile from the first session — not through a self-report quiz, but by reading how the learner naturally describes and frames problems. For workforce re-entry participants, prior operational competency is identified and bridged into the curriculum. Their experience is the starting point, not a deficit.
Sector-Specific Scenario Libraries
Scenarios are generated from libraries calibrated to the learner's sector — manufacturing, healthcare, logistics, legal, and general workforce. Sentinel selects and generates scenarios using a five-condition formula: domain-anchored, knowledge-gap present, interpretively open, stakes-calibrated, and progressive constraint removal. No two learners run the same session.
We Don't Follow What's Trending in AI Training.We Follow What the Research Says Works.
Psychological Review
Miller, G.A. (1956)
Cognitive load and chunking — the basis for the 9±2 questioning structure and three-band design.
Foundation for Critical Thinking
Paul, R. & Elder, L. (2006)
Two-axis question taxonomy (parts of thinking; quality of reasoning) and four directions of diagnostic pursuit — the engineering logic behind every question Sentinel generates.
Studies in the Education of Adults
Katsara, O. & De Witte, K. (2019)
Socratic questioning in adult PBL environments and the Paraskevas-Wickens learner typology — the foundation for passive behavioral profiling.
ECAI Workshop on AI in Education
Favero, L. et al. (2024)
LLM-based Socratic tutor significantly outperforms standard chatbots at critical thinking development after five turns (p<0.001) — validates the AI delivery mechanism.
American Journal of Distance Education
Yang, Newby & Bill (2005)
Socratic questioning in asynchronous learning improves critical thinking skills — and those skills persist after exposure ends. The basis for the quarterly reinforcement cadence.
Recognition of Prior Learning: A Practical Guide for Policy Makers
OECD (2023)
RPL as an instrument for making all knowledge visible and producing better matches between workers and roles — the institutional framework behind the re-entry track.
AI Literacy Framework
US Dept. of Labor — TEN 07-25 (2026)
Five foundational content areas and seven delivery principles for AI literacy — LAI Workforce Readiness aligns with all five content areas and all seven delivery principles, including experiential learning, building complementary human skills, and designing for agility.
Ready to bring this to your workforce?
LAI Workforce Readiness is available for employer licensing, institutional partnership, and workforce re-entry program integration. Government contracting eligible. SAM.gov registered.