1
Data Foundation — 1,200 verified skills across 12 roles
Top skills for each role were sourced from Lightcast job posting analysis, identifying the most frequently demanded skills across active Sacramento-region and national postings for each occupation. Skills were cross-referenced against O*NET occupational definitions (U.S. Department of Labor) to validate relevance and completeness.
Each skill was then scored across three dimensions by the Sacramento K-16 Collaborative research team:
· Importance (1–5) — How critical the skill is to role performance, based on O*NET importance ratings and Lightcast posting frequency
· AI Augmentation (0–3) — The degree to which AI tools currently or imminently assist in performing this skill: 0 = no AI impact, 1 = minor assistance, 2 = meaningful augmentation, 3 = significant AI augmentation
· Displacement Risk (0–3) — The likelihood that AI could reduce human involvement in this skill: 0 = no displacement risk, 1 = low risk, 2 = moderate risk, 3 = high displacement risk
Medical Assistant
100 skills · score 1.630 · ARS 0.792
Dental Assistants & Hygienists
83 skills · score 1.456 · ARS 0.766
Clinical Lab Technicians
99 skills · score 1.943 · ARS 0.730
Respiratory Therapists
100 skills · score 1.466 · ARS 0.708
Surgical Technologists
96 skills · score 1.239 · ARS 0.688
Psychiatric Technicians
100 skills · score 1.380 · ARS 0.663
Pharmacy Technicians
105 skills · score 1.978 · ARS 0.844
Physical & Occupational Therapists
100 skills · score 1.409 · ARS 0.686
Medical Records Specialists
100 skills · score 2.171 · ARS 1.000
LPN / LVN
119 skills · score 1.328 · ARS 0.726
Registered Nurses
100 skills · score 1.663 · ARS 0.726
Radiology Technicians
100 skills · score 1.991 · ARS 0.797
2
Final Augmentation Score — weighted by skill importance
Each role's augmentation potential is calculated as an importance-weighted average across all its verified skills. This ensures that high-importance skills (score 4–5) drive the result more than peripheral skills (score 1–2).
Final Augmentation Score = Σ(Importance × Augmentation) / Σ(Importance)
Scale: 0.0 (no AI impact) → 3.0 (maximum augmentation) → expressed as 0–100% by dividing by 3
Medical Records
738 / 340 = 2.171 — highest (72.4% ceiling)
Surgical Technologists
415 / 335 = 1.239 — lowest (41.3% ceiling)
3
Adoption Readiness Score (ARS) — derived from skill category mix
Not all roles adopt AI at the same rate. A role dominated by administrative and technology skills (e.g. Medical Records) is structurally more exposed to AI adoption today than one dominated by hands-on clinical or soft skills (e.g. Surgical Technologists). The ARS is calculated directly from each role's own skill category composition — no manual assignment.
ARS = Σ(Importance × CategoryMaturityFactor) / Σ(Importance)
Category maturity factors (evidence-based):
Administrative & AI Skill = 1.0 → highest AI maturity (coding, billing, scheduling)
Technology = 0.9 → EMR, PACS, workflow tools AI-ready
Clinical / Technical = 0.5 → mixed — diagnostic AI high, procedures low
Human / Soft Skills = 0.2 → communication, empathy resist automation
ARS is then normalised against the highest-scoring role (Medical Records, ARS = 0.730) so all values sit on a 0–1 relative scale. This preserves real differences between roles without artificially inflating any single role.
4
Scenario Adoption Ceilings — calibrated to healthcare AI research
The adoption factor represents how much of a role's augmentation potential is being realised in practice at a given year. Healthcare AI adoption is uniquely constrained by regulatory approval requirements, clinical liability, EHR fragmentation, and workflow integration barriers — making it substantially slower than general industry AI adoption. The three scenarios span a 3.4× range (Cautious 24% → Accelerated 82% by 2030) to ensure meaningful strategic differentiation between planning assumptions.
| Year |
Cautious |
Measured |
Accelerated |
| 2025 | 0.02 | 0.06 | 0.12 |
| 2026 | 0.04 | 0.12 | 0.22 |
| 2027 | 0.08 | 0.20 | 0.36 |
| 2028 | 0.13 | 0.30 | 0.52 |
| 2029 | 0.18 | 0.40 | 0.68 |
| 2030 | 0.24
24% of potential | 0.52
52% of potential | 0.82
82% of potential |
Scenario rationale & industry anchors:
Cautious (24% by 2030) — Reflects a slow-adoption environment driven by regulatory barriers, clinical liability constraints, EHR fragmentation, and workflow integration challenges. Healthcare AI adoption is uniquely constrained relative to other industries — the FDA had authorised approximately 950 AI/ML-enabled medical devices as of 2024, with each requiring structured regulatory review before clinical deployment (U.S. Food & Drug Administration, AI/ML-Based Software as a Medical Device Action Plan). Assumes only essential administrative AI tools achieve meaningful penetration by 2030.
Measured (52% by 2030) — Reflects steady adoption consistent with current leading health system trajectories. Anchored to McKinsey & Company's finding that generative AI could augment up to 40% of healthcare working hours when deployed at scale (McKinsey Global Institute, "A New Future of Work", 2023). Assumes documentation, scheduling, coding and clinical decision support AI tools achieve broad deployment across mid-to-large health systems by 2030.
Accelerated (82% by 2030) — Reflects an optimistic but plausible scenario under strong investment and policy conditions. The American Medical Association (AMA) reported 38% of physicians using AI tools in 2023, rising to 66% in 2024 — a near-doubling in one year suggesting rapid acceleration is already underway (AMA Digital Medicine Practice Hub, 2024). Consistent with the U.S. Office of the National Coordinator for Health IT (ONC) roadmap for AI integration in certified EHR systems by 2027–2030.
Spread rationale: The 3.4× range between Cautious (0.24) and Accelerated (0.82) reflects the wide disparity in AI readiness currently observed across U.S. health systems. The World Health Organization's guidance on AI in health (WHO, "Ethics and Governance of AI for Health", 2021) explicitly notes that adoption timelines vary substantially by care setting, resource level, and regulatory environment — supporting a wide scenario spread as more defensible than a narrow one for workforce planning purposes.
5
Final Exposure Formula
Exposure % = (Augmentation Score / 3) × AdoptionFactor(year, scenario) × ARS_normalised × 100
The division by 3 converts the augmentation score from its native 0–3 scale to 0–100%. A score of 0 = 0%, a score of 3 = 100% theoretical maximum. The adoption factor applies the time-and-scenario curve. The ARS modifier adjusts for how structurally ready a role's skill mix is for AI adoption.
Worked example — Medical Records Specialists at 2030 Accelerated:
Score = 2.171 | AdoptionFactor = 0.82 | ARS_norm = 1.000
Exposure = (2.171 / 3) × 0.82 × 1.000 × 100 = 59.4% → Orange
Worked example — Medical Records Specialists at 2030 Cautious:
Score = 2.171 | AdoptionFactor = 0.24 | ARS_norm = 1.000
Exposure = (2.171 / 3) × 0.24 × 1.000 × 100 = 17.4% → Green
Worked example — Surgical Technologists at 2030 Accelerated:
Score = 1.239 | AdoptionFactor = 0.82 | ARS_norm = 0.688
Exposure = (1.239 / 3) × 0.82 × 0.688 × 100 = 23.3% → Green