P1
HR Operating Model
LegacyAutonomous
- Legacy
- HR runs reqs, onboarding, and cases manually with no AI in any core workflow.
- Autonomous
- AI-native HR operations self-improve, with each hire and case sharpening the next.
AIR FOR HUMAN RESOURCES
AI Is Already Deciding Who Gets Hired, Coached, and Kept
The AIR Index reads how deliberately your people function runs AI across sourcing, screening, onboarding, performance, and retention.
Bias-audit laws, the EEOC, and new state AI-in-employment rules now make every screening model a legal exposure, so people leaders must own this before it owns them.
What AIR measures
Five pillars, read for Human Resources.
The same five pillars of AI readiness, framed in the work, systems, and stakes that CHROs and people leaders actually face.
P1Operating Model
How AI is embedded in the durable HR workflows of sourcing, screening, onboarding, cases, and reviews rather than improvised by a few recruiters.
P2People Outcomes Model
How AI changes what HR delivers, from time-to-hire and quality-of-hire to retention, internal mobility, and the employee experience.
P3Talent & Capability
How broadly recruiters, HRBPs, and people-ops own AI fluency, and whether roles and reskilling keep pace with the tooling.
P4Governance & Risk
How the function controls bias, consent, candidate-data protection, and the bias-audit and disclosure laws that govern AI in employment.
P5Tool Standardization
Whether the HR AI stack across the ATS, HRIS, and assessment tools is a deliberate, owned, and defensible set rather than scattered point buys.
The AIR rating
Six tiers, Legacy to Autonomous.
AIR places Human Resources on a six-tier readiness ladder from 0 to 100, overall and for every pillar. The climb runs from digital but not intelligent, to a compounding, AI-native edge.
6
Autonomous
85–100AI-native advantage. Compounding intelligence and speed, a durable edge competitors can't copy fast.
5
Integrated
68–84Woven through the business. AI shapes the operating model, pricing, and talent, and ROI is proven.
4
Operational
51–67AI in the core, governed. Embedded at named steps with SOPs, policy, and measured gains.
3
Adopting
34–50Pockets, not a system. Real use in places, uneven and undocumented.
2
Reactive
17–33Experimenting at the edges. Scattered pilots that live in a few people's heads, ungoverned.
1
Legacy
0–16Digital, not intelligent. AI is absent or anecdotal, work is hour-priced, the stack sprawls, and no policy exists.
The point
One score tells you that you are behind; five tiers tell you exactly where to start.
The deep diagnostic
Every pillar, climbed for Human Resources.
The matrix is a 5-by-6 grid: your five pillars of AI readiness scored against the same six tiers, from Legacy to Autonomous. A single overall score tells you roughly where you sit; it hides where you are dangerously behind and where you are quietly ahead. Reading a tier per pillar turns one vague number into five specific, fixable verdicts, so you act on the truth instead of an average.
P2
Talent, Workforce & People Operations Model
LegacyAutonomous
- Legacy
- HR output is manual postings, generic onboarding, and reactive case handling with no AI value.
- Autonomous
- HR delivers AI-native workforce advantage, anticipating needs and acting before gaps appear.
P3
HR Talent & AI Capability
LegacyAutonomous
- Legacy
- HR staff have no AI fluency and roles are defined entirely around manual work.
- Autonomous
- HR is an AI-native team where capability and new roles compound talent advantage.
P4
HR Governance & Risk
LegacyAutonomous
- Legacy
- No AI hiring policy exists and any AI use is unmonitored and legally exposed.
- Autonomous
- Trustworthy-by-design AI governance is a competitive and compliance advantage for HR.
P5
HR Tool Standardization
LegacyAutonomous
- Legacy
- HR has no AI tools, relying on the bare ATS and HRIS with no AI layer.
- Autonomous
- A proprietary, defensible HR AI stack compounds advantage that rivals cannot easily copy.
Governance and compliance
Where the rules bite.
How Human Resources's regulatory reality maps onto AIR readiness. Each row is a control your governance pillar has to carry.
| Regime | Pillar | What AI readiness requires |
|---|---|---|
| NYC Local Law 144 (Automated Employment Decision Tools) | P4 | Independent bias audit within the prior year, public results, and candidate notice before using an AEDT to screen. |
| EEOC guidance on AI and the ADA / Title VII | P4 | Confirm AI screening tools do not produce adverse impact or screen out disabled applicants, and document the validation. |
| ADA (Americans with Disabilities Act) | P2 | Provide reasonable accommodation and alternative formats for AI-administered assessments, interviews, and chatbots. |
| Illinois AI Video Interview Act | P4 | Disclose AI use in video interviews, obtain consent, explain how it evaluates, and honor deletion requests. |
| Colorado AI Act (SB 24-205, high-risk employment) | P4 | Use reasonable care against algorithmic discrimination, complete impact assessments, and notify affected candidates and employees. |
| GDPR Art. 22 / candidate and employee data protection | P4 | Lawful basis and human review for automated hiring decisions, with data minimization and retention limits on applicant data. |
Illustrative mapping for AI-readiness planning, not legal or compliance advice; validate against current regulation with qualified employment counsel.
The stakes
What stalling looks like.
Bias and adverse impact in screening
A resume or video model trained on past hires can systematically screen out protected groups. The disparate impact surfaces in an EEOC charge or a Local Law 144 audit, long after thousands of candidates were rejected.
Unauditable hiring decisions
When a vendor model ranks or rejects candidates and no one can explain why, the function cannot defend a decision or satisfy a disclosure law. The black box becomes the liability.
Candidate and employee data leakage
Resumes, performance notes, and HRIS records pasted into public AI tools expose protected and personal data. This breaches GDPR and internal confidentiality, and the data cannot be recalled.
Disability accommodation failures
AI interviews, gamified assessments, and chatbots can disadvantage applicants with disabilities and screen out qualified people. Without an alternative path, the function risks ADA claims and lost talent.
Inaccurate AI in employee-facing answers
An HR chatbot that hallucinates leave, benefits, or policy guidance creates legal and morale exposure. Employees act on wrong answers and trust in the function erodes.
Start now
Signature quick wins for Human Resources.
Concrete first moves you can make before the full diagnostic, one per pillar where it matters most.
P4Days
Inventory every AEDT in the funnel
List every AI tool that scores, ranks, or screens candidates, and flag which ones trigger Local Law 144 or state disclosure rules.
P4Weeks
Commission a bias audit
Run an independent bias audit on the screening and assessment models and publish the required results before the next hiring cycle.
P1Days
Write the HR AI use policy
Define what recruiters and HRBPs may and may not paste into AI tools, with a clear ban on candidate PII in public models.
P2Weeks
Add an accommodation path
Give every AI interview and assessment a documented human alternative so disabled applicants are never screened out by the tool.
P3Days
Train recruiters on prompt and review craft
Run a short workshop so the team writes effective sourcing and JD prompts and knows when to override AI output.
P5A quarter
Consolidate the HR AI stack
Map AI spend across the ATS, HRIS, and assessment vendors, then standardize on a defensible, vetted set with named owners.
Find out where your team stands.
The old transformation is finished. The new one is scored.