From the Campus Operations edition of September 15, 2026
Georgia Tech drafts admin/employment AI policy with registers, human review, and suspension powers
Georgia Institute of Technology · Higher education administration and employment · United States (Georgia)
- Publisher
- Georgia Tech Policies
- Original publication
- Policy Steering Committee Approval June 2026 (Effective Date TBD)
- Source retrieved
- 2026-09-15
What happened
June 2026-approved (effective date TBD) Georgia Tech policy for administrative and employment AI creates an Institute AI Governance Program and Committee, routes tools to Institute or Local AI Registers after risk-based review, restricts Protected Data and PII to verified tools, prohibits automating employment, grievance, or disciplinary decisions without appropriate human review, and authorizes suspension of tools causing severe operational, privacy, or security consequences, with Internal Audit compliance authority.
Why it matters
Control blueprint for campus-operations AI: registers, employment-action limits, privacy guardrails, and audit/suspension powers complementary to labor demands for human authority.
Evidence and measured results
Primary institutional policy PDF. Steering Committee approval June 2026; Effective Date still TBD, so operational enforceability may not yet be active.
Limitations and uncertainty
Effective date TBD; outcomes of compliance unknown; USG-specific framing; retrieved PDF includes a DRAFT marker in a summary header area.
Put this evidence to work
Lighthouse Advisory interpretation, grounded in this source. Enriched 2026-09-15; this does not change the original publication date. Each role follows the same checklist so takeaways compare across sources.
Sales
Role takeaway
- Customer problem
- Administrative AI spreads faster than inventories, employment safeguards, and auditability.
- Stakeholders
- EVP Administration/Finance
- AI Governance Officer
- CISO
- HR
- Internal Audit
- Procurement
- Discovery questions
- Is there an AI register today?
- Who can suspend a tool?
- Are employment decisions human-reviewed?
- How is Protected Data gated?
- Value hypothesis
- Register-based approval with employment human-review rules reduces unmanaged admin AI risk; Georgia Tech documents the control pattern pending effective date.
- Potential engagement
- Policy-to-operating-procedure implementation for AI registers and employment-use controls.
- Unsupported claims
- Policy already fully in force
- Registers alone prevent shadow AI
- Guaranteed bias-free employment tools
Pre-sales engineering
Role takeaway
- Fit
- Enterprise admin and employment AI governance controls for a research university.
- Architecture and integration
- Central and local AI registers; risk-based approval; integration with cybersecurity assessments and procurement intake.
- Prerequisites
- Governance committee charter
- Data classification
- Tool inventory intake form
- Suspension runbook
- Constraints
- Effective Date TBD
- USG policy stack dependencies
- Local register consistency risk
- Security
- Verified controls before Protected Data use; CISO suspension for severe incidents; prompt-injection awareness.
- Proof of value
- Percent of admin AI tools on register
- Time-to-approval metrics
- Tabletop of suspension authority
Delivery
Role takeaway
- Work
- Publish procedures, migrate known tools onto registers, train managers on employment-use bans without human review, enable audit sampling.
- Dependencies
- OIT/Cybersecurity
- HR
- Procurement
- Internal Audit
- Ownership
- AI Governance Program under EVP Administration and Finance.
- Skills and adoption
- Mandatory compliance training and self-service literacy resources as policy promotes.
- Governance checkpoints
- Register listing before production use
- High-risk automation go/no-go
- Periodic Internal Audit review
- Acceptance criteria
- No unregistered production admin AI in sampled units
- Documented human review for employment-impacting AI
- Tested suspension path
- Risks
- Policy lag because effective date is TBD
- Local register fragmentation
- Shadow tools outside intake
Implementation considerations
Lighthouse Advisory interpretation across the operating dimensions a public-sector buyer must settle before this evidence becomes a design. Each note answers the question under its heading for this specific source.
Architecture and integration
What must connect, and where does the AI sit in the workflow?
Inventory-first architecture—no tool use until register approval; local limited-use registers for constrained tools; intelligent automations may need lifecycle logs and go/no-go approvals.
Governance
Who approves, reviews and stays accountable for outcomes?
Cross-functional committee plus Internal Audit creates enforceable governance beyond guidance memos; employment automation without human review is explicitly out of bounds.
Security and privacy
What data, permissions and controls need testing?
Protected Data and PII only in tools with verified security controls; CISO suspension authority for severe consequences; prompt-injection and hallucination safeguards listed as principles.
Accessibility and workforce
Who is affected, and what skills or accommodations follow?
Accessibility is a listed ethical principle; employment, grievance, and discipline automation limits protect workers from opaque AI decisions. Training expectations create workload for staff.
Procurement
What should contracts, pricing and exit terms secure?
AI tools must pass register approval; procurement intake should block purchases that skip governance review.
Operating model
Which teams own the service once it runs?
Central governance program with local registers balances enablement and control for administrative AI.
Publication history
- 2026-09-15Campus Operations · Issue 107 resources
Stable resource ID: gatech-admin-ai-employment-policy-202606