Louisiana's K-12 AI Playbook: Safe Policies, Smart Tools, Proven Gains

Louisiana's AI Playbook for K-12: What's Working and How to Apply It

Over half of U.S. states have released guidance for using AI in K-12. Louisiana stands out for pairing a clear framework with measurable gains. If your role touches instruction, policy, or school operations, this is a model you can adapt now.

The Framework: Clear guardrails, steady feedback loops

The Louisiana Department of Education (LDOE) launched an AI Task Force in August 2023 and published statewide AI guidance a year later. The guidance uses a simple, cyclical structure that's easy to implement:

• Purpose & Research: Define instructional goals and evidence base.
• Policy & Guidance: Set expectations for consent, data use, and classroom norms.
• Engage Stakeholders: Bring educators, students, families, and vendors into the process.
• Evaluation & Monitoring: Track usage, outcomes, and risk; iterate.

Use is classified with four levels aligned to SAMR: AI-Empowered, AI-Enhanced, AI-Assisted, and AI-Prohibited. Safety is addressed at system and classroom levels, including data storage, software approval, monitoring protocols, and required human oversight.

Infrastructure is not an afterthought: bandwidth planning, cybersecurity practices, and digital citizenship are specified. The state also maintains a curated list of approved tools with age bands, use cases, and an AI glossary. Districts and schools can build local policies from this foundation.

Louisiana Department of Education | BESE

Evidence from Louisiana Classrooms

• Zearn (K-8 math): 272,000+ active students; about 62,000 completed 90+ lessons this year. A Johns Hopkins study found students using Zearn Supplemental scored 5.97 points higher on the state math assessment than non-users. Students completing 90+ lessons outscored lower-usage peers by 4.7 points.
• Amira Learning (early literacy): 57,000+ active students; 4 million+ stories read. In a study of 79,000 K-5 students across 12 systems, K-3 Amira users scored higher on spring 2024 DIBELS, and grades 4-5 Amira users scored higher on spring 2024 LEAP ELA than non-users.
• Khanmigo (AI tutoring and teaching aide): Pilot launched in January. Nearly 50% of participating students activated accounts; 71% of teachers used it to support their work. The team delivered 22+ professional development sessions with ongoing implementation check-ins.

What you can apply this semester

• Start with purpose: Pick one outcome (e.g., fluency in K-3 reading or middle school fractions) and select a tool that directly targets it.
• Define use levels: Map AI-Assisted/Enhanced/Empowered/Prohibited by grade and course; make it visible to teachers and families.
• Tighten approvals: Require an intake form for new AI tools covering data storage, vendor security, and age appropriateness.
• Train for practice: Provide short PD on lesson workflows, prompts, and oversight. Capture model lesson plans and share them.
• Pilot, then scale: Run a 6-9 week pilot with clear entry/exit criteria. Measure usage, assessment deltas, and teacher/student feedback.
• Monitor responsibly: Set classroom protocols for AI output review and documentation of human oversight.
• Build bandwidth and security: Stress-test Wi-Fi where usage peaks; confirm filtering, logging, and MFA are in place.
• Teach digital citizenship: Address academic integrity, bias, and privacy alongside tool use.

If your team needs a focused on-ramp to AI literacy and classroom workflows, explore role-based options at Complete AI Training.

What's next in Louisiana

LDOE is continuing to evaluate programs for impact and privacy. BESE recently visited New Harmony High School in New Orleans to observe one of the first charter school classes teaching about AI. Board President Ronnie Morris put it plainly: "Every student is going to have a tutor in his laptop."

The signal is clear: define purpose, set guardrails, choose proven tools, train educators, and measure results. That's how AI lifts learning outcomes while protecting students.