From the K–12 edition of October 12, 2026
CCERC October 2026: Connecticut AI pilot evaluation finds preparation/differentiation gains, math-error wariness, and demand for clear AI-use policies
Center for Connecticut Education Research Collaboration (CCERC) · State education research collaboration (CSDE-affiliated) · Connecticut, United States (six participating LEAs)
- Publisher
- CCERC AI Pilot Program Evaluation Executive Summary
- Original publication
- October 2026
- Source retrieved
- 2026-10-13
What happened
October 2026 CCERC executive summary evaluates Connecticut’s state-funded AI-in-education pilot across six LEAs (East Hartford, Lebanon, Rocky Hill, Seymour, Westport, Odyssey Community School) in grades 5–12 during 2024–25. Mixed methods: student survey n=735, teacher survey n=58, student focus groups n=32 (3 districts), teacher focus groups n=24 (4 districts). Pilot generally helped students and teachers develop AI skills. Teachers found tools most useful outside class for material preparation and differentiation; in class, teacher-created content-specific resources (e.g., rubric chatbots) were most useful. Both groups raised ethics, accuracy, and reliability concerns; math-related AI mistakes were frequent; most students were ambivalent about expanded use. Recommendations emphasize PD/peer collaboration, AI literacy, content-specific teacher-leveraged use, and clear district AI-use policies.
Why it matters
High for K–12: a current state-commissioned evaluation of multi-district AI pilots with actionable recommendations on PD, content-specific use, and policy clarity.
Evidence and measured results
Executive-summary mixed-methods evaluation funded via CCERC at CSDE request; surveys and focus groups; not an achievement RCT.
Limitations and uncertainty
Executive summary only; self-report methods; six pilot districts; no learning-outcome effect sizes.
Put this evidence to work
Lighthouse Advisory interpretation, grounded in this source. Enriched 2026-10-13; this does not change the original publication date. Each role follows the same checklist so takeaways compare across sources.
Sales
Role takeaway
- Customer problem
- States and districts piloting AI need credible evidence of what worked (and failed) for teachers and students, not vendor claims alone.
- Stakeholders
- CSDE / state education agencies
- District curriculum and instructional technology leaders
- Teachers in pilot grades/subjects
- Students and families
- State legislators funding pilots
- Discovery questions
- Which AI uses were teacher-prep vs student-facing?
- What written AI-use parameters exist, and are they consistent across schools?
- How will math/accuracy failures be monitored before scale?
- Value hypothesis
- Interpretation: bounded engagements that strengthen PD, teacher-designed supports, and policy clarity align with this evaluation; they cannot promise learning gains.
- Potential engagement
- Policy-and-PD readiness assessment tied to a small set of approved tools and teacher-created classroom supports.
- Unsupported claims
- Guaranteed student achievement gains
- That generic chatbots matched teacher-designed supports
- Statewide generalizability beyond six pilot LEAs
- Math tutoring accuracy without human review
Pre-sales engineering
Role takeaway
- Fit
- Stronger for teacher workflow and rubric-grounded classroom aids; weaker for unconstrained student generative use given reported math errors and ambivalence.
- Architecture and integration
- Integrate only within district-approved toolset; prefer teacher-authored bots/study tools bound to course rubrics and materials.
- Prerequisites
- Approved-tool list and accounts
- Teacher PD time
- Clear student-use parameters
- Content/rubric grounding for classroom bots
- Constraints
- Pilot was discrete grade/content groups
- Self-report evaluation methods
- Frequent AI mistakes in math tasks
- Security
- Summary emphasizes ethics/appropriate use; does not document DPA or training-data controls.
- Proof of value
- Compare teacher prep/differentiation time before vs after with human review of outputs
- Sample student accuracy critiques of AI math responses
- Confirm classroom bots stay rubric-bound
Delivery
Role takeaway
- Work
- Stand up PD/peer-collaboration cycles, draft actionable AI-use policies, and pilot teacher-designed classroom supports with monitoring for accuracy and ethics.
- Dependencies
- CSDE/district approved tools
- Teacher schedule time
- Policy adoption by district leadership
- Ownership
- District instructional leadership with CSDE guidance; CCERC evaluated, did not operate, the pilot.
- Skills and adoption
- Teachers need sustained practice building content-specific AI supports; students need guided accuracy/ethics skills.
- Governance checkpoints
- Written AI-use parameters before classroom expansion
- Ethics guidance published to staff/students
- Periodic review of math/accuracy incidents
- Acceptance criteria
- Proposed: majority of pilot teachers complete PD sequence
- Proposed: each participating course publishes allowed AI uses
- Proposed: sampled AI math outputs reviewed by teachers weekly
- Risks
- Ambivalence and ethics concerns stall adoption
- Accuracy failures in STEM undermine trust
- Policy inconsistency across schools
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?
Classroom value skewed to teacher-designed, content-bound AI supports rather than generic student chatbots; tool choice left to districts among CSDE-approved options.
Governance
Who approves, reviews and stays accountable for outcomes?
Recommendations call for clear, consistent district AI-use policies and CSDE model-policy examples; ethical parameters required for engagement.
Security and privacy
What data, permissions and controls need testing?
Ethics and appropriate-use concerns raised by participants; summary does not report FERPA/DPA compliance metrics.
Accessibility and workforce
Who is affected, and what skills or accommodations follow?
PD and peer collaboration framed as vital; teachers request more time/support; AI literacy alongside technology skills for all experience levels.
Procurement
What should contracts, pricing and exit terms secure?
Districts used CSDE-approved AI tools purchased with pilot funds; summary does not score vendors.
Operating model
Which teams own the service once it runs?
Success tied to teacher-created resources and classroom structures that incorporate AI ethically; expanded student use met with majority ambivalence.
Publication history
- 2026-10-12K–12 · Issue 363 resources
Stable resource ID: ct-ccerc-ai-pilot-evaluation-exec-summary-202610