AI literacy and assessment integrity need separate workstreams

Current synthesis

This synthesis integrates the former claim page “Schools should separate AI literacy work from assessment integrity work” into one page, so the concept now lives here rather than as a duplicate claim.

Schools need to separate two related but different jobs: helping students learn to use AI well, and redesigning assessment so teachers can still see what students understand. Separate AI Literacy and Assessment Integrity Mature AI Use vs. Immature AI Use Why Your AI Committee Keeps Stalling

AI literacy asks students to practice judgment, prompting, verification, metacognition, and reflection with AI, not simply avoid AI or use it as a shortcut. AI literacy should teach critical doing with AI AI literacy takes system capacity, not just tool access Why Your AI Committee Keeps Stalling

Assessment integrity asks schools to gather reliable evidence of thinking through process, conversation, visible reasoning, performance, and teacher observation instead of trusting final products by themselves. In an AI world, assessment should focus on watching students think AI has made unsupervised take-home essays unreliable evidence Make Them Teach It

The University of Sydney’s two-lane approach adds a program-level version of the same split: secured assessment of learning should validate outcomes at selected progression points, while more open assessment can support learning and AI practice in ordinary coursework. Program level assessment design and the two-lane approach

The two workstreams should inform each other, but merging them too quickly can make AI literacy feel like cheating prevention and make assessment reform feel like tool training. Separate AI Literacy and Assessment Integrity AI-assisted homework requires redesign, not just policing Why Your AI Committee Keeps Stalling

For district work, this means teams should plan AI literacy instruction, assessment redesign, teacher learning, policy, and tool governance as coordinated but distinct parts of a long-term redesign project. District AI work is a long-term redesign project District AI implementation needs living guidance and teacher-led redesign Why Your AI Committee Keeps Stalling

Practical implications

• Treat AI literacy and assessment integrity as separate but coordinated workstreams, with separate goals, experiments, and success measures.
• Build AI literacy lessons that teach students how to use, question, verify, and reflect on AI use.
• Build assessment systems that make student thinking visible before, during, and after AI-supported work.
• Avoid using every AI literacy conversation as a cheating conversation.
• Avoid treating every assessment-integrity problem as a student behavior problem.

Evidence carried over from the merged claim

• Mature AI Use vs. Immature AI Use supports this synthesis through its discussion of school AI policy, classroom AI-use rubrics, transcript-based assessment, student metacognition, and the need to separate AI literacy from academic-integrity enforcement.
• Separate AI Literacy and Assessment Integrity argues that AI literacy and assessment integrity get conflated in faculty meetings and AI working groups, then recommends separate work groups, experiments, and success metrics.
• I’m (Kind Of) Changing My Mind About Blue Books reinforces the assessment half of the synthesis by arguing that polished final products are weaker evidence of learning in an AI era and that educators should privilege visible thinking and process evidence.
• Program level assessment design and the two-lane approach reinforces the assessment-integrity half of the synthesis by distinguishing secured validation moments from open AI-supported learning tasks.
• The Opposite of Cheating Podcast (Season 2) Episode 59: Joe Clare reinforces the assessment-integrity half of the synthesis by treating misconduct prevention as an opportunity-design problem: courses can increase effort and risk, reduce rewards and excuses, and make honest work the easier path.

Related pages

• AI literacy requires different kinds of AI interaction
• Learning still needs some struggle, even when AI can make things easier
• District AI work is a long-term redesign project

Synthesis history

• Created after the 2026-05-27 weekly wiki synthesis review and Clay’s approval.

• 2026-06-24: Merged duplicate claim page into this synthesis so the wiki has one integrated page for the concept.