{"schema_version":"ICEQC_NEWS_ARTICLE_V1","id":"iceqc-news-ab476a3dbf36b2bd","slug":"2023-12-15-evidence-across-delivery-settings-for-teacher-ai-capability","language":"en","publication_status":"READY_FOR_IMPORT","publication_date":"2023-12-15","last_modified_date":"2023-12-15","title":"Evidence across delivery settings for teacher AI capability","summary":"Sets out the scope, decision criteria and evidential basis relevant to evidence across delivery settings for teacher AI capability.","category":{"code":"STANDARDS_INTERPRETATION","label":"Standards Interpretation"},"article_type":"Standards interpretation","publisher":"International Council for Education Quality Certification (ICEQC)","jurisdictional_scope":"International","historical_reference_basis":"Generative AI policy and practice","reference_authority":"United Nations Educational, Scientific and Cultural Organization","sections":[{"heading":null,"paragraphs":["The generative AI policy and practice provides the immediate context for teacher AI capability. A decision concerning the relevant requirement should recognise that the central issue is the meaning of the expectation in practice, including its scope, the evidence needed to demonstrate it and the circumstances in which it may not apply. Uniform administrative form is not required where equivalent public outcomes can be demonstrated.","Failure in relation to the issue under review may arise even where the stated policy is reasonable. Material concerns include opaque use of personal or inferred data, unverified outputs entering teaching or assessment, unequal performance across learner groups, and automation bias in consequential decisions. In relation to evidence across delivery settings for teacher AI capability, materiality depends on the consequence and extent of an exception, not only on how often it appears in sampled records."]},{"heading":"Applicable scope","paragraphs":["Responsibility for teacher AI capability should be visible at the point where consequential decisions are made. the issue under review, the assessment question is whether the control operates across the relevant sites, programmes, delivery modes and learner groups, including material exceptions. Incomplete evidence, unmanaged conflict, absent learner groups or material learner impact require a higher level of review.","The historical reference basis is the generative AI policy and practice. Its relevance to the issue under review should be assessed against the affected jurisdiction, learner population and form of provision. The international development warrants attention, but a consequential conclusion still requires current, attributable and representative evidence for the affected scope.","International guidance on generative artificial intelligence in education and research was released in September 2023. It calls for a human-centred approach, protection of data privacy, age-appropriate use, validation and institutional capacity. Immediate provider controls should address authorised uses, assessment, disclosure, information security, unequal access and human review while evidence on educational benefit and risk continues to develop.","The analysis of the assurance question should make its decision rule explicit. A decision concerning the control should recognise that the subject should be examined as a connected system of policy, people, resources, decisions and evidence. Transfer of decisions or records can expose weaknesses not visible in separate reviews of individual controls. A stated decision rule enables comparable examination and limits retrospective explanations of adverse evidence."],"bullets":["Retain accountable human decision-makers.","Prohibit uses for which evidence or authority is insufficient, identifying the accountable function and affected scope.","Notify users of material limitations.","Control personal and confidential information.","Review incidents and supplier changes before it is relied on for a decision with material effect."]},{"heading":"Implementation and evidence","paragraphs":["Interpretation of teacher AI capability should avoid two errors: treating a formal commitment as proof of effect, and treating one adverse case as proof that every part of the system has failed. For the stated expectation, a technical capability is not evidence that a use is educationally justified. Accuracy measured in one setting may not transfer to another population, language, curriculum or decision context. Oversight of the assurance question should reflect the principle that a prescribed method should not be treated as the only acceptable method where another approach establishes the same outcome with equivalent evidence.","Assurance of the issue under review should draw on more than one form of evidence. Useful records include documented authority for each consequential use, an inventory of systems and their intended uses, pre-deployment and periodic performance testing, learner information and accessible challenge routes, and supplier change and incident records. Evidence of effectiveness should represent the declared scope, including adverse and exceptional cases.","Records relating to the assurance question should preserve both the conclusion and its limits. The correction record should state what the new evidence changes and which earlier conclusions or decisions require review. In relation to evidence across delivery settings for teacher AI capability, where reliance has occurred, correction may require review of affected decisions as well as amendment of published information."]},{"heading":"Assessment of conformity","paragraphs":["The review method for teacher AI capability should be reproducible. the assurance question, the reviewer should map the complete process, identify the intended result and responsible authority at each stage, and test normal cases together with exceptions. Review should determine whether correction can remain case-specific or must extend across the system. Working papers should allow another competent reviewer to understand the evidence, judgement and treatment of material exceptions.","Interpretation of the control should produce a test that another competent reviewer can apply to comparable evidence."],"bullets":["What outcome is intended?","Where do exceptions occur?","Which evidence establishes operation?","What action is required by the finding?","Who controls each stage?"]},{"heading":"Review and corrective action","paragraphs":["For teacher AI capability, the public interest is not confined to institutional compliance. A decision concerning the assurance question should recognise that technology may support teaching, administration and access, but consequential educational decisions must remain accountable, explainable and open to effective review. Learner protection requires intelligible information and a timely means of reviewing consequential mistakes or unfair decisions.","the issue under review, governing bodies should receive a concise account of the intended result, affected scope, principal risks, evidence limitations and unresolved exceptions. Material action requires a named responsible function and a defined completion point. Closure requires evidence that the condition has changed; completion of planned activity is not sufficient.","The appropriate response to the control is therefore one of controlled implementation and review. The decision record should state the unsupported element and the further work required."]}],"word_count":881,"content_hash":"sha256-7c464cd8c29fbe900587be24d2ac5a3eb86528bdb91920d89593c926388f41fe","seo_keywords":["evidence across delivery settings for teacher AI capability","education quality standards","education quality","ICEQC"],"schema_type":"TechArticle"}