A workable university generative-AI policy does three things at once: sets institution-wide safeguards, leaves room for course and discipline-specific decisions, and gives people practical support to follow the rules. Draft it as part of the university’s existing governance system—not as a standalone technology statement. Cover teaching, learning, assessment, research and relevant professional work; define who is accountable; protect confidential data; require human judgement; and establish a review process that can keep pace with changing tools.
The policy should turn principles such as human agency, equity, privacy and academic integrity into instructions a student, lecturer, supervisor or researcher can apply to a real assignment or project. UNESCO’s guidance, published 7 September 2023 and updated 16 January 2026, frames this as a human-centred approach. Its foreword by UNESCO Assistant Director-General for Education Stefania Giannini states: “AI must not usurp human intelligence.”
As an Amazon Associate I earn from qualifying purchases.
Start with the policy’s purpose and place in university governance
Write a short purpose clause before defining permitted prompts or naming tools. Explain what the policy is intended to achieve: support learning and research, protect people and information, preserve academic and research integrity, and clarify responsibility when generative-AI systems are used.
What’s actually slowing this PC down?
Pick the symptom - the matching free tool is one click away.
State that the document is a university baseline. Faculties, departments and instructors may set stricter rules for a course, assessment or research project when the learning outcome, professional standard or risk requires it. Explain how conflicts are resolved. A sensible hierarchy is:
#1 Best Overall
- [Ultra-Lightweight Ear Buds Designed for Small Ears] Each earbud weighs only 3.7g and features a compact, ergonomic in-ear design made especially for small ears. Secure, low-profile, and comfortable for workouts, all-day wear or overnight sleeping.The ultra-slim profile sits flush in your ear with no hard edges poking or rubbing, so you can even sleep on your side without discomfort.
- [Immersive Stereo Sound with TOZO OrigX Technology] TOZO OrigX tuning delivers clear vocals, balanced mids, and natural stereo sound for music, podcasts, and videos.
- [Long Battery Life for Daily Use] Get up to 7 hours of playtime on a single charge, with up to 32 hours total using the charging case—ideal for workdays, commuting, and extended listening sessions.
- [Bluetooth 5.3 & Stable Connection] Bluetooth 5.3 provides fast pairing, stable wireless performance, and reduced dropouts as you move around home or office.
- [Deep Bass with Clear Vocals] High-performance drivers produce punchy bass while keeping vocals clean and detailed for everyday listening.
- Applicable law, regulation, contractual obligations and ethics approvals.
- Binding university regulations, such as the assessment code, research-integrity procedure, privacy and information-security rules.
- This generative-AI policy.
- Course, programme, laboratory or project instructions that add context-specific requirements.
- Frequently asked questions and operational guidance, which explain the rules but do not silently change them.
Make the owner and approval authority visible. A senate, academic board, research committee or another body with delegated authority should approve substantive policy changes. Name an operational owner—often an education, research, digital, library or information-governance office—for guidance, training and questions.
University examples show why this hierarchy matters. The University of Liverpool says its generative-AI teaching guidance operates alongside Appendix L of its Code of Practice on Assessment, which supersedes the guidance in regulatory terms. Oxford places research use alongside existing standards for transparency, rigour, respect, data protection, intellectual property, export control, information security and compliance. Those arrangements are examples, not universal legal templates; each institution must map its own rules.
Define scope and terms that people can understand
Scope should answer who, what and where. Include students, instructors, teaching assistants, supervisors, researchers, professional and administrative staff, visiting scholars and contractors where they use university systems or information. List the activities covered: teaching preparation, learning, assessment, research design and communication, data analysis, publication, administration and student support. If a faculty recommendation excludes researchers without teaching duties, say so explicitly rather than leaving a gap.
Do these 3 things before closing this tab:
1Fix the driver behind crashes, sound loss and screen glitches2Repair Windows errors before they cause bigger problems3Scan for outdated or missing drivers - takes under a minuteDefine “generative AI” broadly enough to remain useful as products change: systems that produce or transform text, code, images, audio, video, data or other outputs in response to prompts or supplied material. Explain that a tool may be public, institutionally provided, embedded in another application or operated locally. Avoid a product-name list that becomes obsolete; maintain an approved-tool register separately.
Distinguish assistance from delegation. A spelling suggestion, translation checked by the user, brainstorming dialogue or code explanation is not the same as submitting an unverified generated answer, allowing a system to make a final grade decision or outsourcing a research interpretation. The policy should regulate the activity and risk, not only the brand or interface.
Turn values into actor-specific requirements
Begin with a small set of principles, then attach a minimum expectation and an accountable actor to each one. “Use AI responsibly” is not an operational rule unless people know what to disclose, what data may be entered and who checks the result.
Rank #2
- 【IP68 Waterproof Multifunctinal Smart Watch】Torjalph fitness activity tracker smart watch is IP68 waterproof technology ,you can wear it when swimming,handwashing,it's rain-proof which means that you perform multiple exercise modes outside without worrying about sweat or rain corroding your fitness smart watch,and this watch equipped with a 230 mAh large capacity battery magnetic charging ,and this sports activity tracker watch can be used 5-7 days If fully charged.
- 【Larger 1.83'' Full Screen Watch & Multi-dial Selection】1.83 inch full HD color touch screen smart watch for men / women provides excellent 240*284 resolution image quality ,This Bluetooth smart watch offers a better touch experience and remarkable immersion and ,more than 200 watch faces for you to choose from App"Da Fit ", you can change the dial style according to your mood to wear keeping the world on your wrist
- 【Accurate Activity Tracker Smart Watch 】sports watch activity tracker uses sweat and dust resistant breathable silicone strap, our Fitness Tracker Smart Watch for Android iOS Phones supports 12 sports modes,walking,running,bicycle,swimming,badminton,basketball,football,rope skipping. This bluetooth calls smart watch for men women is lightweight and stylish, bringing more convenient and relaxing sports experience while also providing more protection.
- 【Your 24 Hour Personal Daily Health Monitor】Fitness watch for women men full touch screen IP67 waterproof bluetooth smartwatch for Android iOS Phones with heart rate blood pressure monitor multifunction Andriod smart watch with steps counter sleep monitor sports fitness watch, monitoring blood oxygen saturation allows you to understand your physical condition in real time to track fitness progress ,to know your health status in time.
- 【More Functions Bluetooth Call Smart Watch】This fitness activity smart watch full screen for iPhone and Android phones includes a bunch of new features including:The bluetooth smart watch can hang up the phone, receive and display WhatsAPP /Text /Facebook /Twitter /Messenger reminder.Support sedentary reminder, alarm clock, stopwatch, timer,weather information,music control, remote camera control compatible with Android 5.0 and above or iOS 9.0 and above smartphones.
| Principle | Minimum institutional rule | Primary accountability |
|---|---|---|
| Human agency | A qualified person remains responsible for decisions, submitted work, grades, advice and published claims. | Every user; supervisors and assessors for delegated decisions |
| Academic and research integrity | Do not present generated material as unaided original work; verify claims, references, calculations and code. | Students, authors, researchers and instructors |
| Privacy and confidentiality | Do not enter personal, confidential, unpublished or protected information unless an approved process and enforceable protection permit it. | All users; data owners and procurement teams |
| Equity and inclusion | Provide an accessible route to complete required work without an unapproved tool and assess the intended skill fairly. | Course teams, programme leaders and support services |
| Transparency | Record or disclose material tool use, limitations, verification and mitigation in the format specified for the activity. | Students, authors, researchers and staff |
| Security and stewardship | Use approved accounts and storage, follow retention rules and report suspected exposure or harmful output. | Users, IT security, privacy and research-data offices |
Include an explicit statement that generated output is probabilistic and can be biased, incomplete, fabricated or inconsistent. Human review is a control, not a promise that every error will be found. Define escalation routes for uncertainty, suspected misconduct, data exposure and discriminatory or unsafe output.
Use a drafting sequence that exposes gaps early
-
Map the existing rulebook
Inventory the assessment code, academic-integrity process, research-integrity policy, ethics-review requirements, privacy and data-protection standards, cybersecurity controls, intellectual-property policy, records schedule, procurement rules, accessibility duties and staff or student codes. Interview the offices that own them. Record which document prevails when wording differs.
-
Identify risk-bearing contexts
List activities where a wrong or undisclosed output could harm a person, invalidate an assessment, breach confidentiality, compromise a study or create a legal obligation. Treat clinical, professional, laboratory, human-subject, export-controlled and commercially confidential work as contexts requiring specialist review.
-
Define terms, people and boundaries
Set the scope, definitions, institutional baseline and permission for stricter local rules. State whether personal devices, third-party accounts, plugins and locally run models fall within the policy.
-
Write principles, then duties
For each principle, specify what students, instructors, supervisors, researchers, administrators, procurement officers and support teams must do. Include a named contact or service for questions.
Special offer. See more information about Outbyte and uninstall instructions. Please review EULA and Privacy policy. -
Build the teaching and assessment rules
Give permission categories, disclosure and citation instructions, assessment-design safeguards, accessibility provisions and a fair process for questions or suspected violations.
Rank #3
New Bee Guitar Cable 10ft Electric Instrument Bass AMP Cord | 6.35mm Jack, Durable Design, Stage and Home Studio Use, Low Noise Shielded Electric Cord for Bass, Keyboard, Amp, Mixer- [Low Noise High Fidelity] New bee guitar cord with 22AWG oxygen-free-copper center conductors provides enhanced signal transfer, the PVC insulation material and the high-density stranded copper double shield provide noise rejection for quiet operation
- [Ideal for Choice] New bee 1/4 straight and right angle metal connectors, ensures a secure connection in any instrument, pedal, or amp jack. L shape great fro on stage guitar stands
- [Wide Compatible] Suitable for mixer, power amplifier, microphone, effector, decoder, equalizer, electoronic organ, electric guitar, electric bass and other electronic equipments. Widely used in the installation of smart home, karaoke speaker, professional stage audio, etc
- [Easy Disassemble] Making incidental repairs quick and easy! Very nice, flexible cable
- [Dont Hesite to Order] 24 hours customer services and professional technology team are standing by
-
Build a separate research section
Cover data classification, tool approval, human verification, attribution, documentation, reproducibility, ethics and incident handling. Do not assume that a classroom rule adequately protects research data or publication integrity.
-
Attach implementation material
Provide syllabus language, disclosure examples, a prompt or interaction log template where appropriate, approved-tool information, training modules, decision trees and contact points. Label these as guidance if they can be changed without formal policy approval.
-
Test and approve
Run the draft through representative cases: an open-book essay, a closed examination, a coding assignment, a group project, a grant proposal, a study containing personal data and a public-engagement video. Obtain review from students, teaching staff, researchers, disability and library services, privacy, security, legal, procurement and ethics offices before formal approval.
Windows Errors? Fix Them Before They Spread
Repair common Windows errors and clear accumulated junk for a smoother, more stable PC - no reinstall needed.Free scan · no reinstallCrashes, No Sound, or Screen Glitches?
Random freezes, missing sound and display glitches usually trace back to one bad driver. Find and replace yours safely.Free scan · under a minuteSpecial offer. See more information about Outbyte and uninstall instructions. Please review EULA and Privacy policy.
Set visible classroom and assessment expectations
Students should know the rule before they begin work, not after a submission is challenged. Require each course or assessment to state whether generative AI is allowed, restricted or prohibited, what functions are allowed, what must be disclosed, what evidence must be retained and how questions will be answered. Geneva Faculty of Science recommendations (version December 2025) explicitly say teachers may allow or prohibit generative AI and should communicate course-specific rules in writing and in a timely manner.
| Permission level | Typical allowed examples | Controls to state |
|---|---|---|
| Allowed with disclosure | Brainstorming, translation followed by checking, self-testing, summarising a source, coding help, illustrations or alternative explanations. | Identify the tool and purpose; check accuracy, bias and copyright; cite or log material use if required. |
| Restricted | Editing language, generating practice questions, outlining or debugging when the assessment measures independent reasoning or unaided code. | Define the permitted function, prohibit replacement of the assessed work and require a statement or retained interaction record. |
| Prohibited | Submitting generated answers as one’s own, supplying examination questions to a system, or using a system to perform the skill being assessed. | Explain the learning outcome and the academic-integrity process; provide an alternative route for an approved accommodation. |
The category must follow the learning outcome. If an assignment assesses argument construction, a system that writes the argument changes the task; if it assesses editing or translation, limited assistance may be compatible. Course teams should redesign tasks where the intended skill cannot be observed reliably. Oral questioning, a demonstration, staged drafts, version history or an in-class component can check mastery without treating an automated detector as proof. Geneva notes that no reliable method exists for identifying generative-AI use; enforcement should rely on the work process, corroborating evidence and fair institutional procedure rather than detector scores.
Give students a disclosure format
A policy should supply a short, adaptable statement rather than merely demanding “cite AI.” For example:
Rank #4
- Digital Family Calendar: BSIMB digital calendar wall touch screen puts an end to your chaotic life. Designed for busy families, the digital calendar keeps you organised and in step with your family. Assign different colours and tasks to each family member to make family scheduling easier and more intuitive.
- Easy Setup and Auto-Sync: Plug in, connect to WLAN and auto-syn your electric calendar in minutes. Digital wall calendar compatible with Google calendar, iCloud calendar, Outlook calendar, Cozi calendar and Yahoo calendar. Add events directly on the device or through the ecalendar app. Electronic calendar stays current whether you're at work or traveling—no manual updates needed.
- All-In-One Home Assistant: Smart calendar builds habits and encourage independence with personalised chores. Display daily meal plan to reduce the daily ‘What's for dinner?’ stress. To-do lists keep track of everything you want. Turning chores into valuable achievements with stars and rewards, making life easier and more enjoyable.
- Manage Anytime Anywhere: BSIMB electronic calendar for family stay connected with the powerful ecalendar app. Access and manage your digital wall calendar anytime, anywhere with free features like stars & rewards, photo screensaver, chore charts, meals plans, to-do lists etc. Electronic planner paid features like Magic Import tailor advanced tools for users with special needs.
- Stay in Touch with Loved Ones: Share photos or videos from your phone to electronic wall calendar with app, instantly update your life moment, and keep in touch with your family, especially those who are far away; you can also surprise your smart calendar digital wall by sending him a wish card; it's also great for keeping tech-illiterate elders in touch with the youngsters.
“I used [tool and version, if known] for [specific function] during [stage of work]. I reviewed the output against [sources, tests or instructions], corrected errors and remain responsible for the submitted work.”
Recommended Free Tools
Tell students whether they must retain prompts, outputs, dates, settings or screenshots. Do not require disclosure of a trivial spelling correction if the course has deliberately excluded it; define material use so expectations are proportionate.
Tell instructors what they may delegate
Institutional guidance can support brainstorming lesson ideas, drafting lesson plans or supplementary materials, summarising feedback, preparing multiple explanations, drafting communications for review, creating low-stakes test questions, refining rubrics and producing visual or audio/video materials. Liverpool lists these as educational use cases while stating that AI should not generate grades or final assessment decisions. Require an instructor to verify accuracy, accessibility, tone, copyright and alignment with the course outcome before publication or use.
Make research a distinct policy domain
Research use needs controls for information that may be confidential, personal, unpublished, commercially restricted or subject to export controls. Oxford’s policy says sensitive or confidential information should be uploaded only where confidentiality can be guaranteed; it treats personal-data uploads as processing and says research use requires ethics approval. Your policy should direct researchers to the local privacy, security, contracts and ethics processes instead of promising that any particular public tool is safe.
Use a data-classification decision
| Material | Default treatment | Required action before use |
|---|---|---|
| Public, non-sensitive information | May be used with an approved or otherwise permitted tool. | Check terms, output reliability, attribution and retention; do not assume public input produces trustworthy output. |
| Internal university information | Use only an institutionally approved environment with the required access controls. | Confirm contract, retention, access, logging and deletion terms with the information owner. |
| Personal, confidential, proprietary or contract-restricted data | Do not upload by default. | Obtain documented privacy, security, ethics and contractual approval and use a tool that guarantees the required protections. |
| Unpublished results, peer-review material or protected research data | Do not enter into public platforms. | Follow the project’s data-management plan, sponsor terms and research-ethics decision; seek specialist advice. |
Geneva’s recommendations expressly prohibit entering sensitive or confidential data and unpublished research results into generative-AI tools. That is a useful baseline, but local contracts and approvals may impose stricter conditions.
Require verification, attribution and a usable record
- Check every substantive claim, quotation, citation, calculation, translation and code fragment against authoritative sources or tests. Generated references can be fabricated or misattributed.
- Describe material tool use in methods, acknowledgements or another appropriate section. Oxford recommends disclosing the tools used, their limitations and the mitigation measures taken.
- Keep enough information to reproduce the relevant step: tool name and version where available, date, purpose, important settings, prompts or a structured summary, supplied data category, output used and verification performed. Protect logs that contain personal or confidential information.
- Explain how a changing cloud model, unavailable version or non-deterministic output affects reproducibility. Oxford warns that evolving cloud models can create transparency and reproducibility problems.
- Preserve human authorship and accountability. A system cannot approve an ethics amendment, interpret a participant’s welfare, make a final authorship decision or replace a qualified researcher’s judgement.
Connect research use to integrity and incident procedures
State that undisclosed or misleading use may be considered under the existing research-integrity process, while accidental exposure, harmful output or a security event follows the university’s incident route. Include contacts for the research office, ethics committee, privacy officer, information security and intellectual-property advice. Do not create a parallel investigation system unless governance has approved it.
Independent reader supportYour contribution helps us test, update, and keep practical guides available for everyone.Assign responsibilities across the institution
| Role | Required actions |
|---|---|
| Students | Follow course permissions; protect others’ data; verify output; disclose material use; retain evidence when required; ask before acting on uncertainty. |
| Instructors and assessors | Set written, timely rules; align permissions with learning outcomes; provide an equitable alternative; verify teaching materials; never delegate final grades or assessment decisions. |
| Supervisors and principal investigators | Classify project data; approve tools and workflows; ensure ethics, sponsor, privacy and security compliance; train team members; document verification and disclosure. |
| Researchers and authors | Check claims, references, code and images; disclose material use; preserve reproducibility; protect confidential and unpublished material; report incidents. |
| Professional and administrative staff | Use approved systems; review generated communications and decisions; avoid entering personal or confidential information; apply records and accessibility requirements. |
| Procurement and IT | Assess vendors, contracts, retention, model-training terms, security, accessibility, export controls and exit arrangements; maintain an approved-tool register. |
| Library, learning, research-data, privacy, security and disability services | Provide specialist advice, examples, training and accessible alternatives; receive and route questions and incidents. |
| Policy owner and governing body | Monitor cases and feedback, publish guidance, coordinate reviews and approve substantive amendments. |
Pair obligations with training and approved support
A rule without a supported route to compliance will be ignored or applied unevenly. Build a training programme for students, instructors, researchers and professional staff. Cover prompt and output limitations, hallucinations, bias, privacy, copyright and intellectual property, accessibility, disclosure, verification, research-data classification and incident reporting. Use discipline-specific cases rather than a single generic lecture.
Maintain an approved-tool page that states the permitted account type, data restrictions, retention and deletion terms, support contact, accessibility information and review date. Give users a non-generative alternative when a course or workflow requires a tool they cannot safely or equitably access. Procurement should evaluate contracts and security before a tool is promoted as institutionally supported; an informal free account is not an approved environment merely because it is convenient.
The OECD’s 17 July 2026 Education Spotlight identifies five emerging system-level response areas: government and national guidance; compliance and procurement approaches; GenAI competencies; evidence collection and pilot evaluation; and specialised tools for higher education. Treat this as a planning framework, not a statistic about adoption. A university can map each area to an owner and a deliverable.
Design monitoring, review and change control
Assign a standing group to review incidents, appeals, student and staff feedback, accessibility impacts, tool changes, new legal or contractual requirements and evidence from pilots. Set a regular cadence and publish the next review date. Oxford reports a policy advisory group that reviews its policy and FAQs termly; that cadence is Oxford’s institutional choice, not a universal requirement.
Separate changes that require formal approval from administrative updates. A new prohibition, altered assessment consequence or changed scope belongs in the approved policy. A corrected contact, example disclosure, tool-status note or explanatory FAQ may be updated by the policy owner if the governance instrument allows it. Record the effective date and version for every change.
Monitor outcomes without creating a surveillance system. Useful indicators include the number and type of questions, training completion, reported data incidents, appeals, accessibility concerns, assessment redesigns and research disclosures. Interpret them carefully: a rise in reports may indicate better awareness, not worse conduct. Publish a plain-language change log so users can see what changed and why.
Provide ready-to-use implementation language
Course or assessment notice
“For this assessment, generative-AI use is [allowed for the following functions / restricted to the following functions / not allowed]. You remain responsible for the accuracy, originality and integrity of your submission. If you use a permitted system, disclose the tool, purpose and material contribution in the format below. Do not enter personal, confidential or unpublished information. Questions about an ambiguous use should be sent to [course contact] before submission. The university’s assessment and academic-integrity regulations apply.”
Free tools Windows power users keep installed
One-click scans. No signup required.
Research-project notice
“Before using a generative-AI system, classify the material and confirm that the tool, account and contract meet the project’s privacy, security, ethics, sponsor and intellectual-property requirements. Do not upload personal, confidential or unpublished information to a public platform. Verify generated claims, references, code and analysis; record material interactions and limitations; and disclose use in the methods, acknowledgements or other required section. Report suspected exposure or misleading output through the project’s incident route.”
Common drafting failures and their fixes
- One blanket permission or ban. Replace it with activity- and outcome-based categories, because a translation exercise, a closed examination and a research dataset carry different risks.
- Principles without procedures. Add disclosure examples, data rules, contacts, approval paths and consequences under existing regulations.
- Ignoring the regulatory hierarchy. Identify the assessment code, research-integrity process and privacy rules that prevail when guidance conflicts.
- Assuming a detector proves misconduct. Use process evidence, conversation, drafts and fair investigation; detector scores alone are not reliable proof.
- Treating research as an extension of classroom guidance. Add data classification, ethics, contracts, reproducibility and publication disclosure controls.
- Approving tools without contractual review. Check retention, model-training, access, deletion, security, accessibility and exit terms before recommending a service.
- Writing rules that cannot be accessed. Provide training, accessible alternatives, timely course notices and a route for accommodations.
- Freezing the policy. Keep binding requirements stable but maintain versioned FAQs, examples and tool information with a named owner and review date.
Final approval checklist
- Purpose, scope, definitions and policy hierarchy are explicit.
- Teaching, learning, assessment, research and relevant institutional work are covered or deliberately excluded.
- Students receive timely, written course and assessment rules.
- Allowed, restricted and prohibited uses include concrete examples.
- Disclosure, citation, prompt-retention and verification expectations are proportionate and clear.
- Human review is required for grades, decisions, claims, references, code and research interpretations.
- Personal, confidential, protected and unpublished information has a default-safe treatment and an approval route.
- Privacy, security, intellectual-property, procurement, ethics and research-integrity owners have reviewed the draft.
- Training, approved-tool information, accessible alternatives and contact points are funded and published.
- Incidents, questions, appeals and suspected violations follow fair existing procedures.
- A named owner, governing body, review cadence, version history and change-control method are in place.
- Local legal and contractual requirements have been checked before the policy takes effect.
A strong policy therefore acts as a shared safety floor, not a script for every discipline. It protects human judgement and sensitive information while allowing instructors and researchers to make context-specific decisions. Build it into existing governance, publish usable rules before work begins, fund the support needed to follow them and review both the policy and its practical effects on a defined schedule.
Quick Recap
Product prices and availability are accurate as of the date/time indicated and are subject to change. Any price and availability information displayed on Amazon at the time of purchase will apply.
