Policy analysis of higher education institutes in Ontario, Canada: A focus on artificial intelligence

Classified as probably not a substantial claim. May combine two claims.

Policy plays a critical role in governance and equipping students and educators with the knowledge and skills required to navigate and use AI ethically in their academic work.

“It has been suggested that policy plays a critical role in governance and equipping students and educators with the knowledge and skills required to navigate and use AI ethically in their academic work (Chan, 2023).”

• cites Chan, 2023 sourceSupported

Classified as probably not a substantial claim.

Current policy approaches across the higher education sector fail to recognize and adequately address the transformative impacts of AI on teaching and learning.

“Chan (2023) argues that current policy approaches across the HE sector fail to recognize and adequately address the transformative impacts of AI on teaching and learning.”

• cites Chan, 2023 sourcePartly supported

  In “Existing policy on AI in education” (p. 6), Chan says policies are mostly generic and that AI’s use in education is largely absent from policy conversations. In “Key areas versus UNESCO’s recommendations” (p. 19), she says most existing policies emphasize workforce development rather than AI’s transformative potential in higher education. These passages support the criticism, but the claim’s categorical assertion that policy approaches across the higher education sector fail to recognize and address those impacts is stronger than the paper’s qualified account. (LLM, conf=0.85)

  Experimental — accuracy not yet measured

Classified as probably not a substantial claim. May combine two claims.

At the time of Chan's study, educational institutions had not sufficiently funded, regulated, or explored the ethical implications of AI in pedagogical and operational practices and contexts.

“She noted that, at the time of her study, educational institutions had not sufficiently funded, regulated, or explored the ethical implications of AI in pedagogical and operational practices and contexts.”

• cites Chan, 2023 sourcePartly supported

  The “Existing policy on AI in education” section (p. 6) describes generic AI-education policies and insufficient attention to education governance and ethics. But the funding, regulation, and ethics point is presented as Schiff’s warning that policymakers *may* fail in these respects if a trend continues—not as a finding that educational institutions had already failed to do so at the time of Chan’s study. (LLM, conf=0.88)

  Experimental — accuracy not yet measured

Probably not a substantial claim: may not pass S3 (load-bearing).

McDonald et al. (2024) found that 41% of the higher education institutions with policies or documents related to AI use provided clear guidance for use in learning environments.

“They found that 63% encouraged AI use and 41% provided clear guidance for use in learning environments.”

• cites McDonald et al. (2024) sourceSupported

Probably not a substantial claim: may not pass S3 (load-bearing).

McDonald et al. (2024) found that 63% of the higher education institutions with policies or documents related to AI use encouraged AI use.

“They found that 63% encouraged AI use and 41% provided clear guidance for use in learning environments.”

• cites McDonald et al. (2024) sourceSupported

Probably not a substantial claim: may not pass S3 (load-bearing).

AI documents accompanying policies were employed to inform instructors and students about suggested AI syllabi statements, pedagogical approaches, types of recommended AI tools, potential ethical issues, and other general guidance.

“Researchers described how AI documents that accompanied policies were being employed to inform instructors and students about suggested AI syllabi statements, pedagogical approaches, types of recommended AI tools, potential ethical issues, and other general guidance.”

• cites McDonald et al. (2024) sourcePartly supported

  The Overview (pp. 8–9) reports institutional guidance on sample syllabus statements, classroom activities, teaching approaches, and GenAI ethics. The Introduction (p. 2) describes policies as a source of information for faculty and students. The passages do not establish that the accompanying documents advised them on types of recommended AI tools; the discussion of named detection tools (section 4.2.2, p. 12) chiefly cautions against their reliability. (LLM, conf=0.82)

  Experimental — accuracy not yet measured

Probably not a substantial claim: may not pass S2 (checkable). May combine two claims.

The ethical complexities of AI in education demand clear policy, proactive institutional frameworks, and pedagogical strategies.

“Research underscores the ethical complexities of AI in education, such as data privacy, algorithmic bias, AI misuse, misinformation, and disinformation, which demand clear policy, proactive institutional frameworks, and pedagogical strategies (Garcia-López & Trujillo-Liñán, 2025).”

• cites Garcia-López & Trujillo-Liñán, 2025 sourceSupported

Classified as probably not a substantial claim.

Instructors express a need for professional development to build confidence and competence in using AI tools.

“Instructors express a need for professional development to build confidence and competence in using AI tools, with studies identifying AI literacy as a critical factor in successful adoption (Mah & Groß, 2024).”

• cites Mah & Groß, 2024 sourceSupported

May combine two claims.

Two distinct approaches were observed: one outlines roles in the context of responding to breaches, and another emphasizes collective responsibility for upholding academic standards.

“Two distinct approaches were observed: one that outlines roles in the context of responding to breaches and another that emphasizes collective responsibility for upholding academic standards.”

May combine two claims.

In most cases, AI academic misconduct is integrated into existing reporting procedures, with no indication of how or whether these procedures are adapted to address the specific challenges of detecting and verifying AI misuse.

“In most cases, AI academic misconduct is integrated into existing reporting procedures, with no indication of how or whether these procedures are adapted to address the specific challenges of detecting and verifying AI misuse.”

Probably not a substantial claim: may not pass S2 (checkable). May combine two claims.

Across institutional types, language addressing AI is generally clear, free of jargon, and effectively communicates what constitutes dishonest or inappropriate use of AI.

“However, across institutional types, language addressing AI is generally clear, free of jargon, and effectively communicates what constitutes dishonest or inappropriate use of AI.”

Probably not a substantial claim: may not pass S3 (load-bearing).

At the time of review, two institutions had standalone AI policies.

“At the time of review, two institutions had standalone AI policies, although on further analysis, the reviewers deemed only one policy to be a standalone AI policy (see Standalone Policies section).”

Probably not a substantial claim: may not pass S3 (load-bearing).

Seneca College's policy has a dual focus evident in language discouraging discriminatory AI outputs and referencing the institution's broader ethical commitments.

“This dual focus is evident in language discouraging discriminatory AI outputs and referencing the institution's broader ethical commitments.”

May combine two claims.

Most institutions include broad introductory statements outlining the intent of their academic integrity policies, but few extend this framing to explicitly contextualize AI-related content.

“Most institutions include broad introductory statements outlining the intent of their academic integrity policies; however, few extend this framing to explicitly contextualize AI-related content.”

May combine two claims.

All 19 reviewed academic integrity policies that reference AI are publicly available online and can be located within one to five clicks from the institutional homepage.

“All reviewed academic integrity policies that reference AI (n = 19) are publicly available online and can be located within one to five clicks from the institutional homepage (see Table 3).”

May combine two claims.

The dataset included policies from 22 English-language and 3 French-language universities and 22 English-language and 2 French-language technical and vocational colleges.

“The dataset, collected between January and May 2025, included policies from 22 English-language and 3 French-language universities and 22 English-language and 2 French-language technical and vocational colleges (see Table 1).”

Probably not a substantial claim: may not pass S3 (load-bearing).

Seneca College's policy explicitly delineates responsibilities for students, employees, and contractors.

“Further, the policy explicitly delineates responsibilities for students, employees, and contractors, emphasizing adherence to institutional ethical standards and principles of equity, diversity, inclusion, and sustainability.”

Probably not a substantial claim: may not pass S3 (load-bearing).

Seneca College's policy dual focus is evident in language discouraging discriminatory AI outputs and referencing broader ethical commitments.

“This dual focus is evident in language discouraging discriminatory AI outputs and referencing the institution's broader ethical commitments.”

May combine two claims.

Several policies acknowledge AI misuse, often categorized within existing forms of academic misconduct such as plagiarism and contract cheating, but few provide specific criteria or nuanced explanations of the specific actions that constitute a breach involving AI.

“While several policies acknowledge AI misuse-often categorized within existing forms of academic misconduct such as plagiarism and contract cheating-few provide specific criteria or nuanced explanations of the specific actions that constitute a breach involving AI.”

May combine two claims.

One college emphasized ethical practice and integrity, while the other colleges framed AI misuse in terms of academic misconduct or departures from academic integrity.

“One college emphasized ethical practice and integrity, while the others framed AI misuse in terms of academic misconduct or departures from academic integrity.”

Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

Seneca College's policy frames AI as an assistive technology aimed at enhancing efficiency while underscoring the importance of system safety, compliance, and ethical practice.

“Additionally, Seneca College's policy frames AI as an assistive technology aimed at enhancing efficiency while underscoring the importance of system safety, compliance, and ethical practice.”

May combine two claims.

Seneca College's policy frames AI as an assistive technology aimed at enhancing efficiency and underscores the importance of system safety, compliance, and ethical practice.

“Additionally, Seneca College's policy frames AI as an assistive technology aimed at enhancing efficiency while underscoring the importance of system safety, compliance, and ethical practice.”

May combine two claims.

Explicit role definitions are not common, but some institutions incorporate responsibilities regarding ethical AI use within broader academic integrity language.

“While explicit role definitions are not common, some institutions incorporate these responsibilities within broader academic integrity language.”

Probably not a substantial claim: may not pass S3 (load-bearing).

Visual aids, such as flowcharts and tables, are sporadically included in the policies.

“Visual aids, such as flowcharts and tables (which are recommended for clarifying violation and sanctioning pathways), are sporadically included.”

Probably not a substantial claim: may not pass S3 (load-bearing).

The tone and philosophical orientation of academic integrity policies that reference AI generally reflect a formal and constructive stance.

“The tone and philosophical orientation of academic integrity policies that reference AI generally reflect a formal and constructive stance.”

May combine two claims.

Procedures for reporting, recording, and appealing AI-related violations are underdeveloped and inconsistently included.

“Procedures for reporting, recording, and appealing AI-related violations are likewise underdeveloped and inconsistently included.”

Probably not a substantial claim: may not pass S3 (load-bearing).

Seneca College's policy employs clear, precise language to define key terms.

“The policy is easily accessible and employs clear, precise language to define key terms.”

Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

Seneca College's policy adopts an educative and risk-aware tone, highlighting legal, reputational, and third-party risks.

“Although Seneca College's policy does not include dedicated educational directives, it adopts an educative and risk-aware tone, highlighting legal, reputational, and third-party risks.”

May combine two claims.

Three colleges provided explicit statements of purpose and defined AI misuse within the broader framework of academic misconduct.

“Three colleges provided explicit statements of purpose and defined AI misuse within the broader framework of academic misconduct.”

May combine two claims.

A range of support mechanisms (e.g., training, assignment resubmission, workshops, and remediation) are identified in both university and college contexts, but they are not explicitly linked to AI misuse.

“While a range of support mechanisms (e.g., training, assignment resubmission, workshops, and remediation) are identified in both university and college contexts, they are not explicitly linked to AI misuse.”

Probably not a substantial claim: may not pass S3 (load-bearing).

Seneca College's policy is easily accessible.

“The policy is easily accessible and employs clear, precise language to define key terms.”

Probably not a substantial claim: may not pass S3 (load-bearing).

Seneca College's policy emphasizes adherence to institutional ethical standards.

“Further, the policy explicitly delineates responsibilities for students, employees, and contractors, emphasizing adherence to institutional ethical standards and principles of equity, diversity, inclusion, and sustainability.”

Probably not a substantial claim: may not pass S3 (load-bearing).

When present, visual aids typically align with general violation and sanctioning procedures.

“When present, they typically align with general violation and sanctioning procedures.”

The quote only partly states this claim.

Among technical and vocational colleges, 3 institutions required 4 clicks to access GenAI language in academic integrity policies.

“| | 4 | 3 |”

Classified as probably not a substantial claim.

This approach, in many cases, helped make policies relatively easy to locate using standard search terms and minimal navigation.

“This approach, in many cases, helped make policies relatively easy to locate using standard search terms and minimal navigation.”

Probably not a substantial claim: may not pass S3 (load-bearing). The quote only partly states this claim.

The total number of technical and vocational colleges in the access analysis was 10.

“| Total | | 10 |”

Classified as probably not a substantial claim.

This review revealed inconsistencies in tone and approach, with institutions oscillating between educative, punitive, and legalistic language.

“However, this review revealed inconsistencies in tone and approach, with institutions oscillating between educative, punitive, and legalistic language.”

Probably not a substantial claim: may not pass S3 (load-bearing), S4 (based). The quote only partly states this claim.

Among technical and vocational colleges, 2 institutions required 3 clicks to access GenAI language in academic integrity policies.

“| | 3 | 2 |”

Probably not a substantial claim: may not pass S3 (load-bearing).

Of the policies reviewed, 1 artificial intelligence policy was from a technical and vocational college.

“| Artificial intelligence policy | 1* | 1 | 2 |”

Probably not a substantial claim: may not pass S3 (load-bearing). The quote only partly states this claim.

Among technical and vocational colleges, 2 institutions required 2 clicks to access GenAI language in academic integrity policies.

“| | 2 | 2 |”

The quote only partly states this claim.

Among technical and vocational colleges, 3 institutions required 1 click to access GenAI language in academic integrity policies.

“| College | 1 | 3 |”

Probably not a substantial claim: may not pass S4 (based).

In the current landscape, standalone AI policies are rare.

“In the current landscape, where standalone AI policies are rare, an expanded academic integrity policy to explicitly address AI use may offer a practical and scalable strategy.”

Probably not a substantial claim: may not pass S3 (load-bearing).

The total number of universities in the access analysis was 9.

“| Total | | 9 |”

Classified as probably not a substantial claim.

Most universities have not yet developed or published policies supporting the ethical use of AI.

“That is, most universities have not yet developed or published policies supporting the ethical use of AI.”

• cites Stracke et al., 2025Not matched

  Couldn’t be matched to a paper in Crossref, so it can’t be checked.

Classified as probably not a substantial claim.

Students may graduate unprepared for professional environments that increasingly expect competency related to the ethical and proficient use of AI.

“Another concern is that students may graduate unprepared for professional environments that increasingly expect competency related to the ethical and proficient use of AI (Stracke et al., 2025).”

• cites Stracke et al., 2025Not matched

  Couldn’t be matched to a paper in Crossref, so it can’t be checked.

Probably not a substantial claim: may not pass S3 (load-bearing).

UNESCO categorized institutional approaches to AI policy development into three types: independent, integrated, and thematic.

“In an international review, the United Nations Educational, Scientific, and Cultural Organization (UNESCO) categorized institutional approaches to AI policy development into three types: independent (standalone AI policies), integrated (wherein AI elements are embedded within existing technology policies), and thematic (policies addressing focused topics and issues on AI in education) (Fengchun et”

• cites Fengchun et al., 2021Not matched

  Couldn’t be matched to a paper in Crossref, so it can’t be checked.

Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

Emerging ethical concerns include threats to student privacy and social interaction, inequitable access to AI tools, and the risk of learner overreliance on AI technologies.

“Beyond threats to academic integrity, emerging ethical concerns include threats to student privacy and social interaction, inequitable access to AI tools, and the risk of learner overreliance on AI technologies (Vieriu & Petrea, 2025).”

• cites Vieriu & Petrea, 2025 sourceNot checked — no full text of the source

Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

In a review of 15 policies from eight European countries, Stracke et al. (2025) found that AI-specific policies were difficult to find and largely undeveloped.

“In a review of 15 policies from eight European countries, Stracke et al. (2025) found that AI-specific policies were difficult to find and largely undeveloped.”

• cites Stracke et al., 2025Not matched

  Couldn’t be matched to a paper in Crossref, so it can’t be checked.

Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

Within that percentage, certain types of institutions were more likely to have policies: those that offered doctoral degrees and reported high rates of research activity, as well as baccalaureate colleges with Arts and Sciences programs (40% within that subgroup).

“Within that percentage, certain types of institutions were more likely to have policies: those that offered doctoral degrees and reported high rates of research activity, as well as baccalaureate colleges with Arts and Sciences programs (40% within that subgroup).”

• cites Kim and Wu 2024 sourceNot checked — no full text of the source

Probably not a substantial claim: may not pass S3 (load-bearing).

Seneca College's policy tone fluctuates between educational and legalistic, implicitly reinforcing honesty, fairness, and responsibility as core academic integrity values even when not explicitly named.

“Its tone fluctuates between educational and legalistic, implicitly reinforcing values such as honesty, fairness, and responsibility, which are core to academic integrity, even when these are not explicitly named.”

Probably not a substantial claim: may not pass S3 (load-bearing).

The statement in Seneca's policy operates as a supportive safeguard in the context of Bretag et al.'s framework, identifying and articulating specific ethical behaviours to mitigate offence and promote preventative education.

“This statement operates as a supportive safeguard in the context of Bretag et al.'s (2011a;2011b) framework, identifying and articulating specific ethical behaviours to mitigate offence and promote preventative education.”

• cites Bretag et al., 2011a;2011bNot matched

  Couldn’t be matched to a paper in Crossref, so it can’t be checked.

• AssumptionRests on an assumption: The adapted Bretag et al. framework is appropriate for analyzing AI-related academic integrity policies.

Probably not a substantial claim: may not pass S3 (load-bearing).

Kim and Wu 2024 reported that 44.5% of higher education institutions they reviewed (n = 101) had AI-specific policies.

“Looking at the United States, Kim and Wu 2024 reported that 44.5% of HE institutions they reviewed (n = 101) had AI-specific policies.”

• cites Kim and Wu 2024 sourceNot checked — no full text of the source

Probably not a substantial claim: may not pass S3 (load-bearing).

Both researchers independently conducted pilot coding of a subset of policies (n=4) to enhance reliability and minimize bias.

“To enhance the reliability of our analysis and minimize bias, both researchers independently conducted pilot coding of a subset of policies (n = 4).”

Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims. The quote only partly states this claim.

The results of the pilot analyses were compared, and discrepancies were discussed until consensus was reached.

“The results of the analyses were compared, and discrepancies were discussed until consensus was reached, enabling consistent application of the coding framework.”

Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

The University of Guelph's document is outside the scope of policy and excluded from further analysis.

“As such, we regard this document to be outside of the scope of policy and exclude it from further analysis.”

Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

Where AI policies exist, they are organized at the institutional level, through consortia, or via government agencies.

“Where policies do exist, they are either organized at the institutional level, through consortia, or via government agencies.”

• cites Stracke et al., 2025Not matched

  Couldn’t be matched to a paper in Crossref, so it can’t be checked.

Probably not a substantial claim: may not pass S3 (load-bearing).

Seneca College's policy's tone fluctuates between educational and legalistic, implicitly reinforcing values such as honesty, fairness, and responsibility, which are core to academic integrity, even when these are not explicitly named.

“Its tone fluctuates between educational and legalistic, implicitly reinforcing values such as honesty, fairness, and responsibility, which are core to academic integrity, even when these are not explicitly named.”

Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

Coding decisions were reviewed collaboratively, and discrepancies or inconsistencies were resolved through consensus.

“Coding decisions were reviewed collaboratively, and any discrepancies or inconsistencies were resolved through consensus.”

Classified as probably not a substantial claim.

Bretag et al.'s framework for academic integrity policies consists of five categories: Access, Approach, Responsibility, Detail, and Support.

“The framework outlines five key categories for policy review: Access, Approach, Responsibility, Detail, and Support.”

• cites Bretag et al. (2011a;2011b)Not matched

  Couldn’t be matched to a paper in Crossref, so it can’t be checked.

Classified as probably not a substantial claim. May combine two claims.

The Oxford Government AI Readiness Index provides a general framework with three pillars-technology sector, data and infrastructure, and government-but lacks transparency, may contain biases, and does not address education-specific needs.

“They further noted that while the Oxford Government AI Readiness Index provides a general framework with three pillars-technology sector, data and infrastructure, and government-it lacks transparency, may contain biases, and does not address educationspecific needs.”

• cites Stracke et al., 2025Not matched

  Couldn’t be matched to a paper in Crossref, so it can’t be checked.

Classified as probably not a substantial claim.

The University of Guelph's policy functions more as a philosophical statement than a formal policy.

“Upon review, the University of Guelph's policy functions more as a philosophical statement than a formal policy.”

Probably not a substantial claim: may not pass S2 (checkable).

Enhancing AI literacy among higher education stakeholders is instrumental for developing informed, appropriate policy.

“Enhancing AI literacy among HE stakeholders is therefore instrumental for developing informed, appropriate policy.”

• AssumptionRests on an assumption: Without sufficient knowledge among all users (students, instructors, and leadership), AI policy will either lag behind practice or be overly restrictive and hamper student learning.
• AssumptionRests on an assumption: Sufficient knowledge among users is essential to ensure AI policy is effective and meaningful to the institution.

Probably not a substantial claim: may not pass S2 (checkable), S3 (load-bearing).

The findings reported here may quickly become outdated.

“Therefore, the findings reported here may quickly become outdated.”

Classified as probably not a substantial claim.

Human-centric behaviours, e.g., critical thinking, creativity, ethical judgement, are shaped by complex factors, including organizational culture, leadership, and structure.

“Such behaviours are shaped by complex factors, including organizational culture, leadership, and structure (Bratton, 2021;Kumar et al., 2022).”

• cites Bratton, 2021Not matched

  Couldn’t be matched to a paper in Crossref, so it can’t be checked.

• cites Kumar et al., 2022Not matched

  Couldn’t be matched to a paper in Crossref, so it can’t be checked.

Probably not a substantial claim: may not pass S3 (load-bearing).

This fragmentation can hinder the development of institution-wide policies as units may have divergent views on the use of AI in teaching, learning, and research.

“This fragmentation can hinder the development of institution-wide policies as units may have divergent views on the use of AI in teaching, learning, and research.”

• AssumptionRests on an assumption: Some institutions operate under decentralized governance structures, where faculties and departments maintain significant autonomy.

May combine two claims.

The challenge of regulating GenAI may relate to conflicting interests and institutional gridlocks at the heart of regulatory politics, compounded by the novel features of GenAI as a technology capable of processing trillions of parameters.

“As Judge et al. (2025) suggest, this challenge may relate to “conflicting interests and institutional gridlocks at the heart of regulatory politics” (p.86), compounded by the novel features of GenAI as a technology capable of processing trillions of parameters.”

• cites Judge et al., 2025Not matched

  Couldn’t be matched to a paper in Crossref, so it can’t be checked.

Probably not a substantial claim: may not pass S2 (checkable). May combine two claims.

Promoting unreliable detection procedures risks undermining stakeholder trust and may inadvertently penalize legitimate student work.

“Promoting unreliable detection procedures risks undermining stakeholder trust and may inadvertently penalize legitimate student work.”

• AssumptionRests on an assumption: Technology-enhanced detection systems have limitations such as high false-positive rates and variability between tools.

Probably not a substantial claim: may not pass S2 (checkable). May combine two claims.

An exemplary academic integrity policy should adopt an educative and developmental stance, clearly articulating expectations and responsibilities for all members of the learning community.

“An exemplary academic integrity policy should adopt an educative and developmental stance, clearly articulating expectations and responsibilities for all members of the learning community (Bretag et al., 2011a(Bretag et al., , 2011b)).”

• cites Bretag et al., 2011aNot matched

  Couldn’t be matched to a paper in Crossref, so it can’t be checked.

• cites Bretag et al., 2011bNot matched

  Couldn’t be matched to a paper in Crossref, so it can’t be checked.

May combine two claims.

Aside from Québec's work, the regulatory vacuum leaves provinces' institutions to independently determine their approach, leading to varied practices and limited accountability.

“Aside from Québec's work, this regulatory vacuum leaves provinces' institutions to independently determine their approach, leading to varied practices and limited accountability (Sebesta & Davis, 2023).”

• cites Sebesta & Davis, 2023Not matched

  Couldn’t be matched to a paper in Crossref, so it can’t be checked.

Probably not a substantial claim: may not pass S2 (checkable). May combine two claims.

Investing in good policy, strong educational support, and access to AI applications can help institutions mitigate risks and harness AI's transformative potential to enrich teaching and learning.

“Investing in good policy, strong educational support, and access to AI applications can help institutions mitigate risks and harness AI's transformative potential to enrich teaching and learning.”

• AssumptionRests on an assumption: Integrating AI into curricula supports the development of essential fifth industrial revolution skills, including critical thinking, digital literacy, and ethical reasoning, preparing learners for a workforce increasingly shaped by AI technologies.

Probably not a substantial claim: may not pass S2 (checkable). May combine two claims.

Policies related to academic integrity and AI are in flux, and institutional responses are likely to evolve as these technologies mature and pedagogical practices are given adequate space to respond.

“Policies related to academic integrity and AI are in flux, and institutional responses are likely to evolve as these technologies mature and pedagogical practices are given adequate space to respond.”

May combine two claims.

The federal government plays a limited, indirect role in higher education by supporting initiatives such as research funding, student financial aid, and Indigenous education, and by overseeing Canadian Military Colleges.

“The federal government plays a limited, indirect role by supporting initiatives such as research funding, student financial aid, and Indigenous education, and by overseeing Canadian Military Colleges.”

• cites Eastman et al., 2018Not matched

  Couldn’t be matched to a paper in Crossref, so it can’t be checked.

Probably not a substantial claim: may not pass S2 (checkable), S3 (load-bearing), S4 (based).

The reactive integration of AI tools may be a residual effect from the abrupt and sudden changes to higher education post-pandemic that created disruption and chaos within learning institutions.

“This may be a residual effect from the abrupt and sudden changes to HE post-pandemic that created disruption and chaos within learning institutions.”

Probably not a substantial claim: may not pass S2 (checkable), S3 (load-bearing).

Given the rapid pace of AI development, the findings of this study should be interpreted as a temporal snapshot of the policy landscape, rather than a fixed account.

“Given the rapid pace of AI development, the findings of this study should be interpreted as a temporal snapshot of the policy landscape, rather than a fixed account.”

Classified as probably not a substantial claim.

The updates to existing academic integrity policies reflect a more cautious or incremental approach to regulation of AI in Ontario higher education.

“perhaps reflecting a more cautious or incremental approach to regulation in this area.”

May combine two claims.

Each province and territory governs its own higher education system, shaping policies related to governance, funding, and quality assurance.

“Each province and territory governs its own HE system, shaping policies related to governance, funding, and quality assurance.”

Probably not a substantial claim: may not pass S2 (checkable). The quote only partly states this claim.

Policy will help shape the path to balancing technical proficiency with ethical and sociocultural awareness.

“Policy will help shape that path.”

• AssumptionRests on an assumption: Integrating AI into curricula supports the development of essential fifth industrial revolution skills, including critical thinking, digital literacy, and ethical reasoning, preparing learners for a workforce increasingly shaped by AI technologies.

May combine two claims.

Without transparent and reliable procedures for detection, institutions rely on informal or case-by-case judgment, potentially leading to arbitrary enforcement, inequitable treatment, and diminished trust in integrity processes.

“Without transparent and reliable procedures for detection, institutions rely on informal or case-by-case judgment, potentially leading to arbitrary enforcement, inequitable treatment, and diminished trust in integrity processes.”

Probably not a substantial claim: may not pass S3 (load-bearing).

Effective academic integrity policies must be accessible, i.e., easy to locate, read, and understand.

“Bretag et al. (2011aBretag et al. ( , 2011b) ) emphasize that effective academic integrity policies must be accessible, i.e., easy to locate, read, and understand.”

• cites Bretag et al., 2011aNot matched

  Couldn’t be matched to a paper in Crossref, so it can’t be checked.

• cites Bretag et al., 2011bNot matched

  Couldn’t be matched to a paper in Crossref, so it can’t be checked.

Probably not a substantial claim: may not pass S2 (checkable).

A structured approach to AI education may help balance technical proficiency with ethical and sociocultural awareness that can empower both educators and learners to engage with AI responsibly and effectively.

“A structured approach to AI education may help balance technical proficiency with ethical and sociocultural awareness that can empower both educators and learners to engage with AI responsibly and effectively.”

• AssumptionRests on an assumption: Integrating AI into curricula supports the development of essential fifth industrial revolution skills, including critical thinking, digital literacy, and ethical reasoning, preparing learners for a workforce increasingly shaped by AI technologies.

May combine two claims.

Depending on socioeconomic factors, students may have less access to AI tools or may be disproportionately affected by automated decision-making systems used in higher education processes, e.g., admissions and grading.

“Depending on socioeconomic factors, students may have less access to AI tools or may be disproportionately affected by automated decisionmaking systems used in HE processes, e.g., admissions and grading (Hoernig et al., 2024).”

• cites Hoernig et al., 2024Not matched

  Couldn’t be matched to a paper in Crossref, so it can’t be checked.

May combine two claims.

The ambiguity in policies may risk unintentional misconduct and/or uneven and inconsistent enforcement across the institution.

“This ambiguity may risk unintentional misconduct and/or uneven and inconsistent enforcement across the institution.”

Probably not a substantial claim: may not pass S2 (checkable). May combine two claims.

Such variability may hinder comprehension, create a confrontational impression, and undermine the policy's effectiveness.

“Such variability may hinder comprehension, create a confrontational impression, and undermine the policy's effectiveness.”

Classified as probably not a substantial claim. May combine two claims.

This study is limited to the examination of existing institutional policies in Ontario, Canada and does not include the perspectives of key stakeholders, such as students, faculty, or administrators.

“Additionally, this study is limited to the examination of existing institutional policies in Ontario, Canada and does not include the perspectives of key stakeholders, such as students, faculty, or administrators.”