Academic malpractice

1. The aim of this policy is to ensure that there are no actions taken to undermine integrity of the assessment process and certification of the qualifications through deliberate or accidental acts of malpractice or maladministration

1. This policy applies to NSA employees, learners and third-parties

1. NSA learners and employees have a responsibility to read, understand and comply with this policy

1. Malpractice is defined as any deliberate activity, neglect, default or other practice that compromises the integrity of the assessment process, and/or the validity of certificates. Malpractice can take several forms, and is not just limited to deliberate acts. Malpractice may include:
   • Plagiarism, or copying others’ work
   • Collusion between learners
   • Sharing or making an assessment brief publicly available
   • Use of AI in producing assessment submissions (see Appendix 1: AI Misuse Policy)
2. Maladministration is any activity, neglect, default or other practice that results in the centre or candidate not complying with the specified requirements for delivery of the qualifications.

1. NSA does not tolerate actions or attempted actions of malpractice
2. Malpractice or maladministration may be identified in a number of ways, by NSA employees, learners, or members of the public
3. NSA encourages the reporting of malpractice or maladministration
4. Reports should be made in a timely manner by emailing malpractice@nationalsafetyacademy.co.uk
5. All allegations of malpractice and maladministration will be investigated in order to protect the integrity of our qualifications
6. Anonymous reports of malpractice or maladministration will also be acted upon only if there is supporting evidence, or if the nature of the report warrants it
7. Where malpractice or maladministration is deemed to have taken place by an independent Malpractice Panel, sanctions will be imposed proportionate to the severity of the offence
8. Any decision made by the Malpractice Panel is final. However, candidates believing they have grounds to appeal as outlined in the Appeals Policy can do so within 10 days of notification of the investigation outcome.
9. Mitigating or aggravating factors can be taken into account, yet the following will not be considered:
   • that the individual did not intend to cheat;
   • that the individual has an otherwise unblemished academic record;
   • that the individual could face discipline at work or lose their job;
   • that the individual regrets their actions.
10. Investigation reports and outcomes will be retained by NSA for periods of up to five years
11. Where a learner is employer-sponsored and has committed malpractice, we reserve the right to disclose the offence and related sanctions to that employer

1. This policy will be reviewed annually and in line with any significant changes

Policy Statement

1. To protect the integrity, authenticity, and fairness of learner assessments, the National Safety Academy (NSA) strictly prohibits the use of artificial intelligence (AI) tools in the creation, completion, or submission of assessment content unless explicitly approved as part of a reasonable adjustment.
2. Learner submissions generated using AI – either in part or in whole – threaten the ability of the marker to apply the assessment criteria to make a genuine assessment of the learner’s understanding and application of the course content. This has a severe impact on the validity and reliability of the assessment process. Therefore, where AI has been used in any capacity, it is the view of NSA that the submission is null and void as there is no way of reliably differentiating the original thought of the learner from information generated by AI.
3. For this reason, NSA enforces a zero-tolerance policy regarding AI usage, unless approved in a minor capacity as part of a Reasonable Adjustment. Learners found to have engaged in AI-assisted malpractice may face sanctions, up to and including disqualification from their qualification.

Definitions

1. For the purposes of this policy, Artificial Intelligence (AI) refers to technologies or tools capable of generating, modifying, or completing text, audio, or visual content based on prompts or data inputs. This includes, but is not limited to:
   • Large language models (e.g., ChatGPT, Bard, Claude, Gemini, Deepseek)
   • Content generators (e.g., Grammarly AI, Jasper, Quillbot)
   • AI humaniser tools (undetectable AI, StealthGPT, Scribbr)
   • Translation engines that modify meaning beyond simple language conversion
   • Speech-to-text tools (use may be permitted with prior approval)