Streamlining Accreditation Activities With Responsible Use of Artificial Intelligence

Abstract

Many educators have quickly adopted artificial intelligence (AI), leveraging large language models such as ChatGPT, Microsoft Copilot, or Google Gemini to streamline their work and boost efficiency. This technology can assist with brainstorming ideas, refining writing, generating and creating content, comparing documents, analyzing information, and revising educational materials. Additionally, many faculty are designing their teaching, learning, and curricular activities with AI. The National League for Nursing Commission for Nursing Education Accreditation (NLN CNEA), the accreditation division of the NLN, realized that with its widespread popularity, faculty will use AI for their other roles, such as assessment, evaluation, and accreditation responsibilities. While the use of AI can help streamline accreditation activities, it is important to point out that one cannot simply enter program information and have AI create a self-study report that is ready to submit for an accreditation review. Faculty may lack expertise, have limited support, or fear the use of AI, which may transfer to accreditation activities. To avoid problems and address potential challenges, guidelines are needed for the responsible use of AI in accreditation. Thus, we recommend the following basic best practices. First, if you have not already developed internal guidelines for AI use for your institution or program, that will be critical to ensure appropriate and consistent approaches by faculty. You will also want to find out from your accreditors (both institutional and programmatic) and regulatory agencies if AI use is even allowed. Once you have general guidelines to follow for responsible use, consider how AI can assist with your accreditation-related work. For instance, AI can assist with activities like brainstorming ideas for improvement, tracking data trends to identify areas needing attention, analyzing data, presenting and summarizing findings, comparing content to ensure alignment and consistency, and editing reports to improve readability. Regardless of the task, AI must never completely replace human decision-making. Instead, it should be viewed as a tool that, when used in conjunction with human expertise, can enhance the quality of work. Faculty should always carefully and critically review all AI output while using their professional knowledge, expertise, and judgment to evaluate the information generated. This responsible use process ensures the accuracy, verifiability, and defensibility of output. Additionally, remember that program evaluators conducting a site visit will expect faculty to be able to explain information found in a self-study report or interpret data they have collected. Therefore, relying solely on AI for accreditation reporting will not be effective. Program evaluators will quickly uncover this lack of understanding, which could put a program’s accreditation at risk. To help with the accuracy and relevance of AI-generated output, a prompt-engineering framework can be used to assist in developing specific information tailored to the audience’s needs with all output checked and revised as necessary. In addition to the human review elements discussed earlier, it is crucial to ensure that the content produced is unbiased. Since AI models generate responses from historical data, discriminatory or biased results may emerge from the algorithm. Diligent human review is a key part of the responsible use of AI with accreditation activities. Meeting minutes may be a major source of accreditation-related evidence, and some programs may use AI to record meetings and then rely on the AI-generated summary for their records. However, an AI model may generate a meeting summary that misrepresents the main points discussed, requiring human review to correct inaccuracies. It is also important to inform participants that AI is used to record meetings. Faculty should consider how to record personal or sensitive information that may be discussed in the meeting and how to manage it in accordance with institutional and programmatic policies. Important considerations include using password-protected technology, restricting access to confidential files, and adhering to the Family Educational Rights and Privacy Act. Security, confidentiality, and privacy are significant concerns with responsible AI use for accreditation activities. Faculty must follow copyright laws and not upload proprietary or protected information into an open-source system. If AI is used to review student information, student names and other personal identifiers should be removed, or use closed-system AI models that do not share information entered with those external to the system. In conclusion, as faculty consider the responsible use of AI for accreditation-related activities, they should follow guidelines and policies, practice diligent human review, utilize a prompt engineering framework, and protect security, confidentiality, and privacy. As AI technology continues to evolve, educators must engage in ongoing professional development to use AI responsibly for program accreditation-related activities.

Ähnliche Arbeiten