One of the main challenges which academic librarians face when trying to develop programs and services that support AI literacy is the wide array of stances taken by institutions, and individual faculty members, when it comes to teaching with and about AI tools. These dimensions are not only applicable to students, but also to faculty, who also need help and guidance navigating the new technologies. One aspect which becomes central to this conversation is promoting the ethical use of AI tools in the academic environment. Although the topic has gathered considerable attention in recent conversations, these remain fragmented and divisive. This is especially evident when considering what constitutes adequate use of AI in academia. Some advocate for the use of AI tools and even claim that we have entered a new era in which plagiarism is a non-issue since it is not possible to separate human everyday life from technology. While others argue that AI tools are a threat to learning and that people’s capacity for critical thinking and higher order thinking is compromised unless we stop using these technologies.

While these conversations might appear to drive more divide, it is essential for library and information (LIS) professionals to understand the current reality: AI tools are here to stay. The role of LIS professionals now is to adapt, and help others use these by guiding the patrons’ awareness and understanding of the tools’ origins, capabilities, and shortcomings. LIS professionals cannot force people to use or not use AI tools. But they can help educate people so that they can make the most of the available technology within their level of comfort and within ethical boundaries.

The Digital Education Council (DEC), a recently created body aiming to drive sustainable AI innovation and adoption, has created an AI framework. This framework is made up of  five dimensions intended to guide humans in the development of knowledge about AI tools with the goal of promoting appropriate use and successful human-AI collaborations. The framework’s five dimensions revolve around:

• Understanding how AI works
• Critically evaluating AI outputs
• Ensuring the ethical and responsible use of AI
• Focusing on human-centric skills
• Applying AI tools to various contexts

The DEC report identifies the ethical dimensions in the use of AI tools as those encompassing principles related to fairness, transparency, accountability, and privacy. All this while also recognising potential risks for bias, discrimination, and misinformation. However, research findings indicate that undergraduate students demonstrate lower awareness of fairness and understanding related to how AI tools work. This is not an indication of a shortcoming in students’ character or knowledge, but one which is inherent to the way in which AI tools are currently being created. For example, AI transparency is an aspect which depends on people having a basic understanding of how AI tools operate. However, the fact remains that AI tools’ inner workings are intentionally vague to most except for a handful of insiders. As exemplified by most company’s reluctance to divulge which specific information has been used to train their large language models other than to indicate that it is a vast amount of data.

These issues represent major obstacles for anyone trying to ethically use AI tools. The tool’s inherent lack of transparency reduces efforts to preserve human autonomy. In AI use autonomy includes aspects related to making decisions of when AI tools are to be or not to be used. Some scholars, including Luciano Floridi, see this as an essential component in ethical use of any technology.

These issues illustrate some of the aspects which librarians must consider when creating programs for promoting AI literacy. Firstly, effective AI literacy education should not only consider students but also faculty. Furthermore, an aspect such as ethical use of AI tools is a much more complex issue and it is strongly influenced by questions related to how the technologies work and how they are created. It then becomes the LIS professional’s job to create awareness of the shortcomings which interrupt the optimal approach of AI literacy education.

Despite the inherent complexity LIS professionals still have plenty to contribute to the effective deployment of AI tools in higher education. Gary Marchionini has pointed out how the skills accumulated throughout the history of the profession are matched to the needs for AI literacy created by the newest technology. These skills include the profession’s human-centered approach to information, its core concern for information integrity and trust, its commitment to a culture of sharing and equitable access, among others.

The current technological landscape which has been triggered by the advent of generative AI is one which is evolving rapidly. Despite this, it is still the LIS professionals’ responsibility to learn to adapt to it, find ways to leverage professional knowledge and skills in this new information landscape, and help patrons successfully navigate the changes. Just like the old adage states, change is the only constant. Generative AI is the latest technology to bring disruptive change to academia. It is not the first, and it certainly will not be the last. Therefore, it is the LIS professional’s responsibility to remain adaptable and keep serving patrons needs.