You have /5 articles left.
Sign up for a free account or log in.
Nearly three-quarters of university librarians say there’s an urgent need to address artificial intelligence’s ethical and privacy concerns, a survey finds.
Roughly half the librarians surveyed said they had a “moderate” understanding of AI concepts and principles, according to the study released Friday. About one in five said they had a slightly below moderate understanding, and roughly the same amount had a slightly above moderate understanding. Only 3 percent of respondents said they had a “very high” understanding.
The study, conducted in May 2023 by Leo Lo, president-elect of the Association of College and Research Libraries, had 605 respondents who completed the survey. Of those, 45 percent worked in research institutions and 30 percent in institutions with undergraduate and graduate programming.
The need for ethics training was one of the biggest insights, Lo said. Nearly 75 percent of respondents said they agreed or strongly agreed there is an urgent need to address those AI issues. The major worries included violations of privacy and misuse of data, such as generating false citations.
Many librarians are fielding questions and concerns from faculty members about those issues, with some making AI guides and launching pilot projects to address anxieties.
Among those who completed the survey, their most widely used AI tools were chat bots and text- or data-mining tools, the survey found. But about two in five respondents said they had never used a generative AI tool, and of the 28 percent of those who said they had, they used them less than once a month.
Of those surveyed, those involved with administration, management, library instruction and information literacy had a higher understanding of AI, according to the survey. “This could be due to a variety of reasons, such as these [senior] roles potentially requiring a more in-depth understanding of AI and its applications, or these individuals having more access to resources and opportunities for such training,” the report said. “On the other hand, support staff or administrative personnel are less likely to have participated in such programs, which could be due to less perceived need or fewer opportunities for training.”
The biggest concern among the respondents was troubleshooting AI tools and applications, with nearly 70 percent having “low confidence.”
Based on the findings, Lo suggested that librarians focus on seven AI competencies: understanding AI systems’ capabilities and limitations; identifying and evaluating AI use cases; utilizing AI tools effectively and appropriately; critically assessing AI’s quality, biases and ethics; engaging in informed AI discussions and collaborations; recognizing data privacy and security issues; and anticipating AI’s impacts on library stakeholders.
“The rapidly evolving nature of AI and its applications in libraries means that our findings provide a snapshot at a specific point in time,” the report states. “As AI continues to advance and integrate more deeply into academic libraries, the landscape of AI literacy among library employees is likely to shift, necessitating ongoing research.”
