Written by Hey all – Take a look if you have time. I am rather excited about this thoughtful approach to AI. Interested in your opinions as well…
San José State University welcomes its first ever Artificial Intelligence (AI) Librarian, Sharesly Rodriguez, ‘18 MLIS. As AI applications in academia are increasing, Rodriguez joins as one of the first dedicated faculty AI librarians at a university to lead the integration and development of AI technologies for an academic library.
“The Dr. Martin Luther King, Jr. Library at San José State is more than a hub for knowledge — it’s a gateway to the future,” says SJSU President Cynthia Teniente-Matson. “With the addition of our first AI librarian, we are ensuring that students, faculty, staff and the broader community have access to the tools and expertise needed to navigate the evolving AI landscape today and in the future. Our investment in roles like this demonstrates how SJSU truly is at the epicenter of the future. We look forward to the research, collaboration and conversations that will empower our students to apply AI in ways that are innovative, responsible and ethical — ensuring that this technology serves communities and drives progress in meaningful ways.”
@michael – Thank you for sharing! Absolutely fascinating times we live in. I work in legal and in the private sector many law librarians / knowledge management departments have been at the forefront of piloting and implementing AI research tools for attorneys. There have been issues, though, with attorneys not cite-checking / properly vetting their own AI generated work, and some who have submitted briefs incorporating hallucinated AI-generated cases have been called out or even monetarily fined by courts, and then they end up the news, yikes.
@cecinsd I thought this hire was a big deal. Thanks for the insights into what is happening in the legal sector.
Thanks for sharing, Candice!
@georgerothrock – of course! I recently attended a conference hosted by a local law librarian association, and the theme was AI in legal, so I am happy to share my key take-aways with anyone interested, but noting that I still have much to learn and I think I am still digesting all of the content covered. I’m also glad SJSU hired for this role, and you raise valid concerns, especially regarding the environmental impact of AI, so thank you for sharing too!
@cecinsd please share your key takeaways, and I will actually boost them up onto the course blog. I think these are good things to be thinking about as you all move through 200 specifically and through the program.
@michael – sure thing, summary below:
The Southern California Association of Law Libraries (SCALL) is a local chapter of the American Association of Law Libraries (AALL). In February 2025, SCALL hosted its 53rd Institute in San Diego, California. The theme was AI: Beyond the Hype and the program’s speakers represented private, academic, and public (including court) law libraries. As a student attendee, here are my key takeaways:
• AI may be used as a tool or starting point for research, but it is not a substitute for work product. In legal, attorneys must review AI generated research for relevance and applicability (especially given professional and ethical duties governing their state bar licenses), but for research professionals in general, themes of competency, proficiency, and due diligence were emphasized.
• When piloting and implementing AI research applications, users (ideally the same group for all applications being tested) must be engaged in the process from start to finish. Users’ encounters with different applications will vary, so what looks good on paper may be a different experience and vice-versa. Applications that are not practical or realistic for users’ daily research needs are unhelpful and could be a waste of valuable resources ($$$).
• After you graduate, depending on the organization you work for, be prepared to not have access to AI research applications. This point was especially stressed by those working in public law libraries, court libraries, and law school libraries.
Bonus – annually there’s a popular presentation, “30 Sites in 30 Minutes” tailored to the institute’s annual theme, and the below were my top five favorites:
• https://www.wired.com/story/ai-copyright-case-tracker/ – tracks legal disputes between content providers and AI companies. How courts rule “could make, break, or reshape the information ecosystem and the entire AI industry—and in doing so, impact just about everyone across the internet.”
• https://www.ailawlibrarians.com/ – All things AI law librarian-ish, generative ai, and legal research/education/technology.
• https://www.britannica.com/chatbot – Chatbot answers created from Britannica articles using AI. (Note the disclaimer: This is a beta feature. AI answers may contain errors. Please verify important information using Britannica articles.)
• https://copyleaks.com/ai-content-detector – AI Detector. Interesting case study here: https://copyleaks.com/case-studies/oakland-university with an endnote, “To continue addressing the rising concerns around generative AI use within academia, Copyleaks is creating, with support from Oakland and other academic institutions, a faculty guide that helps address common questions and concerns, including how to speak with students about AI use, how to utilize the data provided from the AI content detection, how to implement it into the classroom, and more.”
• https://www.oldmapsonline.org/en – browse historical places and search for old maps with timelines of events.
First I’m am very glad the school has invested in a real person to be engaged in this process and Sharesly Rodriguez’s seems to be a very great addition to the King Library and the school. I have hope that with human oversight what seems to be an inevitable progression can be managed in a human-first way.
I’m not an expert, but as I understand it, large language models are not “intelligent” in any sense of the word – they all rely on an approach that looks at a huge body of language usage (some claim they have trained on the entire internet, at this point) and use a system to mathematically weight the occurrence and order of words – so a chapGPT “answer” is a mathematical average of the responses to the prompt that can be found the “wild.” It can’t tell you, for instance, “how many “r”s are in “strawberry”.
My main worry, and as the response above alludes – is that the allure of “convenience” and “ease” will draw people in and we’ll attempt to use it as a replacement for actual work. Also, that the draw to the eventual goal of eliminating “work” and therefore “workers” will pull in companies. The danger of this was summed up by a friend of mine when they asked “if a robot does my deadlifts, do I get stronger?” Is this a technology in search of a problem?
The question of environmental impact is a very real one – the amount of power this technology takes is enough to “boil an ocean” and while we are still reliant upon dirty energy generation, I’m not convinced its a great tradeoff.
I think back to the changes we have seen in google search – from a system designed for finding information to a system that encourages commercial activity by providing opportunities to consume along the axis of your search, and provide that information at the top. Search “Taylor Swift” and be presented with tour dates and a link to Ticketmaster. So I worry about the LLM technology being hijacked by capitalist and commercial forces before too long.
Apologies – I’ve provided no references for the above thoughts, but I wanted to weigh in and of course there are strong arguments for the positive use of the technology, so I don’t want to come across as unaware of them.
All the best!