AI Agent Operational Lift for Tulsa City-County Library in the United States

Why AI matters at this scale

Tulsa City-County Library (TCCL) is a mid-sized public library system serving a diverse county of over 650,000 residents through 24 branches and a central library. With 201–500 employees and an annual budget around $30 million, it sits at a scale where AI can deliver meaningful efficiency gains without the complexity of a large enterprise. Libraries are trusted community anchors, and TCCL’s mission to provide free access to information and lifelong learning aligns naturally with AI’s potential to personalize services and automate routine tasks.

1. What the company does

TCCL offers traditional lending of books, e-books, and media, along with public computers, Wi-Fi, research databases, and a rich calendar of educational and cultural programs. It also provides outreach through bookmobiles and partnerships with schools and workforce agencies. The library system is funded by ad valorem taxes and governed by a citizen board, making it accountable to the public and focused on equitable service.

2. Why AI matters at this size and sector

At 200–500 employees, TCCL has enough scale to justify AI investments but limited IT staff. AI can amplify the impact of existing personnel by automating repetitive tasks like cataloging, answering common patron questions, and analyzing circulation data. Public libraries face growing demand for digital services and personalized experiences, yet they must operate within tight budgets. AI offers a force multiplier—improving patron engagement and operational efficiency without proportional cost increases. Moreover, libraries are ideal testbeds for ethical AI deployment because they prioritize privacy, equity, and community trust.

3. Three concrete AI opportunities with ROI framing

Personalized reading recommendations – A machine learning model trained on borrowing histories and community preferences can suggest titles patrons are likely to enjoy. This increases circulation, reduces time spent browsing, and boosts patron satisfaction. ROI: a 5–10% lift in checkout rates could justify the development cost within one year.

Automated cataloging and metadata generation – Natural language processing can extract subjects, genres, and summaries from new acquisitions, cutting manual cataloging time by up to 50%. This frees librarians to focus on programming and patron interaction. ROI: staff hours saved translate directly to cost avoidance, potentially $50,000–$100,000 annually.

AI-powered virtual assistant – A chatbot on the library’s website and app can handle routine queries about hours, events, and account issues 24/7. It reduces phone and in-person interruptions, allowing staff to address complex needs. ROI: improved service levels and reduced wait times, with minimal ongoing cost after initial implementation.

4. Deployment risks specific to this size band

Mid-sized public libraries face unique risks: limited in-house AI expertise, reliance on legacy integrated library systems, and strict data privacy requirements. Patron data is sensitive, so any AI initiative must comply with library confidentiality policies and state laws. Budget cycles are annual and political, making multi-year AI projects challenging. Staff resistance to automation is possible, so change management and upskilling are critical. Finally, ensuring equitable access—so that AI benefits all patrons, not just the tech-savvy—must be a core design principle. Starting with low-risk, high-visibility pilots like a chatbot or recommendation engine can build momentum and trust.