Navigating AI Resources at Pitt
By SUSAN JONES
With the rapid advancements in artificial intelligence (AI), understanding how to effectively use these tools in academic settings has never been more crucial. The University of Pittsburgh is making significant strides in integrating AI technologies to enhance research and learning. Here’s an overview of the resources and insights shared during recent presentations.
Upcoming Events
An online session titled “Navigating Pitt’s AI Resources for Research and Learning” will take place from 7 to 8 p.m. on February 23. Register here.
Additionally, a workshop called “Creating a Personal Research & Learning Assistant: Writing Effective Prompts” is scheduled for 7-8 p.m. on February 25, also online. Register here.
For more guidance on the acceptable use of generative AI tools, visit the Pitt Digital website.
Insights from the Presentation
Luke Ferdinand, the Technology Support Services Manager for the University Library System, opened his presentation on “Navigating Pitt’s AI Resources for Research and Learning” with a compelling observation: the pace of technological change is evolving rapidly.
“The slides we are sharing with you today have been revised multiple times since their initial presentation over the summer,” Ferdinand noted.
Key Points About AI
- While AI education is widely available, high-quality resources are harder to come by.
- Data privacy and security present significant challenges as AI systems often gather extensive information. “If you’re a programmer or a writer here at the University, consider what data you’re sharing with these companies,” he emphasized.
- AI is a valuable asset, but its outputs must be validated, as they can be inaccurate.
Currently, many AI tools are accessible at Pitt. One notable example is Google Gemini, which offers comprehensive security features for its users. “Google Gemini is designed to protect your data while utilizing it to enhance your experience,” Ferdinand explained.
He further elaborated that most generative AI tools primarily operate as answer generators. “These systems analyze vast databases of information to predict likely responses based on your queries. However, they can also generate incorrect information or ‘hallucinate’ responses when faced with unrecognized queries,” he added.
Traditional AI applications in academia—such as machine learning and neural networks—are being employed to streamline labor costs and analyze complex data. For example, the library is using machine learning techniques to decipher the often-challenging handwriting of playwright August Wilson, greatly expediting the transcription process.
AI Tools Available at Pitt
Ferdinand discussed an array of free AI tools that faculty, staff, and students can access, with the exception of PittGPT, which is available only to faculty and staff. Users can start by logging in with their Pitt credentials on the respective websites.
- Google Gemini: Integrates with your pitt.edu Google Drive, learning from your data to deliver relevant insights. It’s recommended to log in with your Pitt account to enhance security.
- Anthropic Claude for Education: This tool emphasizes ethical AI usage, assisting in strategic decision-making, brainstorming, and coding tasks.
- NotebookLM: A Google tool that organizes documents and web sources, creating customized learning materials, such as flashcards and summaries.
- Microsoft Copilot Chat: Part of Pitt’s Microsoft Office 365, this tool quickly synthesizes information and automates tedious tasks, promoting efficiency in teaching and research.
- PittGPT: This unique tool connects university data directly to ChatGPT while protecting sensitive information.
Additional Free Tools
- Zoom AI Companion: Facilitates creating summaries and translations from meetings.
- Adobe Creative Cloud: Offers a suite of AI tools for various applications, including document conversion, image editing, and video production, accessible to faculty and students.
Ferdinand encouraged anyone interested in additional AI tools to reach out to Pitt Digital for a security risk assessment of potential vendors. He also underscored the importance of adhering to university policies regarding acceptable AI use.
“Generative AI tools are transformative and are already demonstrating their utility. However, they must be used with caution,” he concluded. “When applied thoughtfully, they can significantly enhance your academic experience at the University of Pittsburgh.”
Susan Jones is the editor of the University Times. Reach her at suejones@pitt.edu or 724-244-4042.
Have a story idea or news to share? Share it with the University Times.
Follow the University Times on Facebook.