Last week we hosted a webinar with Jessica Roberson and Jeremiah Colonna-Romano from the University of Alabama on how they were experimenting with OpenAI's GPT models for their archival metadata workflows. In the Q&A Jessica and Jeremiah were asked: Did you think of any privacy or copyright considerations when you were putting this content into ChatGPT? I had to chime … [Read more...] about Privacy And Copyright Considerations Using GPT Models
How LLMs Work & A Handwritten Text Recognition Sandbox
I’ve got two useful AI tidbits for you this month: The first is the best explanation of how large language models work. This is a pretty opaque topic that few attempt to understand, but I think it’s important to grasp – even in metaphor – how systems like ChatGPT work. When you understand how they work, your intuition for the types of problems they can – or can’t – … [Read more...] about How LLMs Work & A Handwritten Text Recognition Sandbox
Guiding Transcribers: Addressing Racist Material in Crowdsourced Archival Projects
In honor of Black History Month, we asked some of the institutions who run projects uncovering the names and histories of African Americans on FromThePage the following question, "What do you tell volunteers who encounter offensive or racist material in your projects?" Sonya Coleman at Library of Virginia summed it up with a meme: From … [Read more...] about Guiding Transcribers: Addressing Racist Material in Crowdsourced Archival Projects
10 Ways AI Will Change Archives
We recently hosted a webinar entitled "10 Ways AI Will Change Archives," and today, we're sharing the insights from that presentation with you. AI is a transformative force, and often scary. Today, though, we’re sharing specific examples of using AI for good – allowing more people to find and interact with archival resources. 1. Improving Accessibility In the … [Read more...] about 10 Ways AI Will Change Archives
ChatGPT, Microsoft, and the Development of Guardrails
When we were running experiments with ChatGPT this summer, we kept running into guardrails; things that ChatGPT wouldn’t say. None of us wants a chatbot to be racist or sexist, but those of us studying history might want it to be able to speak from the perspective of a 19th century slave holder. (And it won’t.) At the time, we chalked this limitation up to “OpenAI doesn't want … [Read more...] about ChatGPT, Microsoft, and the Development of Guardrails