ACRL Artificial Intelligence (AI) Interest Group

 View Only
last person joined: 23 hours ago 

To provide a forum for discussing the impact of AI on libraries and related topics, facilitating the exchange of ideas, best practices, and collaborative initiatives among library professionals.
Community members can post as a new Discussion or email ALA-ACRL-AI-IG@ConnectedCommunity.org.
Before you post: please note job postings are prohibited on ALA Connect. Please see the Code of Conduct for more information.

"Navigating the Jagged Technological Frontier" - good article on how AI contributes to knowledge tasks

  • 1.  "Navigating the Jagged Technological Frontier" - good article on how AI contributes to knowledge tasks

    Posted Dec 06, 2024 09:28 AM

    Hi folks!  I wanted to share this article that was recently recommended to our AI Working Group by the Harvard CIO - it's titled "Navigating the Jagged Technological Frontier: Field Experimental Evidence of the Effects of AI on Knowledge Worker Productivity and Quality" (HBS 2023).  The "jagged frontier" in the research describes how AI improves or degrades the quality of human work, for various types of knowledge tasks.  In about half of the tasks tested, the use of AI enhanced the quality of the results, for the other half, it degraded the quality of the results.  The paper goes into great depth on the types of tasks that fall into each category, but in a very-reductive nutshell:

    • Tasks that enhance the quality of human results ("Inside the AI frontier"): creative ideation (e.g., brainstorming innovative product ideas), market analysis), writing tasks (e.g., drafting marketing materials or press releases), persuasive communication (e.g., employee memos).
    • Tasks that degrade the quality of human results ("Outside the AI frontier"): business problem-solving requiring subtle interpretations of quantitative and qualitative data, strategic recommendations where AI struggled to synthesize multiple complex inputs, some types of creative tasks - AI-generated responses showed reduced variability and diversity (and greater homogeneity).  

    Curious to hear whether these results resonate with you in your experience, or whether they offer up any new information for our interpretation of what AI is "good" and "not good" at?

    Warmly,

    Heather



    ------------------------------
    Heather Sardis
    Associate Director for Technology and Strategic Planning
    Massachusetts Institute of Technology
    ------------------------------