BEGIN:VCALENDAR
VERSION:2.0
PRODID:icalendar-ruby
CALSCALE:GREGORIAN
X-WR-CALNAME:Agentic AI might be changing everything again: What an LLM wit
 h tools can actually do for information work\, and things to be cautious a
 bout
X-WR-TIMEZONE:Pacific Time (US & Canada)
BEGIN:VEVENT
DTSTAMP:20260711T051621Z
UID:tag:localist.com\,2008:EventInstance_53252720431280
DTSTART:20260710T160000Z
DTEND:20260710T170000Z
DESCRIPTION:Join us for the first presentation in a three-part summer speak
 er series\, "Critical AI Literacy for Information Work: Agentic AI"\n\nReg
 istration Required\n\n \n\nSpeaker: Aaron Tay\, Academic Librarian\, Singa
 pore Management University\n\nTitle: Agentic AI might be changing everythi
 ng again\n\nSubtitle: What an LLM with tools can actually do for informati
 on work\, and things to be cautious about\n\n*This webinar will not be rec
 orded\n\n \n\nDescription: Just as librarians were getting comfortable wit
 h chatbots\, the ground shifted again. Agents or LLMs that use tools in a 
 loop — can now search\, read what they find\, reformulate\, retry and re
 cover from dead ends the way a human expert does. Early adopters are alrea
 dy building powerful flows on harnesses like Claude Code and Codex\, which
  marry the flexible but non-deterministic nature of LLMs to the determinis
 m of code. In academic search specifically\, agentic search has arrived.\n
 \n \n\nThe landscape is moving on two fronts. Search startups such as Unde
 rmind\, Elicit and Consensus now claim to support more agentic search on t
 heir platforms. At the same time\, a widening set of providers is shipping
  MCP (Model Context Protocol) servers that pair with an LLM to allow build
 ing of powerful\, sophisticated home-brew agentic skills: Wiley\, Scite\, 
 Consensus and Elicit are live (together with unofficial MCPs to free searc
 h tools like PubMed\, OpenAlex\, Semantic Scholar etc)\, with EBSCO and Cl
 arivate announced.\n\n \n\nIn this talk I will give some clarity on how ag
 entic search actually works\, how it relates to "deep research" and "deep 
 search"\, and what the existing empirical evidence says about how well it 
 performs on modern tough information-retrieval benchmarks like BrowseComp-
 Plus.\n\n \n\nCloser to home\, I have found that linking even a relatively
  weak LLM to a simple\, home-brew Primo MCP server resulting in a LLM that
  can iterate on its own prior results — produces surprisingly large impr
 ovements in database discovery for inexperienced users\, well beyond simpl
 e synonym expansion.\n\n \n\nI will briefly highlight two further examples
 : an agentic skill that uses PubMed and MeSH tools to construct and pilot 
 highly sensitive PubMed search strategies for systematic reviews\, and an 
 advanced lit-review orchestrator that combines Undermind\, SSRN and other 
 avenues with deduplication and validation against hallucination.\n\n \n\nB
 ut putting an LLM in the search loop is already known to hurt the interpre
 tability and reproducibility of results. Does agentic search make that wor
 se? I will share partial findings from my own mini-study on exactly that q
 uestion.
LOCATION:
SUMMARY:Agentic AI might be changing everything again: What an LLM with too
 ls can actually do for information work\, and things to be cautious about
URL;VALUE=URI:https://events.oregonstate.edu/event/agentic-ai-might-be-chan
 ging-everything-again-what-an-llm-with-tools-can-actually-do-for-informati
 on-work-and-things-to-be-cautious-about
CATEGORIES:Conference or Workshop
END:VEVENT
END:VCALENDAR