AI-BASED ACADEMIC RESEARCH TOOLS

​

On the generative in GenAI

Every generated answer carries a history: the data that trained it, the assumptions that shaped it, the voices the algorithm chose to amplify or silence. Real understanding means tracing those histories, not just accepting their polish.

​

2

Source: Leo S. Lo. (2026). “How Higher Ed Can Adjust to the AI-Answer Economy.” Inside Higher Education. as cited in: Embedding an Equity Lens Janice Kung, MLIS Health Sciences Librarian University of Alberta ViMLoC: Visible Minority Librarians of Canada Feb 3, 2026 to Generative AI’s Impact on Libraries. Presentation by Janice Kung, Feb. 3 2026.

- Leo S. Lo

�ACADEMIC RESEARCH TOOLS 4

1. Sofia (best for locating known items at Concordia & beyond)
2. Library Databases
3. Google Scholar (and competitors like Semantic Scholar)
4. GenAI-based Research Tools

see also the module: Choosing your Search Tool from our Critical Toolkit for Navigating Information

Google Scholar�(and competitors)

​

vs

​

GenAI-based research tools

infographic produced by ChatGPT

�ACADEMIC RESEARCH TOOLS –�AI Research - and Academic - Tools

library.concordia.ca/help/ai/

�

ACADEMIC RESEARCH TOOLS –�AI-based Research Tools

4. AI-based Research Tools
1. for Literature Discovery & Review�
1. Offer searching, synthesis and/or mapping; features evolving almost daily
2. Based on natural language functionality; extremely fast and efficient
3. Searches are usually NOT REPLICABLE - even when you record your prompts
4. Searches mostly limited to open-access sources rather than subscription-based sources; very little book or chapter content
5. Mostly based on fee based personal subscription and/or freemium model
6. Countless ever-evolving ethical, privacy, social, environmental issues

​

​

​

​

​

�AI-based Research Tools �more considerations for you

• Beware of AIsplaining…. or of anything that seems too good to be true
• Don’t let AI preempt your own original thoughts
• Use for understanding vs interpretation, efficiency vs meaning
• You can test an AI tool you are considering using topics well-known to you
• Your profs want to read your take on a topic, not an AI tool’s
• If AI can provide the perfect answer, your research question may need tweaking
• You can do great academic research with or without AI tools

AI & GenAI research tools:��

EXAMPLES�

$$$$ - emphasis on subscription

citation mapping only, does not use large language models (LLMs).

AI-based Research Tools Hallucinations & inaccuracies

Hallucinations - completely made up articles

• This is less likely in academic AI tools as their data sources are based on actual indexes

Inaccuracies - summaries misinterpret/misrepresent the sources

• Summary is not consistent with the sources (e.g., an article supports an argument but the AI summary indicates otherwise)
• AI summary contains inferred information that is not mentioned directly by the sources (e.g., the summary gives wrong connections or illogical conclusions)
• Tendency to overgeneralize and oversimplify – see research by Peters and Chin-Yee, Generalization bias in large language model summarization of scientific research.

​

EXAMPLES OF my �SEARCH RESULTS

​

…watch this context- or workshop-specific examples…..

​

��GenAI-based Research Tools

LEARN MORE

• Concordia Library guide:� AI Literacy
• see especially the section: �AI Academic Tools�

​

​

​

​

​

Databases/Data Sources [NOT AI-based]

Data and descriptions from OpenAlex | About; Semantic Scholar | About; WoS Core Coverage; PubMed | About.

AI tools for discovery

This is not a recommendation or endorsement of any of these tools, just a list to explore along with info for evaluation.

why does it matter where you search?

16