Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Experiment Videos

Practical Use of Advanced AI Frameworks on Real-Life Scientific Problems: Three Case Studies.

Halime S A Gulluoglu, Jibin Baby, Kirti M Bagul

    Biorxiv : the Preprint Server for Biology
    |July 10, 2026
    PubMed
    Summary

    Related Concept Videos

    Issues And Trends In Healthcare Delivery System01:29

    Issues And Trends In Healthcare Delivery System

    The issues and trends in healthcare delivery are constantly changing. The COVID-19 pandemic is one recent issue that wreaked havoc on healthcare systems, causing a shortage of healthcare workers, high demand for medicines and supplies, and increased medical expenditure due to a lack of insurance. Other issues include rising healthcare costs and care fragmentation.
    Cost Containment
    Payment for healthcare services has historically promoted adoption of costly and often unnecessary or inefficient...
    Case Studies01:22

    Case Studies

    There are many research methods available to psychologists in their efforts to understand, describe, and explain behavior and the cognitive and biological processes that underlie it.

    You might also read

    Related Articles

    Articles linked to this work by shared authors, journal, and citation graph.

    Sort by
    Same author

    Toward Ultrafast Depth Sensing via Active Event-Based Stereo Vision.

    IEEE transactions on pattern analysis and machine intelligence·2026
    Same author

    Sn<sup>2+</sup>-Induced In Situ Phase Modulation Enabling a Highly Active o-SnSe/h-NiSe/r-Ni<sub>3</sub>Se<sub>2</sub> Multifunctional Heterostructure for Photo-Thermal/Electronic Urea Water Splitting.

    Advanced materials (Deerfield Beach, Fla.)·2026
    Same author

    Rare but Instructive: Pediatric-Onset Spinocerebellar Ataxia Type 2 and Its Clinical and Genetic Insights.

    Cerebellum (London, England)·2026
    Same author

    Au-Coordinated Motifs with Optimization of Electronic Structure and Hydrogen Bond Network in Zr-Based Metal-Organic Frameworks for Enhanced Photocatalytic Hydrogen Evolution.

    Journal of the American Chemical Society·2026
    Same author

    Network Pharmacology and Transcriptomic Analysis Reveal the Therapeutic Mechanisms of Xiebai Bining Decoction in Hyperuricemic Nephropathy.

    Combinatorial chemistry & high throughput screening·2026
    Same author

    Textile Radio-Frequency Active Devices and Systems: Wireless Communication and Energy Harvesting.

    Research (Washington, D.C.)·2026
    This summary is machine-generated.

    Agentic artificial intelligence (AI) systems show promise but currently struggle to autonomously conduct scientific research. Human-led AI-assisted approaches yield better results, highlighting AI

    Area of Science:

    • Artificial Intelligence in Scientific Research
    • Computational Biology
    • Drug Discovery

    Background:

    • Agentic artificial intelligence (AI) systems are increasingly marketed for automating scientific research.
    • Independent evaluations reveal a gap between claimed capabilities and actual performance.

    Purpose of the Study:

    • To evaluate the performance of frontier agentic AI systems on practical scientific problems.
    • To compare autonomous AI execution versus human-led, AI-assisted approaches.

    Main Methods:

    • Testing agentic AI on three distinct scientific challenges: treatment non-response prediction, chemical structure recognition, and drug property prediction.
    • Comparing autonomous AI runs against human-led, AI-assisted workflows for each task.

    Related Experiment Videos

    Main Results:

    • Autonomous AI frameworks failed in most tested scenarios.
    • Human-led AI-assisted work generated reusable resources and demonstrated modest performance.
    • New insights into treatment resistance mechanisms and a practical benchmark for chemical structure mining were developed.
    • One autonomous AI framework successfully matched human expert performance in drug property prediction.

    Conclusions:

    • Current agentic AI frameworks can support human experts in engineering and analysis but cannot yet independently engineer novel scientific solutions.
    • The application of AI in real-world scientific problem-solving is currently more of an art than a routine technology.