Shariff, M. | Search bioRxiv

bioRxiv2026

Can AI Conduct Autonomous Scientific Research? Case Studies on Two Real-World Tasks

Recent advances in artificial intelligence (AI) have prompted claims about autonomous "AI scientists," yet systematic evaluations of these capabilities remain scarce. This exploratory study investigates whether current AI frameworks can execute scientific research tasks beyond isolated demonstrations. We tested eight open-source AI frameworks (Agent Laboratory, AutoGen, BabyAGI, GPT Researcher, MOOSE-Chem2, SciAgents, SciMON, and Virtual Lab) on two tasks that aimed to reproduce research on algorithm development from recent papers in uncertainty quantification and protein interaction discovery. In our evaluation, no framework completed a full research cycle from literature understanding through computational execution to validated results and scientific paper writing. While all systems showed competence in conceptual tasks such as planning and summarization, they consistently failed at robust implementation. Every framework produced sophisticated hallucinations. Deployment proved demanding, requiring substantial debugging and technical expertise, which undermines common claims about the democratization of science with AI. Despite these limitations, the frameworks showed promise as research assistants for methodological planning and ideation under careful human supervision. Our findings suggest that the explored AI systems cannot yet autonomously conduct scientific research, but may provide real value for specific subtasks within the research workflow. We offer preliminary observations to help researchers and developers better understand the gap between advertised and actual capabilities of AI in science.

bioinformatics↗

bioRxiv2023

Maintenance of pig brain function under extracorporeal pulsatile circulatory control (EPCC)

Selective vascular access to the brain is desirable in metabolic tracer, pharmacological and other studies aimed to characterize neural properties in isolation from somatic influences from chest, abdomen or limbs. However, current methods for artificial control of cerebral circulation can abolish pulsatility-dependent vascular signaling or neural network phenomena such as the electrocorticogram even when preserving individual neuronal activity. Thus, we set out to mechanically render cerebral hemodynamics fully regulable to replicate or modify native pig brain perfusion. To this end, blood flow to the head was surgically separated from the systemic circulation and full extracorporeal pulsatile circulatory control (EPCC) delivered via a modified aorta or brachiocephalic artery. This control relied on a computerized algorithm that maintained, for several hours, blood pressure, flow and pulsatility at near-native values individually measured before EPCC. Continuous electrocorticography and brain depth electrode recordings were used to evaluate brain activity relative to the standard offered by awake human electrocorticography. Under EPCC, this activity remained unaltered or minimally perturbed compared to the native circulation state, as did cerebral oxygenation, pressure, temperature and microscopic structure. Thus, our approach enables the study of neural activity and its circulatory manipulation in independence of most of the rest of the organism.

neuroscience↗