Researchers at the Indian Institute of Technology (IIT) Delhi, in collaboration with scientists from Denmark and Germany, have developed ‘AILA’. AILA, an Artificial Intelligence (AI) agent, autonomously conducts real scientific experiments. This development advances laboratory automation, particularly in the complex field of materials research.
AILA’s core function involves operating an Atomic Force Microscope (AFM) . The AFM is a crucial instrument in nanotechnology and materials science, used to image and manipulate matter at the nanoscale. It measures surface topography by scanning a sharp probe over a sample, providing detailed information about material properties. Operating an AFM requires precise control and careful adjustment.
The AI agent demonstrates a comprehensive set of capabilities. AILA can independently design experiments, operate the sophisticated AFM equipment, and make real-time decisions during the experimental process. It also collects raw data and performs in-depth analysis of the results. All these tasks are completed without any human intervention or supervision.
A Shift in AI Application
IIT Delhi highlighted AILA’s distinction from current AI models. Traditional AI tools, such as ChatGPT, primarily function as digital assistants. These tools excel at tasks like content generation, data analysis, and answering complex queries. AILA moves beyond this, directly interacting with and controlling physical laboratory equipment.
NM Anoop Krishnan from IIT Delhi underscored this change. He explained that AILA signifies a shift in AI’s role, evolving from merely ‘writing about science’ to actively ‘doing science’. This marks a fundamental change in how AI can contribute to scientific discovery.
Increased Efficiency and Automated Expertise
AILA dramatically increases experimental efficiency. Indrajeet Mandal, the study’s first author and a PhD scholar at IIT Delhi, provided a key example. Adjusting microscope parameters for high-resolution imaging, a task that typically took a full day for human researchers, is now completed by AILA in just 7 to 10 minutes . This speed-up accelerates research timelines significantly.
Operating an AFM instrument demands a high level of specialized skill and experience. Nithya Nand Gosvami noted that researchers often require several years of training to master its operation. AILA’s ability to autonomously manage these intricate tasks means complex experimental research can proceed with reduced reliance on highly trained human operators, freeing them for more conceptual work.
International Collaboration and Identified Challenges
The development of AILA was a collaborative effort involving multiple international institutions. Key partners included IIT Delhi, Aalborg University in Denmark, the Leibniz Institute of Photonic Technology in Germany, and the University of Jena. This global partnership brought diverse expertise to the project.
Researchers involved in the project also identified specific challenges. They observed that AI models demonstrating strong performance in theoretical knowledge might face difficulties. These challenges arise when applying that knowledge to practical execution in real-world laboratory environments, where unpredictable variables often occur.
Alignment with National Initiatives
IIT Delhi stated that this technological breakthrough aligns directly with India’s broader national strategy. The ‘Artificial Intelligence for Science’ initiative aims to integrate AI more deeply into scientific research across the country. Furthermore, the development supports recent funding announcements made by the Anusandhan National Research Foundation (ANRF), which promotes scientific and technological research.