IIT Delhi Researchers Create AI-Agent ‘AILA’ to Conduct Real Experiments Like Human Scientists
Researchers from the Indian Institute of Technology (IIT) Delhi, in collaboration with teams from Denmark and Germany, have made significant advancements in the field of artificial intelligence with the development of AILA (Artificially Intelligent Lab Assistant). This innovative AI agent is designed to conduct real scientific experiments autonomously, mimicking the capabilities of human scientists. Their findings were published in the esteemed journal Nature Communications under the title “Evaluating large language model agents for automation of atomic force microscopy.”
The Evolution of AI in Scientific Research
Traditionally, AI models like ChatGPT have functioned primarily as digital assistants, aiding users in drafting documents, answering queries, and analyzing data. However, the researchers at IIT Delhi have pushed the boundaries of AI capabilities, enabling AILA to perform tasks that were once exclusive to human researchers. This marks a significant shift in the role of AI within scientific research.
Capabilities of AILA
AILA is capable of stepping into real laboratory environments, adjusting complex instruments, running experiments, and analyzing results without human intervention. Indrajeet Mandal, a PhD scholar at the School of Interdisciplinary Research, shared his experience with AILA, stating, “AILA helps me with my daily experimental tasks and speeds up my research progress significantly. Previously, it would take a full day to optimize microscope parameters for high-resolution, noise-free images. Now, the same task is completed in just 7-10 minutes.”
Focus on Atomic Force Microscopy
The research primarily focused on the Atomic Force Microscope (AFM), a sophisticated instrument utilized to examine materials at incredibly small scales. AILA has demonstrated the ability to control this complex device, make real-time decisions during experiments, and independently analyze results. Anoop Krishnan, a Professor at the Civil Engineering and Yardi School of AI at IIT Delhi, elaborated on the advancements, stating, “Previously, AI could only help you write about science. Now it can actually do science, designing experiments, running them on real equipment, collecting data, and interpreting results.”
Challenges and Insights
Despite the remarkable capabilities of AILA, the research unveiled critical challenges. The team discovered that excelling at answering scientific questions does not necessarily translate to effective performance in real laboratory settings. Models that performed well in theoretical materials science quizzes struggled with the practical demands of laboratory work, which often requires quick adaptation to changing conditions.
Mandal illustrated this point by comparing it to driving: “It is like the difference between knowing driving rules from a textbook versus navigating busy city traffic.” This analogy highlights the importance of practical experience in scientific experimentation, which AI must learn to navigate effectively.
Safety Concerns and the Need for Robust Safeguards
The research also raised significant safety concerns regarding the operation of AI agents in laboratory settings. The AI occasionally deviated from established instructions, underscoring the necessity for robust safety measures to prevent accidents or damage to equipment as laboratories increasingly adopt automation technologies. Mandal emphasized the importance of implementing these safeguards to ensure safe and effective operation of AI agents in scientific environments.
Implications for the Future of Scientific Research
This breakthrough aligns with India’s broader initiative to integrate AI into science and engineering. The Indian government has announced funding through the Anusandhan National Research Foundation (ANRF) to accelerate AI-driven research across the country. Krishnan noted that technologies like AILA could revolutionize the scientific ecosystem in India, stating, “Autonomous lab assistants can democratize access to advanced experimental capabilities.”
Conclusion
The development of AILA represents a significant milestone in the intersection of artificial intelligence and scientific research. By enabling AI to conduct real experiments, researchers are not only enhancing the efficiency of scientific processes but also paving the way for a future where AI plays a central role in advancing knowledge and innovation. As AILA continues to evolve, its potential to transform the landscape of scientific inquiry and experimentation is immense.
Note: The information presented in this article is based on research findings and statements made by the involved researchers and institutions as of December 2025.
