ACS Central Science | Researchers from Insilico Medicine and Lilly publish foundational vision for fully autonomous “Prompt-to-Drug” pharmaceutical R&D

(Press-News.org) CAMBRIDGE, Mass. – February 20, 2026 – Insilico Medicine (“Insilico”, 3696.HK) The convergence of generative AI, multimodal foundation models, and automated laboratory 

systems is accelerating a fundamental transformation in drug discovery. Despite major technological progress, most pharmaceutical R&D remains fragmented across computational tools and manual experimentation. 

 

To address this challenge, researchers from Insilico Medicine and Lilly published a landmark perspective in ACS Central Science describing a comprehensive framework for fully autonomous, AI-orchestrated drug discovery. The article, “From Prompt to Drug: Toward Pharmaceutical Superintelligence,” outlines how an advanced reasoning system can integrate AI-driven target discovery, generative chemistry, automated synthesis, biological validation, and clinical planning into a single workflow.

 

In this vision, a scientist could simply request, “Design a drug for idiopathic pulmonary fibrosis,” and a central AI controller would autonomously delegate and coordinate drug discovery tasks, such as identifying targets, proposing optimized chemical structures, executing in vitro and in silico validation, and generating clinically informed development strategies, for agentic AI programs to carry out. The paper provides a conceptual overview of this end-to-end “prompt-to-drug” workflow.

 

The publication traces the evolution of AI in biotechnology, spanning traditional machine learning, deep learning, and transformer-based generative models, highlighting how each step expanded AI’s capability across target identification, molecular design, clinical prediction, and automated experimentation. 

 

The paper also presents a next-generation architecture composed of sequentially called subsystems which are based on current standard practices of drug discovery but are each autonomously driven and orchestrated by AI algorithms. Biology modules mine data, generate hypotheses, and validate disease-relevant targets. Chemistry modules use generative chemistry, docking, free-energy calculations, and microfluidic synthesis to iteratively design and optimize compounds. Clinical development modules leverage predictive engines such as InClinico to forecast trial outcomes, patient populations, and design strategies. 

 

How an advanced reasoning controller coordinates these diverse systems and legacy laboratory equipment via APIs is illustrated throughout the paper. Additional highlights include the integration of emerging advanced reasoning models capable of planning multi-step workflows, coordinating specialized AI agents, and revising strategies based on readouts. While powerful, these systems still require safeguards to mitigate hallucinations, error propagation, and data-driven biases. Auditability, human oversight for high-stakes decisions, and the use of “AI arms” in clinical trials is recommended to validate predictive tools in real-world settings.

 

The paper also emphasizes the importance of humanoid-in-the-loop automation to interact with legacy laboratory systems and enable uninterrupted 24/7 experimentation, reducing downtime between chemical and biological steps.

 

Although the authors’ envisioned self-driven, closed-loop drug discovery workflows may seem far beyond what is possible today, they present cases that demonstrate that smaller, individual steps along the pipeline have already been automated and offloaded to AI-based programs to speed up pieces of the full process in proof-of-concept studies. Insilico has worked toward that end as well, developing AI-based tools to facilitate various biological and chemical discovery tasks. For example, Insilico’s DORA and PandaOmics engines can comb through scientific literature and propose new biological hypotheses, Chemistry42 can design novel molecules from user prompts based on 3D structural analyses, and the Chemistry42 Retosynthesis module can plan synthesis of new compounds. Linking all the steps together under one unified framework will be the next frontier in maximizing speed and efficiency of the drug discovery pipeline, the authors argue.

 

“The foundational components for this vision are already operational,” the authors note. “However, to bring this vision to full fruition, collaboration across academia, biotechnology companies, and regulatory agencies is imperative. Achieving truly end-to-end, autonomous drug development will require buy-in from the entire sector, with each player contributing a necessary piece of the puzzle.”

 

This ACS Central Science publication adds to Insilico Medicine’s growing body of peer-reviewed research across target discovery, generative chemistry, and clinical-stage AI applications, supporting the company’s broader vision for Pharmaceutical Superintelligence and closed-loop drug discovery ecosystems. 

 

Target Discovery

Journal of Chemical Information and Modeling: PandaOmics: An AI-Driven Platform for Therapeutic Target and Biomarker Discovery

Aging: Hallmarks of aging-based dual-purpose disease and age-associated targets predicted using PandaOmics AI-powered discovery engine

Generative Chemistry

Nature Biotechnology: Deep learning enables rapid identification of potent DDR1 kinase inhibitors

Journal of Chemical Information and Modeling: Chemistry42: An AI-Driven Platform for Molecular Design and Optimization

Nature Biotechnology: A small-molecule TNIK inhibitor targets fibrosis in preclinical and clinical models

Chemical Science: nach0: multimodal natural and chemical languages foundation model

Clinical

Nature Medicine: A generative AI-discovered TNIK inhibitor for idiopathic pulmonary fibrosis: a randomized phase 2a trial

Clinical Pharmacology & Therapeutics: Prediction of Clinical Trials Outcomes Based on Target Choice and Clinical Trial Design with Multi-Modal Artificial Intelligence

Harnessing state-of-the-art AI and automation technologies, Insilico has significantly improved the efficiency of preclinical drug development. While traditional early-stage drug discovery typically requires 3 to 6 years, from 2021 to 2024 Insilico nominated 20 preclinical candidates, achieving an average turnaround - from project initiation to preclinical candidate (PCC) nomination - of just 12 to 18 months per program, with only 60 to 200 molecules synthesized and tested in each program.

About Insilico Medicine

Insilico Medicine, a leading and global AI-driven biotech company, utilizes its proprietary Pharma.AI platform and cutting-stage automated laboratory to accelerate drug discovery and advance innovations in life sciences research. By integrating AI and automation technologies and deep in-house drug discovery capabilities, Insilico is delivering innovative drug solutions for unmet needs including fibrosis, oncology, immunology, pain, and obesity and metabolic disorders. Additionally, Insilico extends the reach of Pharma.AI across diverse industries, such as advanced materials, agriculture, nutritional products and veterinary medicine. For more information, please visit www.insilico.com.

Forward-Looking Statements:

This press release contains forward-looking statements relating to the likely future developments in the business of the Company and its subsidiaries, such as expected future events, business prospects or financial performance. The words “expect”, “anticipate”, “continue”, “estimate”, “objective”, “ongoing”, “may”, “will”, “project”, “should”, “believe”, “plans”, “intends”, “visions”, “schedule” and similar expressions are intended to identify such forward-looking statements. These statements are based on assumptions and analyses made by the Company at the time of this press release in light of its experience and its perception of historical trends, current conditions and expected future developments, as well as other factors that the Company currently believes are appropriate under the circumstances. However, whether actual results and developments will meet the current expectations and predictions of the Company is uncertain.  Actual results, performance and financial condition may differ materially from the Company’s expectations. 

All of the forward-looking statements made in this press release are qualified by these statements. Consequently, the inclusion of forward-looking statements in this press release should not be regarded as representations by the Board or the Company that the plans and objectives will be achieved, and investors should not place undue reliance on such statements. 

The Company, its Board, the employees and the agents of the Company assume (a) no obligation to correct or update the forward-looking statements contained in this press release; and (b) no liability for any losses in the event that any of the forward-looking statements do not materialise or turn out to be incorrect.

 

END