Automating the Entire Laboratory
Recognizing Chemical Molecules Like Digital Code
Connecting AI-Discovered Chemical Molecules to Real-World Experiments

Imagine a computer that, with the press of a button, could instantly produce everything from advanced plastics to pills. Such an invention may not be far off in the future. A startup backed by global venture capital is developing the world's first chemical computer, also known as the "Chemputer."


The First Computer to Recognize and Synthesize Chemical Molecules

Experimental computer. University of Glasgow homepage

Experimental computer. University of Glasgow homepage

View original image

The Chemputer is a neologism combining "chemistry" and "computer," and it is being developed by Lee Cronin, a professor at the University of Glasgow and founder of the UK-based synthetic chemistry startup Chemify. Professor Cronin first introduced the term in a paper published in the international scientific journal Science in 2018, where he defined it as a "modular robotic system operating with a chemical programming language."


According to this paper, the Chemputer is used to produce chemical substances. When commands are entered via keyboard, a robotic arm and pump connected to the computer directly manipulate reagents, pipes, beakers, and condensers to induce chemical bonding at the molecular level. In essence, it is a device that fully automates the chemical laboratory.


The Chemputer operates with its own unique language system, known as the "chemical programming language." This language is a database that converts all molecular structures currently known in academia into digital code. Chemists can use the Chemputer to search this data and select only the desired molecules for synthesis. When the synthesis process is executed, the Chemputer directly operates the experimental equipment to trigger chemical reactions. As a result, it can produce new chemical substances thousands of times faster than a human chemist.


When commands are entered into the computer, it directly mixes reagents and initiates chemical reactions. University of Glasgow website

When commands are entered into the computer, it directly mixes reagents and initiates chemical reactions. University of Glasgow website

View original image

Dr. Cronin began his Chemputer research early on with support from the US Defense Advanced Research Projects Agency (DARPA), and in 2019, he founded the startup Chemify to commercialize the Chemputer. Over the past six years, Chemify has attracted more than 90 million dollars (approximately 131 billion won) in investments from global venture capital firms and is now on the verge of full-scale mass production.


Bridging the Gap Between Theory and Experiment

Why are top US research institutions and investors so interested in the Chemputer? The answer is linked to artificial intelligence (AI). Recent rapid advances in AI have accelerated research into new materials and pharmaceuticals. AI accesses vast databases of scientific papers to identify new chemical formulas with potential applications in industry and medicine. Isomorphic Labs, a spin-off from Google's AI research division DeepMind, secured 600 million dollars (approximately 870 billion won) in investment this April, and OpenAI announced in June that it would enter the drug development business. All of these investments were made based on the expectation that AI will drive new discoveries in chemistry.


An engineer from Chemify operating the first unit computer 'Chemipham'. Chemify

An engineer from Chemify operating the first unit computer 'Chemipham'. Chemify

View original image

The problem is that even if AI discovers a new chemical molecule, it remains a virtual substance that exists only in computer simulations. In an interview with the biotechnology media outlet GEN News this February, Dr. Cronin pointed out, "AI-driven drug development companies often start with molecular design before even verifying whether the candidate compound can be produced in reality."


The process of manufacturing new substances is extremely challenging. A measurement error of just one milligram (mg) or a slight temperature variation can cause the entire experiment to fail. Sometimes, the skill of the scientists conducting chemical experiments in the lab determines the outcome. By replacing the uncertainties of manual work with machines, the Chemputer serves as the missing piece that bridges the gap between chemical theory and experiment created by AI.



Chemify's first Chemputer unit, "Chemifarm," is being installed at its headquarters in Glasgow, UK, and is expected to expand to the United States in the future. Experiments to produce new substances using the Chemputer are already underway. Chemify has received funding from the Gates Foundation, operated by Microsoft founder Bill Gates's fortune, and has begun developing new drug compounds to combat malaria and tuberculosis.


This content was produced with the assistance of AI translation services.

© The Asia Business Daily. All rights reserved. Unauthorized AI training and use prohibited.

Today’s Briefing