Lee Cronin and his team at the University of Glasgow have developed an “organic synthesis robot” – a reaction system controlled by a machine learning algorithm. According to Cronin, the bot is capable of automatically performing chemical reactions and analysis much faster than its human counterpart.

Cronin says he wanted to make the process of creating molecules faster, asserting that the process, when using a human chemist, is more lengthy – it can potentially take up to ten years- and is a waste of the chemist’s time. In an effort to speed up the process of creating molecules, he decided to create and trial a robot that would be able to start mixing random chemicals together, without prior knowledge.

When creating the bot Cronin was keen to ensure that the data fed into the system had no bias: “Chemists want to give the robot their ideas. I was very adamant that I didn’t want to. I wanted to see what happened when I remove the bias because I am pretty sure now that we are going to discover completely new reactions and completely new reactivity the chemist could not imagine.”

Game Changing Technology

The robot that was created is capable of running six experiments in parallel, reactions are assessed in real-time by mass spectrometers, nuclear magnetic resonance machines and infrared spectrometers. This means it can perform up to 36 experiments per day, which is considerably faster than the average human chemist.

The first results of this AI robotic chemist have now been published in the journal Nature. Cronin and his team have reported in their new paper that the machine learning system was able to predict the reactivity of roughly 1,000 reaction combinations with accuracy greater than 80% after considering the outcomes of slightly over 10% of the dataset.

After the predictions were confirmed by a chemist the team were surprised to learn that the robot has also discovered four new reactions in the process, with one of the reactions classed to within the top 1% of the most unique reactions known  .

The team have also started compiling a database of failed reactions, which means chemists will not waste time repeating experiments already proven to not work in the future. “What the robot allows us to do is basically make these most reactive discoveries many times faster. If this robot gives us nothing else what we’re able to do is cut down the workload of the chemist by 90%” says Cronin.

“No Chemists, No AI” Says Cronin 

These initial results are extremely positive as the robot could potentially optimise and discover new molecules and by taking their molecular code it could produce chemicals on demand. This robot could be a real game-changer in the field of chemistry as it could, in theory, discover and produce genetically tailored medicine or even create new non-polluting plastics.

For chemists worried they might be replaced by AI, Cronin offers reassurance saying: “There are a lot of people overhyping AI as some kind of sentient thing.

“The fact is that AI is little more than a regression algorithm. The training comes from the chemist. No chemist, no AI.”

Instead, Cronin believes the robot will be a help to the modern chemist, taking on the brunt of laborious tasks allowing the humans to spend more time on research. Speaking of this revolutionary new approach he said: “This is a key step in the digitisation of chemistry, and will allow the real time searching of chemical space leading to new discoveries of drugs, interesting molecules with valuable applications, and cutting cost, time, and crucially improving safety, reducing waste, and helping chemistry enter a new digital era.”