What is AI’s role in the lab?

If an AI system was tasked with overcoming this challenge, its solution could be to find the different applicational uses of sticking sand to paper, a perfectly valid response.

In a 2020 survey conducted by Cognilytica, 90 per cent of respondents said they were looking to implement more artificial intelligence (AI) technology in their business. Despite the fixation on AI as the next step in technological advancement, AI solutions are not always the best for the job and sometimes it’s better to stick with a personal touch.

The decisions AI can't make

US corporation, 3M is well known for their manufacturing tapes and adhesives but they originally started out as the Minnesota Mining and Manufacturing Company. They specialised in mining corundum, a crystalline form of aluminium oxide commonly used as an abrasive, which would be used to manufacture sandpaper. However, when the Great Depression hit in the 1920s, manufacturing plummeted and the company needed to go in a different direction if it wanted to survive.  

If an AI system was tasked with overcoming this challenge, its solution could be to find the different applicational uses of sticking sand to paper, a perfectly valid response. Generally speaking, traditional AI solutions work by analysing large amounts of data, using a predetermined set of rules and limitations that have been programmed into it to achieve a specific goal. In this case, because the AI solution operates based on inputted patterns to produce predictions, it rarely results in creative solutions.

AI would not have suggested for 3M to transition from sandpaper to tape. The decision, made by two of the buyers, transformed the small business into a Fortune 500 corporation. They stopped selling into the manufacturing industry and started selling to corner shops and supermarkets. Today, you’ll find 3M’s brand Scotch Tape in households across the globe. The creative aspect of problem solving comes more naturally for human intelligence than it does for its artificial counterpart.

The decisions AI can help make

This does not mean that AI is not a useful tool in a scientific setting. IBM Watson Health, a healthcare focused AI solution, is currently being trialled in several countries to support medical professionals with patient diagnosis. The software supports physicians needing to quickly siphon through vast amounts of medical information.   

By using natural language capabilities, hypothesis generation and evidence-based learning, IBM Watson can support critical decision making, and the results look promising. However, this does not mean that we can remove the role of human intelligence. The investigative nature humans possess is necessary, in a healthcare environment or a lab environment, just as creativity is.

Currently, AI’s main purpose in the laboratory is to interpret the large datasets that are often present during research and development. But, developments in AI could see greater involvement in data analysis, some of which we’ll be able to see at Lab Innovations. In the future, it would be great for industries like pharmaceuticals and chemicals to follow in the footsteps of other sectors like the automotive industry. Here, supervisory control and data acquisition (SCADA) systems that have been collecting data for decades have moved towards an Industry 4.0 business model. However, I believe that the laboratory industry will remain experimental and human-centric for some time yet.

Innovation DB will exhibit at Lab Innovations, the UK’s only exhibition dedicated to the entire laboratory industry, on November 3 and 4, 2021. The show will host more than 160 leading scientific suppliers and manufacturers showcasing the latest laboratory developments and equipment. To find out more, visit https://www.lab-innovations.com/

Author: Gerald Law is chief executive at Innovation DB