For example, Insilico Medicine was able to design a drug used for idiopathic pulmonary fibrosis using AI, and in the process was able to reduce the cost to one-tenth of the original price and cut the duration by three and a half years. MIT researchers have also developed their own algorithmic framework to be able to identify optimal molecular candidates that will minimize synthetic costs, known as Synthesis Planning and Rewards-based Route Optimization Workflow (SPARROW).

They are specifically using this AI model to curate ways to make the discovery process more cost-effective; SPARROW is able to take on the challenges of batch synthesis by considering the shared intermediary compounds involved in synthesizing molecules and incorporating information into its cost-versus-value function. The AI model goes even further to consider the costs of starting materials and the number of successful reactions with each route. From here, SPARROW is able to select the best of the pathways and create a route and compound that is most suitable for the cost and quality. The AI models go beyond developing different routes more efficiently for researchers; they tap into creating the most efficient routes for targeting disease and financial costs.

However, there are also other innovative ways that AI is contributing to the pharmaceutical industry. Aside from cutting costs and time for drug researchers, AI is allowing companies to expand on their discovery of drugs. Enveda Biosciences is using AI to identify plants with the highest likelihood of providing cures. Viswa Colluru, founder of Enveda, has been able to use the AI technology with Enveda to tap into the world’s digital information about how humans across cultures have used plants to cure pain and disease, noting that this information is thousands of years of experiential human wisdom.

Why wasn’t this done previously? It’s because isolating molecules to identify and test them is much more time-consuming than synthesizing compounds. However, with Enveda’s AI, they are able to gather materials and test them in the lab with the AI model. In return, the AI model can decipher the “chemical language” of the entire sample. This new AI use and discovery allows Enveda to take pharmaceuticals to the next level, deriving from ancient remedies to create so many potential drugs that we don’t know what to do with them.

From an investor’s standpoint, this may seem like the next best thing to dip your toes into. However, some are saying that this AI-powered drug discovery requires a lot of patience. This is because, with the use of AI and its success rates, there is a fluctuation and variance in the outcomes. There is also only so much that can be pursued when developing potential drug candidates. Depending on the relevancy and publications available, AI is only able to predict models using information already provided to train the models to predict future compounds or models.

So, while AI is promising in an industry fueled with drive and ambition to create new drugs for the market, AI also has data gaps that can only be filled through time and deep pockets.