It seems that Artificial Intelligence (AI) could begin to have great importance in the design of medicines. At a pharmaceutical company called Eli Lilly, generative AI is already used to research molecules.
It is expected that AI capabilities will be able to do in 5 minutes what traditional methods can do in a year. Even though the capabilities are many, one question always remains to be answered: despite the speed, will the results be satisfactory?
It is possible that there are medicines 100% manufactured using AI
At an early stage, Lilly scientists were expected to balk at implementing AI. However, the fact is that they were surprised by its technological potential. One of the scientists compares the use of AI to a kind of “epiphany”.
“We always talk about training machines, but another art is where machines produce ideas based on a set of data that humans wouldn't be able to see. This further stimulates creativity, opening up paths in the development of medicine that humans would not have explored otherwise” – says Diogo Rau (via CNBC).
In the opinion of many experts, it is quite possible that, in a short time, there will be medicines 100% manufactured using AI. There are those who even consider that this could be the norm very soon.
For Kimberly Powell, vice president of health at Nvidia, it is possible to establish a parallel between Chat GPT and this new way of thinking about the pharmaceutical industry.
As he himself states, the systems are also trained on “each book, each web page, each PDF document”. Thus, it is possible to bring together all existing knowledge in an AI, which can be useful in the design of medicines.
What now takes years may soon take months to discover
Illustrative Image (via Copilot AI)
If this new technology is implemented, there could be a natural departure from the methods we know. It is believed that the use of AI can save a lot of time and culminate in the most important thing of all: superior results.
Based on a Nature study, Powell highlights the huge difference between drug discovery times. In practice, this considers that what previously took many years can now be achieved in a few months.
Another aspect that stands out is the ability that AI may have in discovering what was previously non-existent. In practice, experts in the field believe that this will be able to find new proteins that we have never heard of.
In this sense, a scholar and postdoctoral fellow at the University of Texas, Daniel Diaz, believes that AI will have two main focuses. The first is to discover molecules. The second, from its most important perspective, is the development of new biological products (via CNBC).
At the moment, the period is one of studies and rehearsals. However, the potential of AI in the pharmaceutical world depends on one big factor: the results in human trials. For this reason, it is still necessary to find “concrete evidence”, as Powell says.
In any case, expectations within the scientific community are quite good.
The post AI-made medicines could be on the way: everything you need to know appeared first on Aroged.