Pharmaceutical companies are tapping into artificial intelligence (AI) to help reduce costs by streamlining the drug discovery process, analyzing vast amounts of data quickly and predicting outcomes more accurately.
An in-depth report from Deep Knowledge Group notes there has been a substantial increase in investment rates in AI-driven pharmaceutical companies since 2015.
The report also found nearly 40% of all AI companies specialize in early drug development, making it the dominant field and one to be trusted from the point of investments.
Out of a total of 800 AI-focused companies in the drug discovery sector, 91 were established between the years 2022 and 2023 – all startups that leverage AI technology. The report also found there is an increasing tendency for new collaborations to emerge, both between Big Pharma companies, as well as with new startups with new technologies and Big Pharma.
The report offers a database of the “Top 100 AI in Drug Discovery Experts”, with profiles on the researchers, scientists, entrepreneurs and technologists.
Over the past 9 years, annual investments in 800 companies skyrocketed 27-fold, totaling $60.3 billion by August 2023, with the most rapid growth occurred in 2021, reaching $13.68 billion. However, due to the global economic downturn, AI investments in drug discovery companies saw reduced growth in 2022 ($10.2 billion vs. $13.68 billion in 2021).
In 2023, total investments hit $60.3B, projected to maintain this trajectory but at a slower pace, reaching a cumulative $80 billion by 2025.
Andrii Savytskyi, strategic director of the life sciences division of Deep Knowledge Group, says AI isn’t just a buzzword – it’s reshaping clinical trials and drug discovery.
“Companies adopting end-to-end AI-enhanced drug development or infusing AI into their clinical pipeline are surging forward, promptly entering clinical trials and regulatory phases,” he says.
He points to Insilico Medicine as an example: In 2022, they introduced 9 preclinical candidates, totaling 14 since 2020. Four are now in human trials, one in Phase II, all while this newcomer to pharma, under a decade old, advances their IND filing.
“Insilico’s journey underscores AI’s power, especially in crucial clinical trial moments, propelling treatment development into unexplored realms,” he says.
Savytskyi explains internal drug discovery spending has notably risen, ranging from $2.6 billion to $6 billion in various assessments.
“Addressing this challenge involves three key strategies: Refining business models for early collaboration and diversified pipelines, embedding AI as a data-driven paradigm shift in drug discovery, and unearthing novel therapeutic modalities like biologics,” he says.
He explains while the second strategy is AI-centric, it profoundly impacts the first and third strategies as well.
“During clinical development, AI algorithms enhance outcome prediction accuracy in trials, notably boosting success rates for studies and future marketable drugs,” Savytskyi says.
Despite the evident advantages of AI in drug discovery, potential risks warrant consideration, with data quality and biases posing a key challenge—AI model efficacy hinges on its training data.
“Biased, incomplete, or subpar data may lead to inaccurate predictions,” Savytskyi warns. “Additionally, interpretability remains elusive as AI operates as a ‘black box’, lacking transparency crucial in regulated fields like pharmaceuticals.”
Dr. Harvey Castro, a physician and health care consultant, adds AI can significantly reduce the costs associated with drug development.
“Introducing a new drug can take over a decade and cost up to $6 billion,” he says. “AI’s ability to analyze and predict outcomes can streamline the process, potentially saving billions.”
He says platforms like Deep Pharma Intelligence’s AI in Drug Discovery offer sophisticated analytical and market intelligence capabilities, including automated SWOT analysis and AI-driven smart-matching tools.
Like Savytskyi, he cautions AI models are only as good as the data they are trained on, and says any biases in the data can lead to skewed results, potentially affecting drug efficacy and safety.
“A clear understanding and communication of how AI models make decisions, especially in critical areas like drug discovery, is necessary,” Castro explains. “The bias of the AI might lead to errors and make these errors worst due to preferences in the air. Another way of thinking is that bias can create barriers to creating correct solutions.”
He says he thinks the investment in AI by pharmaceutical companies will likely continue increasing, as evidenced by the cumulative $60.3 billion investment between 2014 and 2023, adding both significant Pharma and startups stand to benefit.
“The use of AI in drug development is indeed a revolutionary step in health care. It offers promising solutions to longstanding challenges but presents new complexities and ethical considerations,” Castro says. “The continued collaboration between AI experts, health care professionals, and regulatory bodies will be vital in harnessing AI’s full potential in this field.”
