{"id":12591,"date":"2025-07-16T09:00:00","date_gmt":"2025-07-16T07:00:00","guid":{"rendered":"https:\/\/haimagazine.com\/uncategorized\/humans-and-ai-together-in-a-lab-how-do-we-develop-medications-today\/"},"modified":"2025-07-16T14:55:39","modified_gmt":"2025-07-16T12:55:39","slug":"humans-and-ai-together-in-a-lab-how-do-we-develop-medications-today","status":"publish","type":"post","link":"https:\/\/haimagazine.com\/en\/uncategorized\/humans-and-ai-together-in-a-lab-how-do-we-develop-medications-today\/","title":{"rendered":"\ud83d\udd12 Humans and AI together in a lab. How do we discover medications today?"},"content":{"rendered":"

According to SNS Insider data<\/mark><\/a>, the global market value of artificial intelligence in pharmaceuticals was $1.73 billion in 2024. By 2032, it is expected to grow to $13.46 billion. This represents an annual growth rate (CAGR) of 29.33% from 2025 to 2032.<\/p>

It’s worth taking a look at the budgets right away. The traditional method of developing drugs can cost an average of 2.6 billion dollars. AI reduces this cost to about 1.5 billion dollars. According to a study by Boston Consulting Group<\/mark><\/a>, 75 molecules discovered by AI had entered clinical trials by 2023, with 67 remaining in active tests. Today, these numbers are certainly much larger. Interestingly, oncology alone accounts for about 50% of the share in molecules discovered in the early phases of research.<\/p>

AI is no longer just a futuristic curiosity in pharmaceutical research, but a fully-fledged participant in the drug discovery process \u2014 from analyzing biological data to designing molecules and planning preclinical studies. Not only has the pace of work in the laboratory changed, but also the quality of the decisions made.<\/p>

Pharmaceutical research strategy…<\/strong><\/h4>

…primarily relies on machine learning, which allows to analyze vast volumes of biological data in a short time. In practice, this means we can select molecules faster, predict drug interactions more accurately, and identify new therapeutic targets.<\/p>

AI supports scientists when it comes to making decisions, suggests hypotheses, generates models, and automates processes, which speeds up not only laboratory work but also the documentation process and risk analysis.<\/p>

The impact of AI on global research<\/strong><\/h4>

Developing a new drug in the traditional cycle usually takes several years. AI can reduce this time by half. Models that learn from historical data allow for the identification of promising molecules within just a few months.<\/p>

AI tools analyze data from in vitro studies, support the identification of biomarkers, predict the body’s response to the active substance, and enable earlier detection of toxic side effects. What does this look like in practice? For example, like this:<\/p>

Insilico Medicine (China\/Hong Kong) developed rentosertib<\/mark><\/a> \u2014 the first drug where both the therapeutic target and the molecule itself were discovered entirely by generative artificial intelligence. What’s even more impressive, the entire process \u2014 from identifying the need to completing the research \u2014 took less than 30 months and cost just $2.6 million.<\/p>

Meanwhile, DeepMind and its AlphaFold have released structural models of millions of proteins<\/mark><\/a>, which are now regularly used by research institutes and medical teams around the world. This means that, in practice, AI is currently helping to solve structural biology problems that have resisted traditional methods for years.<\/p>

The impact of AI on research in Poland<\/strong><\/h4>

Similarly, though on a smaller scale, this is also happening in our own backyard. Molecule.one has established a strategic partnership with Chemical Abstracts Service (CAS), a division of the American Chemical Society<\/mark><\/a>. The company uses generative AI for chemical synthesis planning. This technology enables the prediction of optimal production pathways for chemical compounds and significantly accelerates the early stages of drug discovery. Meanwhile, Ingenix.ai has raised 9 million euros<\/mark><\/a> in a seed funding round from venture funds. The company’s goal is to create a multimodal AI model for simulating clinical trials \u2014 a model that would predict drug efficacy and adverse effects even before costly testing begins. In turn, NaturalAntibody, which collaborates with AstraZeneca, leverages machine learning systems to discover and optimize antibodies<\/mark><\/a>, drawing on scientific data that had previously remained untapped.<\/p>

AI as an element of the research standard<\/strong><\/h4>

Artificial intelligence is not just heralding a new era of drug discovery, but is also co-creating it. It is expected that in the coming years, models specialized in specific therapeutic areas, such as rare diseases and neurology, will be developed.<\/p>

Thanks to precise predictions, it will be possible to develop therapies tailored to the patient’s individual genetic and biological profile. For healthcare systems, this also represents an opportunity to optimize costs and improve treatment efficacy.<\/p>","protected":false},"excerpt":{"rendered":"

Artificial intelligence supports drug research in an increasingly practical and systematic way. What does a pharmaceutical laboratory look like with AI by its side?<\/p>\n","protected":false},"author":8,"featured_media":12460,"comment_status":"closed","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"_acf_changed":false,"rank_math_lock_modified_date":false,"footnotes":""},"categories":[1],"tags":[],"popular":[],"difficulty-level":[],"ppma_author":[660],"class_list":["post-12591","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-uncategorized"],"acf":[],"authors":[{"term_id":660,"user_id":8,"is_guest":0,"slug":"kamil-swidzinski","display_name":"Kamil \u015awidzi\u0144ski","avatar_url":{"url":"https:\/\/haimagazine.com\/wp-content\/uploads\/2025\/06\/freepik__retouch__63609.png","url2x":"https:\/\/haimagazine.com\/wp-content\/uploads\/2025\/06\/freepik__retouch__63609.png"},"first_name":"Kamil","last_name":"\u015awidzi\u0144ski","user_url":"","job_title":"","description":"\u015aledz\u0119 najnowsze technologiczne trendy, w tym AI. Jako Innovation Manager jestem blisko nowych rozwi\u0105za\u0144 wsp\u00f3\u0142pracuj\u0105c ze startupami."}],"_links":{"self":[{"href":"https:\/\/haimagazine.com\/en\/wp-json\/wp\/v2\/posts\/12591","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/haimagazine.com\/en\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/haimagazine.com\/en\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/haimagazine.com\/en\/wp-json\/wp\/v2\/users\/8"}],"replies":[{"embeddable":true,"href":"https:\/\/haimagazine.com\/en\/wp-json\/wp\/v2\/comments?post=12591"}],"version-history":[{"count":3,"href":"https:\/\/haimagazine.com\/en\/wp-json\/wp\/v2\/posts\/12591\/revisions"}],"predecessor-version":[{"id":12594,"href":"https:\/\/haimagazine.com\/en\/wp-json\/wp\/v2\/posts\/12591\/revisions\/12594"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/haimagazine.com\/en\/wp-json\/wp\/v2\/media\/12460"}],"wp:attachment":[{"href":"https:\/\/haimagazine.com\/en\/wp-json\/wp\/v2\/media?parent=12591"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/haimagazine.com\/en\/wp-json\/wp\/v2\/categories?post=12591"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/haimagazine.com\/en\/wp-json\/wp\/v2\/tags?post=12591"},{"taxonomy":"popular","embeddable":true,"href":"https:\/\/haimagazine.com\/en\/wp-json\/wp\/v2\/popular?post=12591"},{"taxonomy":"difficulty-level","embeddable":true,"href":"https:\/\/haimagazine.com\/en\/wp-json\/wp\/v2\/difficulty-level?post=12591"},{"taxonomy":"author","embeddable":true,"href":"https:\/\/haimagazine.com\/en\/wp-json\/wp\/v2\/ppma_author?post=12591"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}