Anthropic CEO Dario Amodei has made a bold claim. He says cancer, “Alzheimer’s, heart disease and mental illness could become relics of the past in the next 10 years”. However, experts remain divided over whether such ambitious predictions can become a reality even as artificial intelligence (AI) is opening new possibilities in medical research and drug discovery,
“I think if we do this right, we can create an age of miracles where humanity is able to solve problems that have eluded us for millennia,” he added.
Amodei believes AI could dramatically accelerate biological research, drug development and the discovery of new treatments. His optimism is shared by other prominent technology leaders, including Elon Musk and Google DeepMind CEO Demis Hassabis.
Hassabis has suggested that AI could compress years of drug discovery into weeks, potentially bringing humanity closer to an era where diseases can be cured or prevented. He has cited AlphaFold, Google’s AI system that has predicted the structures of more than 200 million proteins, as an example of how artificial intelligence can accelerate scientific research.
However, AlphaFold’s achievement does not necessarily mean AI will be able to develop cures for complex conditions such as cancer and Alzheimer’s within the next decade.
Musk has also predicted a major transformation in healthcare, claiming that AI could eventually provide everyone with access to medical care comparable to that received by the President of the United States.
Experts remain sceptical of AI’s medical promises
Not everyone is convinced by these predictions. Jeff Leiden, executive chairman of Vertex Pharmaceuticals, has dismissed some of the claims surrounding AI-driven drug discovery as “complete BS”.
Leiden pointed to an exercise in which Vertex evaluated around 50 AI companies tasked with analysing 10,000 drug compounds. According to him, none of the companies independently produced a working drug.
Pharmaceutical experts argue that identifying a promising drug candidate is only the first stage of a lengthy and complicated process. Potential treatments must undergo extensive laboratory research, animal testing and human clinical trials before they can be approved for medical use.
Drug development typically takes 10 to 12 years, with clinical testing and regulatory reviews accounting for a substantial part of the process. Even if AI significantly reduces the time required to identify promising molecules, it cannot eliminate the need to establish their safety and effectiveness.
A compound that appears promising in a computer simulation may fail during laboratory experiments or clinical trials. Some treatments may also produce unexpected side effects or prove ineffective in patients.
AI breakthroughs vs the reality of medical research
Critics, including cognitive scientist Gary Marcus and cardiologist and digital medicine expert Eric Topol, have questioned whether today’s AI systems possess the deeper biological and causal understanding needed to deliver major medical breakthroughs.
As of 2026, AI has contributed to drug discovery and development, but no AI-designed drug has yet achieved full clinical adoption, highlighting the gap between computational predictions and established medical treatments.
There are also concerns that sweeping predictions about ending diseases could create unrealistic public expectations. Medical history offers several examples of ambitious promises that failed to meet their projected timelines.
The 1970s ‘War on Cancer’ is often cited as one such example. Despite decades of research and significant advances in diagnosis and treatment, cancer remains a major global health challenge.
AI could undoubtedly transform how scientists understand diseases, identify drug candidates and develop treatments. However, turning these discoveries into safe and effective medicines requires years of testing, clinical evidence and regulatory approval.
While the technology may shorten parts of the drug development process, whether it can eliminate some of humanity’s most challenging diseases within a decade remains an open question.
Also Read:
Can ultra-processed foods raise colorectal cancer risk? What a study of 461,000 adults found