AI in Medicine Under Scrutiny: OpenEvidence's JAMA Paper Raises Questions Amid Isomorphic's Trial Breakthroughs

A new paper in the Journal of the American Medical Association (JAMA) on OpenEvidence, a widely used AI tool for doctors, reveals a surge in its use by clinicians across various specialties but leaves critical questions about its actual impact on patient care unanswered. This comes as Isomorphic Labs, Alphabet's AI drug discovery firm, makes significant strides, launching human trials for its AI-designed cancer drugs and reportedly seeking a massive $40-50 billion valuation. These contrasting developments highlight the rapid advancement of artificial intelligence in medicine, but also the growing need for rigorous scrutiny and ethical guardrails. The JAMA study, which looked at 2.5 years of OpenEvidence usage by nearly a million US clinicians, shows a clear shift towards 'retrieval' of medical knowledge via AI tools, with many doctors using it to explore options outside their primary expertise. OpenEvidence itself has seen explosive growth, recently securing $250 million in funding at a $15 billion valuation and expanding its AI models and global reach through partnerships with major institutions like Memorial Sloan Kettering Cancer Center and AI giant Anthropic. However, the paper's 'open questions' about patient outcomes underscore a broader industry-wide debate on clinical validation and ethical considerations for AI in healthcare, a topic also addressed by recent JAMA summits and AMA frameworks. Meanwhile, Isomorphic Labs' progress from predicting protein structures with AlphaFold to actual human trials for new medicines demonstrates the immense, but still unproven, potential of AI in drug discovery. Looking ahead, the healthcare world will be keenly watching how regulatory bodies and medical communities respond to the need for clearer guidelines and robust clinical validation for AI tools like OpenEvidence. The ongoing ethical discussions around algorithmic bias and transparency remain crucial as these technologies become embedded in daily clinical practice. For Isomorphic Labs, the focus shifts to the results of its human clinical trials, which will be the real test of whether AI's predictive power can translate into safe and effective new drugs for patients, bridging the gap between digital prediction and proven medical efficacy.