Summary 

Artificial intelligence in healthcare - historical roots, technological shifts, and societal implications

Artificial intelligence (AI) has become a central component of contemporary health care, sometimes referred to as »Sund AI« in the Danish context. Early visions of AI in medicine emerged in the 1960s with rule-based expert systems such as MYCIN and INTERNIST-1, which demonstrated the potential to match physicians in narrow diagnostic tasks but never reached broad clinical implementation due to technological, organisational, and ethical barriers. Following periods of stagnation – the so-called »AI winters« – progress in health AI was enabled by the digitalisation of health data, electronic medical records, and advances in machine learning. From the 2000s onward, data-driven approaches, exemplified by IBM Watson and DeepMind’s AlphaFold, marked a new era of learning systems with practical applications in diagnostics, treatment, and biomedical research. Today, health AI is embedded in a broader political, economic, and technological landscape, shaped by massive investments from governments and tech companies, regulatory initiatives such as the EU AI Act, and rapid advances in multimodal systems. At the same time, critical questions remain about responsibility, trust, patient autonomy, and social justice. AI in health care holds transformative potential, but its future impact depends on how societies manage ethical risks, secure trust, and ensure equitable access.

https://pure.au.dk/portal/da/publications/kunstig-intelligens-i-sundhedsv%C3%A6senet-historiske-r%C3%B8dder-teknologi/