Gottlieb predicts AI could replace doctors sooner than anticipated

In a recent opinion piece published with CNBC, former Food and Drug Administration (FDA) Commissioner Scott Gottlieb discussed the potential for artificial intelligence (AI) to take over tasks typically performed by doctors. Gottlieb believes that this future may arrive sooner than expected, stating that “for some tasks, this medical future is sooner than we think.”

Gottlieb outlines two categories of AI tools in healthcare: machine learning, which uses algorithms to learn patterns from data and make predictions, and natural language processing, which understands and generates human language. He explains that, in certain cases, large language models are already analyzing patient medical records and providing diagnoses and treatments directly to patients, without the need for a physician’s involvement. However, one significant hurdle in implementing these AI technologies is establishing a suitable regulatory path, as regulators are hesitant due to concerns about errors and the potential replication of medical mistakes.

Nevertheless, Gottlieb believes that overcoming these hurdles can lead to promising outcomes, such as improving patient care and addressing financial challenges associated with hiring non-physician providers to reduce labor costs. Large language models are also being utilized for administrative tasks like processing medical claims, analyzing records, and providing clinical decision support. In cases where a doctor is involved, the need for FDA regulation may be altered.

Gottlieb emphasizes the use of machine learning technology for analyzing clinical data, images, and scans. These software tools are often categorized as medical devices by the FDA and are trained with verified data sets, which boosts the FDA’s confidence in assessing their integrity. While Gottlieb acknowledges that current AI development may not completely remove doctors from the decision-making loop, he believes that these tools will increasingly enhance providers’ productivity and, in some cases, substitute for them.

Gottlieb cites OpenAI’s ChatGPT as an example, as it successfully passed the U.S. Medical Licensing Exam, while a significant portion of medical students fail this exam each year. This discussion on AI’s potential impact on the medical field coincides with studies showing that AI could lead to job automation in the future. According to a recent report by McKinsey Global Institute, almost 30% of the hours currently worked in the U.S. could be automated by 2030.

In summary, Gottlieb presents a forward-thinking perspective on the role of AI in healthcare, highlighting the benefits it can bring while acknowledging the challenges in regulatory approval and professional acceptance. As technology continues to advance, the integration of AI into the medical field may revolutionize patient care and provider productivity.