Artificial intelligence in health Artificial intelligence in health

By about 2050, up to the smallest detail, every human will be the focus of an artificial intelligence-assisted universe (AI), states an expert.

From what is in your fridge to the temperature of the room, digital assistants will regulate all these spaces: screens will turn on your favorite TV show as soon as you enter the room, your car will not need a driver and your bartender will probably be an android.

There are a lot of predictions regarding a future shaped by artificial intelligence (AI). Antoine Blondeau, who was the forefront of the technology leading to the development of Siri (Apple’s voice assistant), shares his insights with us. “In 30 years, the world will be very different,” states the 48-year-old from France, who left his country a long time ago and now lives both in California and Hong Kong. “Everything will be designed to meet your personal needs.”

Health will also be completely transformed, he asserts. Patients will have their entire medical records available to them and AI will be able to make diagnoses. “Having an appointment with your doctor will just serve to make you feel more at ease by being able to talk with someone or because humans will be the only ones authorized to prescribe medications.”

Antonie Blondeau is more of a skeptic, however. Blondeau was the general director of the technology company Dejima when it developed the CALO project (one of the most recognized AI programs in the United States) and he was also the developer of Siri’s predecessor. “We will achieve a certain level of AGI, but it is unclear whether we will ever be able to create something that matches our intuition,” he states. For him, with the impending changes of the future, it will be necessary to have a different take on education and careers.

Trial cases

Along with the U.S. university, MIT, Sentient is also developing what is being referred to as an “AI nurse.” In testing that was done on the blood pressure of thousands of patients, this nurse was able to identify people who may develop sepsis -a potentially fatal condition- 30 minutes before the onset of its initial visible symptoms and its diagnosis had a 90% accuracy rate. “This is a crucial window of time that gives doctors a greater opportunity to save lives,” adds Blondeau, acknowledging that these concepts can also bring up resistance in people. “People believe they can trust their physician, but will they be able to trust in a machine?”