Are We Rubbing Sticks Together to Create Fire?
Google’s CEO has said that AI is akin to the invention of fire. Karen DeSalvo, the company’s chief health officer, compared itto the discovery of penicillin. In the 1930s and 40s, infectious disease was the leading cause of premature mortality, and many died of bacterial infections after something as simple as an encounter with a thorn.
Today, chronic diseases cause most early deaths, and the cure will be much more complex because it has to bring together data about social determinants, genetic risk, and other causes. It then has to make sense of it and make it personal. That is the promise of generative AI, which is why it may be theantibiotic of our age.
DeSalvo said information is a crucial determinant of health, and that is already Google’s business. It is a company already in people’s lives every day, so it is well-positioned to deliver the knowledge people need for 99 percent of their lives when they are not patients but are trying to be healthy.
When we do need treatment, AI may make it more available. DeSalvo pointed to the use of AI in analysing clinical images. Some of these currently require two “reads” (expert reviews) in the United Kingdom, and specialists are scarce. AI can now provide the second read and cut weeks or even months off the time until treatment begins.
The impact may be most felt in countries with far more serious shortages of health professionals. In rural Kenya, frontline healthcare workers use an AI-linked smartphone for screening and diagnosis. Google is working with the Government of India and a leading private hospital chain to improve the diagnosis of respiratory diseases in multiple settings using AI.
AI can, though, also conserve and extend the capabilities of workforces everywhere. “There is a cognitive burden,” DeSalvo said, that leads to burnout and a decreasing willingness to work in medicine. De Salvo believes we will become used to the new generative AI environment very rapidly. For example, many young people would rather talk to a chatbot than to a human about mental health issues.
Google does not want to deliver many health products; it wants to be an enabler for partners, including developers, governments, and care providers. Sometimes, it partners in lighthouse experiences: partnerships designed to show the way or even chart the course.
Alphabet, Google’s parent, made AlphaFold available to predict the structure of proteins. It is Open Source and available for use by multiple entities, but Alphabet has set up its own new subsidiary, Isomorphic Labs, to deploy AlphaFold capabilities. It is currently particularly important in the search for treatment for orphan diseases, but the protein prediction capabilities may be just as important in the future in countering food scarcity.
DeSalvo spoke at HLTH Europe, the first Amsterdam version of the annual event that has come to dominate the US health ecosystem.
