It’s hard to envision all the ways artificial intelligence will impact our lives this year (much less five or ten years!). We’ve already seen huge leaps in things like image recognition, strategic gameplay, cancer recognition & more general medicine, as well as natural resource extraction and a host of others. These fields and applications represent such a small percentage of what AI will eventually help us do. One such application recently featured in the New York Times? The AI diet; and it’s pretty fascinating stuff.
For all the ‘common knowledge’ or self-professed experts out there, it turns out that nutrition is still very poorly understood from a scientific perspective. There’s a reason sooooo many ‘authoritative’ diets are out there — no one can definitively prove one’s incontrovertible superiority over another. Paleo diet; keto diet; Mediterranean diet; Atkins; Whole30… the lists go on and on.
But why is it so hard to nail down nutrition at a scientific level, and how can AI help?
The NY Times article lays it out pretty perfectly, so I’ll just block quote it for context and background:
It turns out, despite decades of diet fads and government-issued food pyramids, we know surprisingly little about the science of nutrition. It is very hard to do high-quality randomized trials: They require people to adhere to a diet for years before there can be any assessment of significant health outcomes. The largest ever — which found that the “Mediterranean diet” lowered the risk for heart attacks and strokes — had to be retracted and republished with softened conclusions. Most studies are observational, relying on food diaries or the shaky memories of participants. … Perhaps not surprisingly, these studies have serially contradicted one another. Meanwhile, the field has been undermined by the food industry, which tries to exert influence over the research it funds.
The bottom-line message doctors and nutritionists are coming to after the advent of machine learning? The concept of a singular optimal diet for all people rests on flawed logic that’s proving more and more tenuous over time. It turns out, that with the ability to analyze huge data sets effectively (hello, AI!), we’ve “learned how simplistic and naïve the assumption of a universal diet is,” according to Dr. Topol.
The concept of a universal diet is “both biologically and physiologically implausible: It contradicts the remarkable heterogeneity of human metabolism, microbiome and environment, to name just a few of the dimensions that make each of us unique. A good diet, it turns out, has to be individualized.”
I mean, on a certain level that makes a lot of sense — no two people are identical (even identical twins have different fingerprints, irises, etc.). It would stand to reckon that individualizing diets to account for that uniqueness would be ideal for achieving health goals. So what does an AI diet / testing look like in practice today?
For Dr. Topol, he used “a smartphone app to track every morsel of food [he] ate, every beverage [he] drank and every medication [he] took, as well as how much [he] slept and exercised.” He also wore a “sensor that monitored [his] blood-glucose levels, and [he] sent in a sample of [his] stool for an assessment of [his] gut microbiome. “All of that data was collected and combined with “similar input from more than a thousand other people” before it was “analyzed by artificial intelligence to create a personalized diet algorithm. The point was to find out what kind of food [individuals] should be eating to live a longer and healthier life.”
That may not be feasible for consumers just yet, but it hints at the brave new future an AI diet could help usher in. Living longer, healthier lives through the advent of highly personalized nutrition algorithms.
So, Siri, what should I eat next?