Forget ‘anti inflammatory’ diets; 2019 could be the year of another type of AI: artificial intelligence. And according to experts like Dr Eric Topol, cardiologist and New York Times Op Ed writer, the two are not necessarily mutually exclusive (although there were some surprising findings regarding stereotypically ‘healthy’ vs. ‘unhealthy’ foods). But more on that later. First: what exactly is an artificial intelligence diet?
The general explanation is this: a diet which involves inputing as much of your data as possible into a mobile application, religiously recording what you eat, taking tests to see how your body reacts on a molecular level and working out exactly what foods suit you—rather than going off a ‘universal’ (however well intentioned) diet like Keto or Whole30.
As Dr Topol explained in a recent piece for The New York Times, it looks a little something like this: “I… (used) a smartphone app to track every morsel of food I ate, every beverage I drank and every medication I took, as well as how much I slept and exercised. I wore a sensor that monitored my blood-glucose levels, and I sent in a sample of my stool for an assessment of my gut microbiome. All of my data, amassed with similar input from more than a thousand other people, was analysed by artificial intelligence to create a personalised diet algorithm.”
“The point was to find out what kind of food I should be eating to live a longer and healthier life.”
Why is this necessary? Well, as Dr Topol discovered, just because something is generally healthy (or unhealthy), that doesn’t necessarily mean it is beneficial (or bad) for you. For example, for him, personally, “Cheesecake was given an A grade, but whole-wheat fig bars were a C -. In fruits: Strawberries were an A+ for me, but grapefruit a C. In legumes: Mixed nuts were an A+, but veggie burgers a C.”
“Needless to say, it didn’t match what I thought I knew about healthy eating.”
This is because, despite decades of research, we are still floundering in the shallows when it comes to the science of nutrition. And while we have developed some basic guidelines (which are better than nothing), the more we study this topic the more we are aware of our ignorance.
In other words: next time your overly scrupulous mate tells you to lay off the donuts, science (might) have your back in telling them to mind their own…
This is because high-quality trials require people to adhere to a diet for years, and are thus extremely hard to come by. So much so that even the gold standard of randomised trials, which found that the “Mediterranean diet” lowered the risk for heart attacks and strokes, as Dr Topol puts it, “Had to be retracted and republished with softened conclusions.”
Worse still, “Most other studies are observational, relying on food diaries or the shaky memories of participants,” (NYT). In light of this, Dr Topol argues, most studies can only prove association, not anything about cause and effect. Which explains why you often hear one headline one week, and the opposite the next. Not to mention the problem is exacerbated by the food industries’ skewed funding of what one might call ‘convenient’ research.
So what can we do about it? Well although it is an emerging technology which will surely have a few wrinkles to iron out, artificial intelligence—though time consuming—could be a worthwhile investment for you if you want to learn what best to fuel your body with.
Interest piqued? Check out companies like EverlyWell and Day Two, which build on the precedent set by legacy nutrition applications like Noom, but which take physiological readings into their calculations as well (as opposed to just cold data points, which applications like Noom exclusively rely on).
That said: a word of warning. As of now, this technology is as green as a first day university student. So it may be wise to wait until it becomes more widely tested (and available), because—as Dr Topol points out in his NYT article—although a number of companies boast of of their ‘nutrigenomics’ advances, despite their ostensibly all inclusive approach, they don’t get everything right.
“For a fee, they’ll sample your saliva and provide a rudimentary panel of some of the letters of your genome, but they don’t have the data to back their theory up.”
To illustrate the point, he explains how—despite the company’s best intentions—when he tested the ‘artificial intelligence diet’, he ended up with a big problem: “I’ve had two bouts of kidney stones. To avoid a third, I need to stay away from foods high in oxalate, a naturally occurring molecule abundant in plants. But if you look at the recommendations for my personalised diet, many — like nuts and strawberries — are high in oxalate.”
“That’s a big miscue, because my pre-existing medical conditions were not one of the test’s inputs. And as we undergo significant changes through our lives, like pregnancy or aging, we’ll need re-assessments of what our optimal diet should be.”
In light of this he concludes that, “Coming up with a truly personalised diet would require crunching billions of pieces of data about each person,” which would require developing an artificial intelligence smarter than anything currently on the market.
“In addition to analysing the 40 trillion bacteria from about 1,000 species that reside in our guts, as the project I participated in did, it would need to take into account all of the aspects of that person’s health, including lifestyle, family history, medical conditions, immune system, anatomy, physiology, medications and environment.”
The takeaway? Although the rudimental tenants of an AI diet can still (carefully) be put to some use right now, don’t get too excited until the technology is refined.