I knew that I was very probably not going to die, of course. Very few people get ill from pufferfish in restaurants. But I still felt giddy as I took my first bite, as though I could taste the proximity of death in that chewy, translucent flesh. I swilled my saki, squeezed some lemon onto the rest of my sashimi, and looked up. Through the serving window I could see the chef who held my life in his busy hands. We made eye contact for a moment. I took another bite. This is absurd. I am absurd. I pictured the people I love, across the ocean in sleeping California, stirring gently in their warm, musky beds.
My experience in Tokyo eating pufferfish, a delicacy known as fugu, was rich and profound. Fugu has an unremarkable taste. But pufferfish is poisonous; it can be lethal unless it is prepared in just the right way by a highly trained chef. My experience was inflected with my knowledge of the food’s provenance and properties: that this flesh in my mouth was swimming in a tank a few minutes ago and was extracted from its lethal encasement by a man who has dedicated his life to this delicate task. Seconds ago, it was twitching on my plate. And now it might bring me a lonely death in an unfamiliar land. This knowledge produced a cascade of emotions and associations as I ate, prompting reflections on my life and the things I care about.
Fugu is an unfamiliar illustration of the familiar fact that our eating experiences are often constituted by more than physical sensations and a drive for sustenance. Attitudes relating to the origin or context of our food (such as a belief that this food might kill me, or that this food was made with a caring hand) often affect our eating experiences. There is much more to food, as a site of human experience and culture, than sensory and nutritional properties.
You would be hard pressed to find someone who denies this. Yet we are on the cusp of societal changes in food production that could systematically alter our relationship to food and, consequently, our eating experiences. These changes are part of broader trends apparent across nearly all spheres of life resulting from advances in artificial intelligence and other automation technologies. Just as an AI system can now drive your taxi, process your loan application, and write your emails, so AI and related automation tools can now make your food, at home or in a restaurant. Many technologists in Silicon Valley are trying to make automated food production ubiquitous. One CEO of a successful company I spoke with said he expects that almost no human beings will be cooking in thirty years’ time, kind of like how today very few humans make soap, toys, or clothing by hand. It may sound ridiculous, but I’ve found that this vision is common in influential industry spaces.
What might life look like if this technological vision were to come about? This question can appear trivial relative to louder questions about autonomous weapons systems, AI medicine, or the existential threat of a superintelligence. It is not a question of life and death. But I think the question points to a more insidious possibility: that our technological advances might quietly erode the conditions that enable us to experience our day-to-day lives as meaningful.
On the one hand, the struggle for sustenance is a universal feature of human life, and everyone is a potential beneficiary of technology that streamlines food production, like AI that invents recipes or performs kitchen managerial work and robots that prepare food. Home cooking robots could save people time and effort that would be better spent elsewhere. A restaurant that staffs fewer humans could save on labor costs and pass these savings on to customers. Robots could mitigate human errors relating to hygiene or allergies. And then there is the possibility of automated systems that can personalize food to each consumer’s specific tastes and dietary requirements. Virtually every technologist I have spoken to in this industry is excited about a future where every diner can receive a bespoke meal that leaves them totally satisfied and healthy, every time.
Automation brings interesting aesthetic possibilities, too. AI can augment human creativity by helping pioneer unusual flavor pairings. The knowledge that your food was created by a sexy robot could enhance your eating experience, especially if the alternative would be a miserable and underpaid laborer.
These are nice possibilities. But one thing that automation tends to do is create distance between humans and the things that are automated. Our food systems already limit our contact with the sources of our food. For example, factory farming hides the processes through which meat is produced, concealing moral problems and detracting from pleasures of eating that are rooted in participation in food production. AI and robotics could create even more distance between us and our food. Think of the Star Trek replicator as an extreme case; the diner calls for food, and it simply appears via a wholly automated process.
Why is the prospect of losing touch with food processes concerning? For some it might not be. There are many sources of value in the world, and there is no one right way to relate to food. But, personally, I find the prospect of losing touch with food concerning because my most memorable food experiences have all been conditioned by my contact with the processes through which my food came to be.
I have a sybaritic streak. I enjoy being regaled at fancy restaurants with diseased goose livers, spherified tonics, perfectly plated tongues, and other edible exotica. But these experiences tend to pass for me like a kaleidoscopic dream, filled with rarefied sensations that can’t be recalled upon waking. The eating experiences I cherish most are those in which my food is thickly connected to other things that I care about, like relationships, ideas, and questions that matter to me. These evocative connections are established through contact with the process through which my food was made.
I’ve already mentioned one example, but I can think of many others. Like when, in the colicky confusion of graduate school, Sam and I slaughtered and consumed a chicken in the living room of his condo so that we might, as men of principle, become better acquainted with the hidden costs of our food. Or when I ordered tripas tacos for Stephen, my houseguest in Santa Barbara, which he thoroughly enjoyed until, three tacos in, he asked me what ‘tripas’ meant. Or when I made that terrible tuna-fish casserole filled with glorious portions of shredded cheese and Goldfish crackers for Amy, Jacob, and Allison so that they might become sensuously acquainted with a piece of my childhood. Or when Catelynn and I sat in that tiny four-seat kitchen overlooking the glittering ocean in Big Sur and were served sushi, omakase style, directly from the chef’s greasy, gentle hands, defining a shared moment of multisensory beauty.
These experiences fit into the fabric of my life in unique and highly meaningful ways. These experiences are mine, but you probably have some like it. The thing to notice is that these sorts of experiences would be inaccessible without contact with the provenance of food. They would not be possible in a world where all food was produced by a Star Trek replicator. This suggests that food automation threatens to erode an important source of human meaning.
Really, there are all sorts of concerns you might have about AI and robotics in the culinary sphere. Many of these have been identified by my colleague Patrick Lin. But for me, the erosion of meaning is worth emphasizing in discussions about technology because this kind of cost resists quantification, making it easy to overlook. It’s the sort of thing that might not show up on a cost-benefit of a tech CEO who speaks glibly about eliminating human cooking.
The point I’m making is not that we should reject automation. The point is that as we augment and replace human labor in restaurants, home kitchens, and other spheres of life, we need to be attentive to how the processes we hope to automate away may enrich our lives. An increase in efficiency according to quantifiable criteria (time, money, waste) can diminish squishier but no less important things. Sometimes this provides a reason to insist on an alternative vision in which humans remain in contact with the processes in question. I would argue this is true in the kitchen; humans should retain robust roles in the processes through which our food comes to be.
After my meal in Tokyo, I used my phone to find an elevated walkway on which to smoke. I took a drag on a cigarette and watched a group of men under an overpass producing music, in the old way, by a faint neon light. I could feel the fugu in my belly, and my thoughts flashed to my loves and hopes. One of the men playing a guitar looked up. We made eye contact for a moment. I took another drag. This is nice. I am happy.
Note: This material is based upon work supported by the National Science Foundation under Award No. 2220888. Any opinions, findings and conclusions or recommendations expressed in this material are those of the author and do not necessarily reflect the views of the National Science Foundation.
