A Robot (Sort of) Made My Lunch

(SeaPRwire) –   Earlier this week, I dropped by a lunch eatery in London where several people in their 20s were hard at work behind the counter. Just to their left sat something far from ordinary: a food-assembly robot that could one day put these workers out of a job.

I visited the location to check out a pilot program run by London-based startup Kaikaku AI. With roughly $1.8 million in funding, it is one of a growing group of startups that see recent breakthroughs in AI and robotics as a chance to cut out repetitive kitchen tasks, lower labor costs for food businesses, and—at least according to Kaikaku CEO Josef Chen—reduce prices for consumers while also improving the taste of food.

As someone who never shies away from first-hand reporting for the public good, I arrived with a healthy appetite.

I entered my order into an iPad and walked over to the robot. From above, it looked just like any standard food-assembly counter, with trays holding shredded carrots, onions, salad leaves, and sweet corn. I then crouched down and peered through a clear plastic window, where I saw funnels leading down from each tray to a conveyor belt below.

My empty paper bowl moved in short jerks along the belt, pausing under each funnel in turn. A cascade of spring onions came first, followed by a heap of mango chunks and edamame beans. Next, the bowl stopped under a device that looked like a downward-pointing revolver, with each chamber holding a different sauce. My requested chili-garlic sauce was dispensed as ordered. Finally, a handful of raw salmon chunks tumbled out of the last chute. The conveyor belt carried my bowl to a small lift, which raised it up through an opening at waist height, ready to be picked up by a human worker.

What came next was not automated at all. A human worker brought over a separate bowl of rice and sliced cucumber. (The machine can dispense rice, Chen explained, but it cannot yet do so as neatly as this restaurant prefers.) The worker mixed the ingredients portioned by the robot by hand, tipped them into the bowl of rice, then seasoned the mix by hand with sesame seeds and crispy onions. I dug in, and it tasted just like any other salmon poke bowl I’ve ever had.

Kaikaku is not the first company to try automating salad assembly. Back in 2021, chain Sweetgreen acquired a startup called Spyce Kitchen to pursue a nearly identical project. The idea never gained traction, and Sweetgreen sold off the venture last year. But according to Chen, a mix of current factors gives Kaikaku a real shot at success: robotics parts have gotten cheaper, new innovations in food-safe 3D printing have emerged, and human labor has grown more expensive. Kaikaku uses proprietary machine learning systems that leverage recent AI advances to accurately weigh and dispense food, Chen says. The company’s machine, called Fusion, can theoretically process 360 bowls per hour—far more than even a team of human workers working at full speed. What I witnessed during my visit, which took place during an admittedly low-footfall lunch hour in London, was far slower than that maximum rate; roughly one bowl per minute.

After my experience eating a lunch that was (sort of) prepared by a robot, I could have left the restaurant feeling skeptical. It’s true that assembling a salad is far simpler than cooking a complex meal. And most of the work required to get the food to my plate was still done by humans: farmers, fishermen, drivers, suppliers, and of course the worker who mixed everything together. Many of these workers already use machines, such as automatic crop planters that have automated the tedious work of sowing edamame beans by hand. What difference does one more machine make, simply because it is added at the final step of the process?

Instead, I left thinking about what the future holds for the worker who handed me my bowl. When I asked her how she felt about the robot, she had nothing but good things to say. “It’s definitely a good idea,” she replied. “Right now, we struggle a lot to find good team members. So extra help like this really makes our work go faster.” But speaking with Chen made it clear that Kaikaku’s end goal is the full automation of the type of work she does. “Instead of having 20 people in the kitchen slicing vegetables, or three chefs whose only job is timing how long pasta cooks, all of that will be automated,” he told me.