“The physical world is where we have most of our problems because it is so complex and rapidly evolving that we cannot fully understand things,” said Brandon Barbello, co-founder and COO of Archetype. “We put sensors in all kinds of things to help us, but sensor data is too difficult to interpret. There is a potential to use AI to understand that sensor data – then we can finally understand and solve these problems.”
When I visited Archetype’s five-person team, currently working out of a cramped room in the Palo Alto office of its lead funder, venture capital firm Venrock, they showed me some illuminating demos that, they assured me, only gave a hint of Newton’s enormous potential impact. . They placed a motion sensor in a box and had Newton imagine that the container was an Amazon package with fragile cargo that needed to be carefully monitored. When the box fell, the display with the model reported the news that the package may have been damaged. One can easily imagine a shipment of vaccines where motion, temperature and GPS sensors are monitored to verify that they arrive with full effectiveness.
A key use case is using Newton “to talk to a house or chat to a factory,” Barbello says. Instead of needing a complex dashboard or custom software to understand the data from a home or industrial facility with sensors, you can let Newton tell you what’s happening in plain language, ChatGPT style. “You’re no longer looking sensor by sensor, device by device, but you actually have a real-time mirror of the entire factory,” says Barbello.
Naturally, Amazon – owner of some of the most digitally advanced logistics operations in the world – is one of Archetype’s backers, through its Industrial Innovation Fund. “This has the potential to further optimize the flow of goods through our fulfillment centers and improve delivery speed for customers, which is obviously a big goal for us,” says Franziska Bossart, head of the fund. Archetype also explores the healthcare market. Stefano Bini, a professor in UC San Francisco’s Department of Orthopedic Surgery, has been working with sensors that can assess recovery progress after someone undergoes knee replacement surgery. Newton could help him in his search for a single metric, perhaps from multiple sensors, that can “literally measure the impact of any healthcare intervention,” he says.
Another early Archetype customer is Volkswagen, which is doing some early testing of the Archetype model. Surprisingly, this isn’t about autonomous driving, although Archetype is keen to see its technology used for that purpose. One Volkswagen experiment involves a scenario in which a car’s sensors can analyze motion, perhaps in combination with a sensor on a driver’s person, to figure out when the owner returns from the store and needs extra help. “If we recognize human intent in that scenario, I can automatically open that back gate and maybe put my stuff in specially heated or cooled locations.” said Brian Lathrop, senior chief scientist at Volkwagen’s Silicon Valley Innovation Center. This mundane task, Lathrop believes, is just the beginning of what will become possible when AI can process large amounts of sensor data into human-centric insights. Volkswagen’s interests include the safety of people outside vehicles, as well as passengers and drivers. “What happens if you connect all those cameras of those millions of vehicles on the road, in parking lots and on driveways into a network?” he says, “If you get AI to look at all these data feeds, it opens up an incredible amount of possibilities and use cases.”
It is not difficult to imagine the dark side of a performance trillion sensor monitoring system that provides immediate answers to questions about what is happening at every location in the dense network. When I tell Poupyrev and Barbello that this seems a bit dystopian, they assure me that they have thought of this. Unlike cameras, they say, radar and other sensor data are more beneficial. (Camera data, however, is one of the sensor inputs that Archetype can process.) “The customers we work with focus on solving their specific problems with a wide range of sensors without compromising privacy,” says Poupyrev. Volkswagen’s Lathrop agrees. “When we use Archetype software, I detect behavior, not identity. If someone walks up to my wife and tries to grab her bag, that’s behavior you can detect without identifying the person.” On the other hand, there is evidence that the way people walk – something a high-quality radar could very well detect – is also as distinctive as a fingerprint. Just sayin’.