Much of the world might have been gobsmacked by DeepSeek and AI’s Sputnik moment, but EnCharge AI has been thinking about how to make artificial intelligence more efficient and more powerful for years.
The company’s proprietary analog in-memory chip—developed initially out of a Princeton lab by CEO Naveen Verma and his team—is 20X more efficient and 9X more powerful than the highest-end digital chips on the market, according to the company.
EnCharge says it will revolutionize technology from drones to iPhones. And, turns out, they’re not the only ones that think so: This morning, EnCharge announced that it has closed a $100M Series B led by Tiger Global with participation from RTX Ventures, In-Q-Tel, and Constellation Technology Ventures, among others.
Small but mighty: Now you might be sitting here thinking—what the hell is an analog in-memory chip, and why invest so much money in something so tiny? Lucky for you, Tectonic asked Jonathan Morris, the company’s VP of government affairs and communications, to explain it.
“The really big challenge of AI is twofold,” he said. “You have challenges with memory and moving data back and forth. You also have challenges with just the sheer compute needed to do the processes on that data. Analog in-memory computing solves both of those problems simultaneously.”
- EnCharge’s technology integrates computing and memory in the same architecture, which means less data movement. It also means AI processing can happen on-board, not in a data center.
- Digital computing uses binaries (0 and 1), while analog computing works with continuous signals, making it better at processing real-time complex data.
- The company uses precise metal-wire switch capacitors instead of noise-prone transistors, which increases both efficiency and compute power.
- EnCharge employs existing semiconductor production materials and processes, meaning these chips are no more complicated or expensive to make than your standard digital chips.
Embrace the duality: EnCharge is pretty much the textbook definition of a dual-use company—these chips are positioned to enable more powerful, more efficient artificial intelligence in iPhones, laptops, and cars. But they could also supercharge AI-powered defense tech, like drones and C2 platforms, all without having to send data back to a center for processing.
“The efficiency breakthrough of EnCharge AI’s analog in-memory architecture can be transformative for defense and aerospace use cases where size, weight, and power constraints limit how AI is deployed today,” Dan Ateya, President and Managing Director of RTX Ventures, said in a statement.