A startup founded by two ex-Unit 8200 officers closed one of the largest seed rounds in robotics history and immediately opened 100 of its machines to anyone with a browser.
The money came fast. Enigma, a physical AI company that didn't exist a year ago, emerged from stealth on July 27 with $71 million in seed funding and a live public experiment that lets anyone on the internet take control of a real robot. No waitlist. No partnership agreement required. Just a browser and robots.online.
Index Ventures and Ribbit Capital led the round, joined by Conviction Partners and a constellation of angel investors drawn personally from OpenAI, Anthropic, DeepMind, xAI, Cognition, and Wiz. That list is worth pausing on. These aren't passive check-writers hedging a sector bet. These are researchers and operators who spend their days at the frontier of AI development, and they collectively decided that what Enigma is doing with physical robotics is the next place smart money belongs.
Enigma's thesis is direct: the problem with robots isn't the hardware, and it's not raw AI capability either. It's the interface. Programming a robot today requires specialist engineers, detailed task scripts, and hours of configuration. Enigma wants to replace all of that with something closer to turning a volume knob. Its hardware-agnostic AI models layer onto any robot, and its control interface is designed so that operating one feels intuitive rather than technical. As TechCrunch noted when the story broke, the company's bet is that human-machine communication is the real adoption bottleneck, not the machines themselves.
To test that thesis at scale, Enigma built more than 100 proprietary robots and put them in hangars in Israel and California. Then it opened them to the world. At robots.online, visitors can direct these machines in real time, watching them draw pictures with a paintbrush, fence with swords, and pick up and mix flasks in simple chemistry experiments. It's genuinely novel. No company has done this before at this scale, and the stunt is more than marketing. Every session generates human-robot interaction data that feeds directly back into Enigma's models, meaning the public is, in effect, helping train the system while playing with it.
The founders built toward this with unusual speed. Jonathan Jacobi finished a computer science degree in high school. He then became Microsoft's youngest-ever employee at 17, before co-founding the company with Gal Niv, who began hardware hacking at 10, completed a four-year degree in a single year, and became Unit 8200's youngest cyber-operations manager. The two met competing in hacking competitions as teenagers and later shared a bunk bed during their time in the elite Israeli intelligence unit. In eleven months they assembled a team that spans AI researchers, Math Olympiad medalists, physicists, and low-level systems engineers. It reads less like a founding story and more like a recruiting poster for technical ambition.
Where the sector's money is actually rotating #
Enigma's raise lands at a moment when humanoid robotics investment has become almost feverish. Figure AI, Physical Intelligence, and 1X Technologies have all attracted nine-figure rounds in the past eighteen months, and every major chip and cloud company has taken at least an exploratory position in physical AI. But most of that capital has flowed to companies building their own robot hardware, competing on form factor and payload and degrees of freedom. Enigma's bet is different. Hardware-agnostic AI models mean it can ride the wave of whoever wins the robot manufacturing race without needing to win it themselves. That's a materially different risk profile, and it may explain why investors from right across the AI field were willing to write personal checks at seed stage.
The company is already piloting with partners in entertainment, retail, and health, though it hasn't named them publicly yet. Funding will go toward growing the research and engineering team, scaling compute, and expanding real-world deployments. Enigma declined to comment on revenue or a timeline to Series A.
Frankly, the robots.online platform is the most interesting signal here. A controlled enterprise pilot tells you a company has customers. An open public experiment with 100 machines running live tells you a company has conviction. Enigma is generating data at a pace no closed beta could match, and every hour robots.online stays online is an hour its competitors aren't collecting the same thing.
Also read: The model isn't what makes one AI coding agent 10x cheaper than another • Together AI hits $8.3 billion valuation as enterprises flee closed AI models for open source • Eliyan raises $145 million at a $1 billion valuation as AI's real bottleneck moves between chips