Enigma is entering the robotics race with a bet that the next leap in AI robots may depend as much on human interaction as on raw model capability. The research lab, set to emerge from stealth on Monday, has raised a $70 million seed round to test how people actually want to communicate with machines.
The startup is not only building robots. It is trying to learn what the right interface for robots should be before assuming the answer is text, voice, video, or direct controls.
A different starting point for robotics AI
Many robotics companies are working on foundation models that can perform tasks they were never explicitly trained to handle. The approaches described in the source article include studying millions of web videos, running computer simulations, and collecting motion data from humans using gloves with built-in sensors.
Enigma is taking another route. Instead of focusing only on model capabilities, the less-than-one-year-old startup wants to study how humans engage with robots. The goal is to use those interactions to develop intuitive interfaces and possibly a different kind of robotic brain.
That framing matters because capable robots are not automatically useful robots. If a person must spend too much time explaining a simple task, the machine can become more burden than tool. Enigma’s core question is therefore practical: what kind of interaction makes robot control feel natural?
The $70 million seed round
To fund that work, Enigma raised a $70 million seed round led by Index Ventures and Ribbit Capital, with participation from Sarah Guo of Conviction Partners.
The company was co-founded by Jonathan Jacobi and Gal Niv, longtime friends who met while competing in hacking competitions as young teens. They later served together in Israel’s elite Unit 8200, where they conducted cybersecurity research.
Jacobi is described in the source article as Microsoft’s youngest-ever employee, recruited by Wiz founder Asaf Rappaport during his time there. When Jacobi and Niv decided to launch a startup together last year, they chose AI for robotics despite lacking direct experience in the field.
Their view was that robotics held some of the most exciting unsolved problems in tech. To pursue that idea, they built a team from Israel’s tech ecosystem and community, including alumni from top AI labs, math Olympiad winners, and several people who were convinced to drop out of PhD programs.
The public experiment with more than 100 robots
Enigma’s first major public move is a large-scale online experiment. The company says anyone in the world will be able to interact online with more than 100 of its proprietary AI robots.
Those robots are housed in hangars in Israel and California. According to the source article, they can carry out tasks such as drawing pictures with a paintbrush, fighting each other with swords, and performing simple chemistry experiments by picking up and mixing flasks with liquids.
Enigma claims it developed both the robotic arms and their underlying models entirely from the ground up. The online experiment is meant to test not just whether the robots can act, but how people choose to direct them.
The company plans to compare different modes of communication, including text, audio, video examples, and tap, drag, and drop interactions. That range shows how open the question remains. The right interface may not look like a chat box, a remote control, or a training video alone.
Why the interface may be the product
Jacobi told TechCrunch that Enigma wants human-robot interaction to become completely effortless. He used a household example to describe the friction: if someone has to spend 15 minutes explaining to a robot where to put dishes, the task may stop feeling worth automating.
He also compared future robot control to adjusting a car’s volume knob. The point is not that robots and audio systems are similar. The point is that a good control gives users fast feedback, clear intent, and an easy path to correction.
In robotics, that kind of simplicity is difficult. A robot has to understand what a person wants, connect that intent to the physical world, and act without requiring exhausting instruction. Enigma’s experiment is designed to collect data from real interactions so the company can see which communication methods work best in practice.
That data could also influence how Enigma trains its foundational AI model. If the company learns how humans naturally guide robots, it may be able to shape its model around those patterns instead of treating the interface as an afterthought.
Still searching for the business case
For now, Enigma’s commercial direction remains undefined. Jacobi declined to share specific use cases for the company’s AI, though he said Enigma is already partnering with companies in healthcare, logistics, and entertainment.
Those partnerships suggest the company is testing beyond a research demo, but the source article does not identify specific products, deployments, or customers. The business use case remains open, which fits the company’s broader approach: learn first how people and robots should communicate, then build around that evidence.
Enigma’s wager is that the future of robot control will not be solved only by more capable models. It may also depend on finding the interaction pattern that makes people trust the machine, correct it easily, and keep using it when a task gets complicated.