Enigma Raises $70M to Make Robot Control as Easy as Volume

A significant challenge in artificial intelligence, the development of foundation models capable of performing tasks without explicit prior training, is being actively addressed by numerous robotics companies. Their methodologies are diverse, encompassing the analysis of millions of web videos, conducting extensive computer simulations, and gathering human motion data via sensor-equipped gloves.

Enigma, a research laboratory poised to publicly launch on Monday, is adopting a distinctly different strategy. Instead of solely concentrating on model capabilities, the less-than-one-year-old startup aims to investigate human-robot interactions. The objective is that these engagements will foster intuitive interfaces and potentially lead to an innovative form of robotic intelligence.

To support its ambitious mission, Enigma successfully secured a $70 million seed funding round. This round was led by Index Ventures and Ribbit Capital, with additional investment from Sarah Guo of Conviction Partners.

To explore optimal human-machine communication, Enigma is initiating a large-scale online experiment, inviting global participation to interact with over 100 of its proprietary AI robots. These robots, situated in specialized hangars in Israel and California, demonstrate capabilities such as painting with a brush, engaging in sword combat, and conducting basic chemistry experiments by manipulating and mixing flasks of liquids. Enigma asserts that both the robotic arms and their underlying AI models were developed entirely in-house.

Jonathan Jacobi, noted as Microsoft's youngest-ever employee (recruited by Wiz founder Asaf Rappaport), co-founded Enigma alongside his long-standing friend, Gal Niv. Their connection began during teenage hacking competitions, solidifying into a close friendship while they served together in Israel's elite Unit 8200, where they specialized in cybersecurity research.

When Jacobi and Niv decided to launch a startup last year, they chose to apply their technical expertise to AI for robotics. Despite lacking direct prior experience in the field, they perceived it as containing some of the most compelling and unsolved problems in technology.

They assembled a formidable team, which Jacobi describes as comprising some of their "smartest friends" from Israel's vibrant tech community and ecosystem. This group includes alumni from leading AI laboratories, mathematics Olympiad winners, and several individuals who were persuaded to leave their PhD programs.

Shardul Shah, a partner at Index Ventures, commented on their unique position: "There are a lot of robotics industry insiders participating in the next wave of embodied intelligence, but Jonathan and Gal are outsiders — they’re not roboticists. It affords them more room for originality." He added, "Someone who’s an insider may start with the capability of teleoperation or dexterity, but Enigma is starting from a very different place: ‘What’s the ultimate experience?’"

Jacobi conveyed to TechCrunch Enigma's aspiration to render human-robot interactions completely effortless.

"If you had to do your dishes and spent 15 minutes explaining to a robot where to put everything, everyone reaches the point of ‘Forget it, I’ll just do it myself,’" Jacobi illustrated. "Right now, everyone is at that point — even with the most capable models."

Jacobi envisions that manipulating robots should eventually be as intuitive as adjusting a car's volume knob. He posits that users would experience frustration if, instead of a simple dial, they were forced to adjust volume by set percentages without clear feedback on whether the outcome would be too loud or too quiet.

Enigma anticipates that the data collected from its online experiment will unveil an interface that serves as the robotics equivalent of the intuitive car volume knob. The startup's public testing phase will meticulously evaluate various methods for human-robot communication.

"We’re going to learn a lot about what is the right way to interact with robots," Jacobi stated. He pondered, "Do we want to just talk to them over text or audio? Do we want to show them an example as a video? Or maybe do we want to tap, drag, and drop?"

Jacobi acknowledged the highly open-ended nature of Enigma's experiment. The overarching hope is that by gathering real-world data on human-robot interaction, the startup will not only discover superior interfaces but also uncover more effective methods for training its foundational AI model.

While the company may eventually pinpoint the preferred modes of human-robot communication, its specific business applications remain somewhat undefined for now. Jacobi refrained from disclosing particular use cases for Enigma’s AI but confirmed that the startup is already forging partnerships with companies across healthcare, logistics, and entertainment sectors.