Robot Learns Empathy Working 24/7 in Hotel Lobbies

Forget kung fu moves and warehouse lifting. The next frontier in robotics is teaching machines to nod sympathetically when you're talking and sense when you need help.

IntBot Inc., a Sunnyvale startup barely a year old, has deployed humanoid robots named Nylo that work full shifts as hotel concierges in New York, Las Vegas, and Tulsa. They're multilingual, never take breaks, and they've learned something remarkable: how to make humans feel heard.

The secret sauce isn't fancy hardware. IntBot built social intelligence software that runs on off-the-shelf robot bodies, training it with thousands of hours of real human interactions in noisy, unpredictable environments like hotel lobbies and convention centers.

CEO Lei Yang says their robots use audio-visual fusion, combining what they hear with what they see, to figure out who's talking and what they might need. The result is subtle head nods, natural body language, and the kind of presence that says "I'm listening" without a single word.

At NVIDIA's recent GTC conference, Nylo greeted thousands of attendees as a navigation assistant. The robot wasn't following a script. Its emotions and movements were generated in real time based on the social context around it.

The Ripple Effect

IntBot just announced their IntEngine software now works with multiple robot brands, not just their own. That means hotels, hospitals, and businesses won't need to choose between different robot manufacturers. They can pick the hardware they want and layer on the social skills.

Three hotels already run Nylo robots 24/7 alongside human staff, handling routine questions so people can focus on complex guest needs. At The Nap York in Manhattan, Otonomous in Vegas, and a Marriott in Tulsa, the bots work every shift without complaint.

The company also showcased the first edge deployment of NVIDIA's Cosmos Reason-2 vision model, which lets robots understand crowded, complex spaces in real time without sending data to the cloud. That means faster responses and better privacy.

While other robotics startups chase perfect walking or delicate manipulation, IntBot is betting on something simpler but perhaps more important: teaching machines to exist comfortably in human spaces. Earlier robots like Pepper struggled in real-world noise and chaos, often reduced to novelty attractions.

By mastering natural conversation and social awareness first, IntBot wants their robots to become everyday coworkers rather than trade show curiosities. The future of helpful robots might not be about what they can lift, but whether they know when to make eye contact.