Can Safeworld convince people that GenAI robots won’t hurt them? — SkimNews
Get the Tech newsletter
Daily tech — startups, AI labs, chips, the launches that shape the next decade. Free.
- Safeworld emerged from stealth with more than $12 million in seed funding led by Shine Capital and a16z Speedrun, joined by Box Group, Carnegie Mellon University Endowment, Innovation Endeavors, and SV Angel
- Dr. Ding Zhao, director of the Safe AI lab at Carnegie Mellon University, co-founded Safeworld to tackle the unpredictability of generative AI in robotics through advanced probabilistic risk evaluation and simulation-based testing
- Safeworld evaluates robotic control systems using simulations in platforms like Genesis or MuJoCo, inserting realistic human models to test thousands of scenarios including tripping, falling, and blind-corner encounters
- Kyle Wong, Safeworld co-founder and veteran startup executive, emphasized testing edge cases such as robots detecting humans carrying boxes or navigating factory blind spots to ensure collision avoidance
- Gritt Robotics, via CTO Vishal Dugar, is partnering with Safeworld to empirically validate safety for robots operating alongside human workers on industrial solar farms
- Jonathan Lai, partner at a16z Speedrun, stated that establishing an industry safety standard must happen now during robot design and deployment, not after household incidents occur
Why it matters: Robot makers face rising pressure to prove safety as generative AI introduces unpredictable behavior in unstructured environments. With $12M in backing and early partnerships like Gritt Robotics, Safeworld positions itself as a necessary third-party validator before mass deployment — turning safety verification into a scalable business where failure isn’t just costly, it’s unavoidable.
Ask SkimNews
