CMU Safe AI Lab director launches Safeworld with $12.2M seed to test GenAI robot safety
Scobleizer · x · 2026-10-06
Safeworld emerges from stealth with a $12.2M seed round co-led by Shine Capital and a16z Speedrun, founded by Ding Zhao, who directs Carnegie Mellon's Safe AI Lab, alongside Kyle Wong and Simo Rachidi.
- The problem: robots increasingly hand control to generative AI models, whose probabilistic behavior isn't predictable like traditional algorithms — how do you verify a new humanoid is safe?
- The product: robot safety testing and simulation software that lets teams replay hazardous scenarios (pedestrians stepping out, obstacles blocking sensors) in simulation instead of running dangerous tests around people
- Mission: improve robot safety for the next billion machines
- Zhao argues deployment needs both probabilistic risk underwriting for GenAI systems and public trust
Covered by TechCrunch.
More from Venture
- a16z's 7th Top 100 Consumer AI Apps adds revenue leaderboard; ChatGPT tops 1B MAU — NousResearch · 2026-10-06
- Munich robotics firm RobCo hits $1B+ valuation, doubling in 9 months — lukas_m_ziegler · 2026-10-06
- Tasklet raises $12.8M at $500M valuation to bring AI agents to enterprise — binarybits · 2026-10-06
- From $1k to $15k MRR by fixing his funnel instead of adding new channels — tibo_maker · 2026-10-06
- 19-Year-Old Student Asks How to Build a ChatGPT Plugin Hitting $100-200 MRR — wallstreetbest0 · 2026-10-06
- Teen prodigy Zain raises seed round led for ghostai, launches personal AI computer Core — kirbyman01 · 2026-10-06