Robotics
Jun 6, 2026

The Gist
Columbia professor Yunzhu Li says simulation software is becoming the key to training robots safely before they operate in real warehouses and factories. Generalist AI raised $400 million to develop AI brains that can control multiple types of robots across different industries. Companies are focusing on making robot control systems more reliable and predictable, especially for safety-critical applications.
Today's Stories
- 1
Generalist AI raises $400 million to build universal robot brains
Generalist AI, a startup developing AI software that can control different types of robots, secured $400 million in funding at a $2 billion valuation on June 5. The company is creating what it calls 'physical AGI' - AI systems that can operate robots in warehouses, factories, and eventually homes without needing separate programming for each robot type.
Instead of buying specialized robots for different tasks, companies could eventually use one AI system to control multiple robot types, potentially reducing costs and complexity.
- 2
Columbia professor emphasizes simulation as key to robot training
Columbia University professor and SceniX co-founder Yunzhu Li highlighted in a June 5 interview that simulation software is becoming central to robotics development. Companies use virtual environments to train robots safely before deploying them in real-world settings like warehouses and manufacturing plants.
Robots working in your local Amazon warehouse or factory will likely be safer and more reliable because they practiced millions of scenarios in virtual reality first.
- 3
QNX pushes for more predictable robot control systems
Winston Leung from QNX discussed on June 5 why deterministic real-time systems (computing that responds within guaranteed time limits) are becoming more critical in robotics. These systems ensure robots react predictably and safely, especially in applications where timing matters for human safety.
Factory robots and autonomous vehicles will become safer because their computer brains will respond to emergencies in precisely measured timeframes rather than unpredictable delays.
- 4
GMEX Robotics develops integrated robot control systems
Jun Wu of GMEX Robotics explained in a June 5 interview how his company provides 'integrated terminal + brain closed-loop systems' - complete robot solutions that combine hardware and AI software. The approach contrasts with companies that focus only on AI software or only on robot hardware.
Businesses buying industrial robots may get more reliable systems because the robot body and AI brain are designed to work together from the start.
- 5
Humanoid robot demonstrates stair climbing without falling
IEEE Spectrum featured a June 5 video showing a humanoid robot from DEEP Robotics successfully running down stairs and recovering from near-falls. The demonstration highlights advances in balance control systems that help robots navigate complex terrain without human intervention.
Future household or workplace robots will be able to safely navigate stairs and uneven surfaces, making them practical for multi-story buildings and outdoor environments.
- 6
AI CEO proposes robot tax as companies replace entry-level workers
Andrew Yang, CEO of Noble Mobile, stated on CNBC's Squawk Box on June 4 that his company is 'replacing junior employees with AI' and called for a tax on AI and robots. Yang argued this tax revenue could help retrain displaced workers as automation spreads across industries.
Entry-level job seekers may face increased competition from AI systems, while governments consider new taxes on companies that replace human workers with automation.
What to Watch
The IEEE International Conference on Robotics and Automation (ICRA 2026) concluded June 5 in Vienna, where multiple companies debuted new robot platforms. Watch for commercial announcements from companies like GENISOM AI and others who demonstrated their technologies at the conference.
Sources
- Why deterministic real-time systems are more critical than ever in robotics
- Interview with Columbia professor and co-founder of SceniX Yunzhu Li: ‘Simulation is central’
- Interview with Jun Wu of GMEX Robotics: ‘We provide an integrated terminal + brain closed-loop system’
- Video Friday: Watch This Running Robot Not Fall Down Stairs
- GENISOM AI debuts deployable robotics platforms at ICRA 2026
- Robot Talk Episode 159 – Robot sensing and manipulation, with Maria Koskinopoulou
- AI CEO On CNBC: ‘We Are Replacing Junior Employees with AI.’ The Only Solution is a Tax on AI and Robots
- Generalist AI raises $400 million to scale robot intelligence platform
- Newly released Wuji Hand 2 at ICRA2026
- If you want to watch a pendulum learn to swing up
Share this with a friend
Send today's roundup to anyone who wants to keep up.
Get daily AI news free with AIToday
200+ AI sources, summarized in 1 minute. Email / LINE / Slack.
Sign up free