Shenzhen-based LimX Dynamics shows off one of its humanoid robots.
Limx Dynamics
In robotics, advances in generative AI can help machines perceive and understand their environment, said Li Zhang, chief operating officer of Shenzhen-based LimX Dynamics.
About three months after joining the two-year-old startup, Lee said he lowered his expectations about how long it would take LimX to develop a humanoid robot that could not only work in factories but also help in households. to be able to.
Lee originally expected the entire process to take eight to ten years, but now expects some use cases to be ready in five to seven years. “After working for a few months, I saw how AI improved the capabilities of various tools,” he said in Mandarin, as translated by CNBC.
“It has accelerated our entire research and development cycle,” he said.
Companies are rushing to the opportunity. OpenAI itself is supporting humanoid robots. The beginningWhile Elon Musk's Tesla is developing its own, called Optimus.
Electric car giant BYD invested in Shanghai-based Agibot last year, months after its founding, according to Pitchbook.
And at a higher level, Chinese state media published a photo in November of Chinese President Xi Jinping viewing a humanoid robot at an exhibition center during his first post-pandemic trip to Shanghai. The robot is developed by Fourier Intelligence.
Before humanoid robots reach households, as LimX eventually intends, factories could be a profitable, connected landscape in which to deploy them.
According to Stanford's latest AI Index report, China overtook Japan as the world's largest installer of industrial robots in 2013, and now accounts for more than 50% of the global total.
The report said that electronics, automotive and metal and machinery were the top three sectors for the installation of industrial robots in China.
When it comes to completely replacing human workers, however, AI advancements alone are not enough.
Even if AI allows a robot to think and make decisions on a par with humans, mechanical limitations are a major reason why humanoids cannot yet replace human workers, said Li of LemX.
One of LimX's backers, Future Capital, has also invested in Pan Motor, a company that specializes in motors for humanoids.
Generative AI doesn't directly help robotic movement, pointed out Eric Xia, partner at Future Capital, an investor in LimX. But “advances in large language models could help humanoid robots plan advanced tasks,” he said in Chinese, as translated by CNBC.
Other LimX investors include Lenovo Capital.
As the cost per robot falls, the shift to factory robots may accelerate.
Steve Hoffman, chairman of FoundersSpace, a startup accelerator, said he is working with a Chinese startup called Fastra, which he expects could begin mass-producing robots within a year. He said he spent time this year in China teaching local businesses how to integrate generative AI.
“We've already received six orders from research institutes,” he said, adding that the startup aims to reduce the cost per robot by $50,000 to $100,000 through the rollout.
“If we can get to $50,000, we can sell a lot of robots,” he said, adding that the robot's batteries can be charged while operating 24 hours a day. “I can pay for a robot in a year.”
In pharmaceutical research, creative AI can reduce costs without reducing human labor.
“You don't save costs in our business by having fewer people. You actually save costs by having fewer experiments that fail,” said Alex Zavronkov, executive director and CEO board chairman of Ansilico Medicines, which has offices in Hong Kong. , New York and other parts of the world.
He notes how large pharmaceutical companies typically spend thousands of dollars to replicate a single molecule for testing — and will run a few thousand such tests per program. They claimed that with the help of AI, Insilico only needs to synthesize 70 molecules per program.
The company published a paper in Nature in March claiming that it had reached Phase 2 clinical trials of the AI-developed drug.