The Zhitong Finance App learned that Agility Robotics will soon become the first listed company in the US to focus on humanoid robots, which is an important milestone for an emerging category that still needs to prove its worth.
The startup has previously completed a merger with special purpose acquisition company Churchill Capital Corp. XI and began trading on the NASDAQ on Thursday under the stock code “AGLT.” According to the statement, the company had about 465 million US dollars in capital before deducting expenses, including about 201 million US dollars of private equity investment led by Foxconn. Other prominent supporters include Nvidia (NVDA.US), Amazon (AMZN.US), and SoftBank Vision Fund No. 2.
Agility is one of many companies developing humanoid robots. These humanoid robots can walk upright or move on wheels. Until now, these machines are mainly used in commercial scenarios such as warehouses and factories. The Chinese company Yushu Technology went public in August in Shanghai, but Agility was the first company among its peers to go public in the US.
The Oregon startup, which has about 400 employees, is expanding its collaboration with GXO Logistics (GXO.US) to deploy its robots to more facilities, and is also carrying out a manufacturing pilot with AGCO.US.
CEO Peggy Johnson said in an interview, “I think what's more important is what happens next after AI leaves the digital world.” She was referring to the way artificial intelligence is being applied to robots and other hardware today. “Bringing intelligent robots to more workplaces is a real opportunity. When machines and humans actually work together, it will fundamentally change the boundaries of what is possible.”
Agility believes its newest humanoid robot, the Digit 5, will help fill labor shortages, particularly in warehouses. This new model, which was released in September, is designed to perform complex tasks such as loading materials onto equipment, sorting parts, and inspecting items. In contrast, the earlier Digit 4 could only take on simpler tasks such as stacking plastic boxes. The new models can also work safely in places close to people, while previous generations of products had to be operated across physical security fences, limiting their ability to take on human tasks.
Agility said Digit 5 has received a multi-year commitment order of more than 300 million US dollars. The company's humanoid robot has been used by 9 customers and is in active sales negotiations with more than 30 potential commercial customers. The startup has promised to deploy Digit at 12 customer facilities, and its goal is to build a pool of 100 potential customers in various industries such as automotive, retail, and pharmaceuticals by the end of 2027.
Citing a 2024 Deloitte study, Johnson said the number of unfilled job vacancies in the US manufacturing industry is expected to rise to about 1.9 million by 2033.
Previously the CEO of augmented and mixed reality company Magic Leap, she said, “The labor gap that Digit is good at filling is only growing. Many companies don't know how to get the manpower to complete these difficult to fill, high attrition rates.”
Johnson believes that as Wall Street adapts to this new category, there will still be an education process. But she said the company should be assessed using the same metrics as other automation technology providers, including high throughput and accuracy.
She said, “What I think is really interesting is that Digit 5 can be updated via an over-the-air upgrade (OTA), so we can teach it new skills. As these additional deployments gather more data, all of this will be transformed into a training flywheel.”
Currently, humanoid robots with legs like the Digit 5 present some unique challenges. In particular, they are not as predictable as similar devices with wheels and pose a greater safety risk to humans. Johnson said that the listing made her feel some pressure, “because we were the first to run, and will be the target of reference for other companies, not only those indicators, but also safety.”
But the team has learned a lot from operating in various environments in the past, she added.
She said, “Demonstrations can show what's possible, but actual deployment will teach you what works. We've learned a lot in the past three years. We can walk into these facilities, do real work, and get paid for it.”