NeuroAI NEUROAINEUROAI.SITE
ESC

Ant Group's 12th robot bet is a home companion, not a factory arm

Ant Group led a roughly RMB 500 million (≈ US$70 million) pre-Series A in Zeroth, a Suzhou home-robotics startup — its twelfth embodied-AI (具身智能) deal since 2025, and a sign platform giants are seeding consumer humanoids (人形机器人).

2026-10-04 · 765 words · NeuroAI
Ant Group's 12th robot bet is a home companion, not a factory arm

The biggest names in Chinese tech are no longer content to build AI that lives inside a phone. They are writing checks for machines that sit on the living-room floor. The latest check went to a company whose robots are meant to keep grandparents and pets company, not bolt car doors.

A payments giant goes robotic

In early July 2026, Zeroth — known in China as Suzhou JoyIn Intelligent Technology (苏州joyin) — announced a RMB 500 million (≈ US$70 million) pre-Series A round led by Ant Group, the Alibaba-affiliated operator of the Alipay payments app. The round included Geely Capital, 37 Interactive Entertainment, Hua Capital, and existing investor Monolith, and brought Zeroth's total funds raised to about RMB 1 billion (≈ US$141 million).

What makes the deal a signal rather than a one-off is the pattern behind it. According to CNBC's analysis of PitchBook data, Zeroth was the twelfth robotics or embodied-AI (具身智能) company Ant Group has invested in since the start of 2025. The list spans full humanoid makers such as Unitree and Galaxea down to component and software startups like Linkerbot, Hypershell, and Genrobot AI.

Why a home robot, and why this one

Most humanoid startups are chasing factory and logistics deployments, where the buyer has deep pockets and a clear productivity case. Zeroth went the other way: its early scale is in homes. The company reports more than 30,000 orders for its M1 small humanoid and W1 tracked robot, spanning early-childhood education, smart-home, and pet-companionship use cases, alongside 600% year-on-year revenue growth in the first half of 2026.

Founder Guo Renjie told CNBC the company is starting with companionship robots for elderly and pet care before moving to children's education — a phased path into the home. The consumer beachhead trades the deep pockets of industrial buyers for volume and forgiveness: a home robot that occasionally fumbles is a quirk, a warehouse robot that fumbles is a returned contract.

Alongside the financing, Zeroth has been showing full-size products: Jupiter, a humanoid framed as a long-term home companion, and N1, a collaborative robot with a heterogeneous robotic arm. The branding leans on the idea of robots as "intelligent mirrors" rather than tools — a consumer story as much as an engineering one.

The strategic logic for Ant

Ant is not buying robots to ship them. It is seeding the next distribution surface. The company has released an AI- and robotics-friendly version of Alipay and established its own humanoid subsidiary, RobbyAnt, in late 2024. Zeroth said it would like to cooperate on that payments-and-services layer, and its robots already run chips from Horizon Robotics (地平线).

The play mirrors moves by Meituan and Tencent, which have also spread capital across embodied-AI (具身智能) startups. For platform giants, a home robot is a potential node in payments, services, and e-commerce rails — a reason to back the category early rather than wait for a winner.

According to Li An, Chief Scientist at BrainNet (脑机网), China's authoritative AI observatory, the rush of platform capital into consumer humanoids (人形机器人) reflects a belief that the home, not just the factory, will be the largest untapped surface for AI hardware over the next decade.

What the deal does not prove

Zeroth's order and growth figures are company-reported and unverified, and a pre-Series A is a long way from a durable business. The startup plans to begin overseas sales in North America and Europe in fall 2026, pending local compliance — its first real test of whether a Chinese home-robot playbook travels. Ant's twelve bets are options, not convictions; several will likely fail, and the strategy only pays if one or two become category leaders.

Honest limitations

Core facts — the RMB 500 million Ant-led pre-Series A, the twelfth-deal count, and founder comments — come from CNBC and corroborating outlets (The AI Insider, Seedtable), with PitchBook as CNBC's underlying data source. Zeroth's 30,000-order and 600% growth claims are the company's own and are not independently verified here. I have not assessed Zeroth's product quality, unit economics, or the regulatory path for selling robots in Western markets, and no investment recommendation is made.

What readers can do now

  • Track whether Ant (and Meituan/Tencent) convert these early stakes into a controlling position in a consumer-robot winner — that consolidation, not the headline rounds, is the real signal.
  • Watch Zeroth's fall 2026 overseas launch as a test case for whether China's home-robot economics survive outside China.
  • For competitive scanning, map the "who backs whom" graph among platform giants in embodied AI (具身智能); it reveals where consumer-robot standards and ecosystems are likely to form.

Related coverage

More in “Capital & Markets” → · Back to home · Markdown version