Pushing Boston Dynamics past 2027 adds 1.6 trillion won of losses at last year's rate

The delay costs money every year until the listing. Boston Dynamics lost 528.4 billion won last year, roughly 31% of the 1.7 trillion won it has lost since 2021 ^[3][4][1]. Hold that rate flat and the two years between the ruled-out 2027 window and Meritz Securities' 2029 estimate cost about 1.06 trillion won; three years, to 2030, cost about 1.59 trillion ^[6][2]. Cumulative losses on the morning of the first trade would then be near 3.3 trillion won ^[3].

The group is paying for that itself. Chung Euisun moved in July to lift his personal stake to 25% with an additional 120 billion won. That was part of buying out SoftBank's remaining holding at roughly 500 billion won, about $371 million, giving Hyundai Motor Group full ownership before the debut ^[14]. Hyundai has declined to name a listing date or a target valuation ^[7].

Money comes in today from Spot, the four-legged robot, and Stretch, the box-unloading machine ^[10]. Atlas demand named so far is captive: more than 25,000 units across Hyundai Motor and Kia plants, about 0.83 of a single year at the 30,000-unit annual capacity the company told a JPMorgan-hosted investor session it would have by 2028 ^[9][6]. Kim Hyun-su, a principal investment manager at IBK Asset Management in Seoul, is skeptical that target is reachable ^[8]. Outside the group, the record shows one small order: Immigration and Customs Enforcement plans to spend at least a million dollars on Boston Dynamics legged hardware to improve "officer safety" ^[11]. Last year's loss converts to about $435 million at the rate implied by the source's own 1.7 trillion won and $1.4 billion pairing, so the ICE spend is roughly two-tenths of one percent of it ^[5].

The price has not moved with the timetable. Boston Dynamics has been marked at around 30 trillion won privately ^[15]. Samsung Securities puts current value between 50 trillion and 100 trillion won, or 1.7 to 3.3 times that private mark ^[16][7]. IBK Securities models as much as 141 trillion won by the end of the decade on revenue nearing 11 trillion won, about 12.8 times the revenue it forecasts ^[16][8].

There is a second reading, in which the delay is a treasury decision. The group can fund Atlas internally, and each additional year of deployment history raises the price at which it eventually sells shares. Hyundai Motor stock is up 25% this year, having slipped from the high it set after the commercial Atlas reveal at CES in January ^[17]. Cryptopolitan reported that Unitree Robotics' shares fell below 500 yuan, erasing more than 240 billion yuan of paper value from the debut peak. It also reported that China's securities regulator afterwards told humanoid firms informally they needed recurring revenue, reduced losses, or actual technological innovation before attempting to list ^[12][13].

In my view an allocator should underwrite the loss run-rate and the 2029 or 2030 date, and treat 141 trillion won as a forecast resting on revenue that has not arrived, from a humanoid with no large-scale commercial deployment ^[2][16]. That view breaks if the 2026 loss comes in well below 528.4 billion won while Atlas capacity builds; that combination would mean the past five years were capital spending on a schedule ^[3].