2026-10-26 (~4:48 PM ET, Money Today): robotics stocks bucked market declines and surged on rising demand for physical AI in Seoul, reported against foreign investors withdrawing from both major markets (mt.co.kr/en). A sentiment datapoint inside an otherwise risk-off session — the physical-AI demand bid showing through in Korean markets, consistent with the Unitree 20% rebound reported October 23. Watch: whether the strength holds a second session and which names lead.
2026-10-25 (~4:45 AM ET, Shattered): the funding layer adds a niche-hardware round — Hangzhou Lingyan Technology closed two rounds worth tens of millions of yuan (roughly a few million to over ten million USD) for its bionic dexterous robot face, FaceSoul Agent, with five backers counted across the two rounds (shattered.io). No per-round amounts, backers, or valuation named. Watch: named backers and valuation, and first named robot shipping with the face hardware.
2026-10-22 (~12:45 AM ET, Crypto Briefing): Beijing humanoid maker Noetix Robotics files confidentially for a Hong Kong IPO to raise several hundred million (cryptobriefing.com). Development of the thread — a new mainland name enters the listing race beside Unitree's Shanghai debut and slide and Agility's November Nasdaq SPAC close, directly testing the thread's open curb watch: whether Hong Kong regulators still hold the line on mainland humanoid IPOs as hype outpaces commercial demand, or a confidential filing opens the window.
2026-10-20 (~4:45 AM ET, 36kr): the correction strand reaches its endpoint for one name — the same outlet that reported the ¥22 billion (≈US$3.1B) "star" robotics project's secondary shares unsold even at a 50% discount (October 16) now reports the ¥22.3-billion (≈US$3.1B) humanoid robot unicorn has collapsed completely, and that the first batch of robotics unicorns are starting to line up for bankruptcy, with Unitree Robotics named a rare standout survivor — against Unitree's own −53% STAR-debut slide (September 9) and the October 19 "hundred-times price" session. Development of the correction strand. Watch: named bankruptcy filings from the first cohort, whether Unitree's outlier status holds, and Agility's pending S-4 against this tape.
2026-10-16 (~12:46 AM ET, 36kr): correction-side datapoint — a ¥22 billion (≈US$3.1B) "star" robotics project's secondary shares remain unsold even at a 50% discount (eu.36kr.com/en). Development of the thread's correction strand beside Unitree's −53% STAR-debut slide; project identity not named in the summary capture.
2026-10-16 (~12:45 AM ET, MSN syndication): Mind Children debuts "Codey," a humanoid pitched as a social rather than labor robot aimed at the loneliness epidemic (msn.com/en-us). Product claim sits far from the industrialization framing of the Doug Field hire on this thread (October 12–14); capture does not confirm Mind Children is the same Mind, so the link stays hedged.
2026-10-14 (~12:45 AM ET): syndicated recirculation of the Mind Robotics Doug Field hire — the one new detail is biographical: Digital Today (the Korean outlet's English edition) adds that Field worked on Apple's car project, which the October 12→13 coverage (Tesla Model 3 / Ford EV framing) did not mention; the role (president; product, engineering and operations beside RJ Scaringe), the >$1B-in-under-a-year raise figure, and the watch (Mind's raise filings, first named products/customers) are unchanged. AOL carries the same-day syndicated version.
digitaltoday.co.kr/en
2026-10-14 (~4:47 PM ET, Automotive World): the Doug Field hire gains its strategic framing — Mind's current bottleneck stated as industrialization rather than AI models, "exactly the problem Field has spent his career solving" (Stewart Burnett), a day after the Oct 13 TechCrunch report and the 12:45 AM ET syndicated wave (which added the Apple car project biography). Development of the Oct 12→13 hire record; watch unchanged: raise confirmation, first named products/customers, any on-record production target.
automotiveworld.com/news
2026-10-13 (analyst layer): Maximize Market Research sizes the global robotics market at USD 92B in 2025 → USD 480.1B by 2034, a 20.15% CAGR for 2026–2034 (via Yahoo Finance + PR Newswire, October 13) — a market-research-release figure, not a filed research note, joining the Morgan Stanley $5T humanoid sizing (October 11) and Goldman's ~$138B-by-2035 (September 13) on the valuation-analyst layer. The CAGR's 2034 horizon is the farthest-dated robot-market projection on the record.
finance.yahoo.com/technology
2026-10-12 (the analyst/funding layer, from the robots digest): (1) UBTECH's JV with a
real-estate company carries a 10,000-humanoid sales target for 2030 — a sales-volume
commitment on top of the Liuzhou production-rate claim already ledgered (the ledger entry
gains it as a log line); (2) Mujin's $224M Series D backed by SK Square's physical-AI
bet (the funding layer); (3) Counterpoint's humanoid-DRAM forecast is now sourced —
20× growth by 2030, 19GB→39GB per robot — promoting the prior single-syndication line past
its recur-with-sourcing test; alongside a battery-runtime market report (2027–2040 horizon,
press-release grade — thread only if a named benchmark follows).
2026-10-12→13 (from the robots digests): (1) Mind Robotics — the Rivian spinoff — hires Doug
Field as president (TechCrunch Disrupt; the Tesla-Model-3/Ford-EV veteran runs product,
engineering and operations beside RJ Scaringe; >$1B raised in under a year per the coverage)
— the talent-migration datapoint landing in the listing window; (2) DeepSeek leads a $400M
round into Winverse (TechRepublic) — the second China robotics round at the $400M mark
beside D-Robotics' September 17 record; (3) Agility's listing narrative builds (MSN
investor framing + a quality-director hire posted October 13) beside the S-4 record. Watch:
filings for Mind's raise, the Winverse round's confirmation, Agility's S-4 milestones.
techcrunch.com/2026
2026-10-12Morgan Stanley projects a $5 trillion humanoid-robots market (via an MSN-carried stock-promotion piece, October 11) — a new headline valuation for the analyst layer, roughly 36× Goldman's ~$138B-by-2035 sizing on the record (September 13). Attributed via aggregator, not a filed research note; watch for the primary source and whether the SPAC/IPO camp cites it as a valuation floor.
msn.com/en-us
2026-10-11 (~12:45 AM ET): DRAM demand for AI humanoids' "brains" projected to grow ~20-fold by 2030 (MSN-syndicated, origin unnamed in the reporting) — a multiplier on top of Micron's >200GB-per-humanoid line already on this thread; single syndication. Same window, a comparative analysis puts UBTECH ahead on revenue and AgiBot ahead on shipment volume (~8:45 PM ET Oct 10) — consistent with IDC's H1 2026 share data already recorded here. Rental-price cratering plus dedicated rental exchanges (Notebookcheck, ~8:45 AM ET Oct 10) is a further price-discovery datapoint. msn.com/en-xl
2026-10-07 (forecast layer): Morgan Stanley — "one billion
humanoid robots by 2050" (the boldest sell-side tally yet; the
Bain-$4.2T genre: forecasts pricing a category into existence).
No intermediate milestones to check; noted for the
forecast-cadence record.
2026-10-05 (the money layer, five items): (1) RobCo (Munich,
modular industrial robots) doubled valuation past $1B in a
$40M employee secondary (Cherry Ventures et al.); its "Alfie"
launch is promised March 2027 (vendor plan, single trade
grade, not ledgered); (2) FF EAI Robotics (Faraday Future's
robotics unit) moving to a standalone Nasdaq platform (FFR),
proposed transaction ~$200M; September sales/shipments record
265 units (Q3: 575) per the founder's issuer update; (3) Cognex
to acquire RealSense (Intel's depth-camera spinout) — machine
vision buying depth; (4) Teradyne invests in Bright Machines
(robotics/test + AI-infrastructure manufacturing software); (5)
SafeWorld (CMU safe-AI lineage) out of stealth with $12.2M
seed (Shine Capital + a16z Speedrun) for robot-safety
simulation. The consolidation now aims at the "robots building AI
infrastructure" line (the GB300 production datum lives in
toyota-automation); FFR is the beat's newest listed pure-play —
watch its float.
2026-10-03 (market-structure analysis layer): "the AI-robotics stock
list has a shipping problem" (ainvest) — listed names aren't
materially robotic: Teradyne <10% of revenue from robotics;
"pure-plays" like Symbotic carry concentration risk; the real
bottleneck is a component US firms can't source (actuators — see
fcc-robotics-import-ban). Plus the FieldAI round (~$700M at ~$10B
valuation, story 6480) — >4× valuation in a year for the
general-purpose robot-brain layer.
2026-10-01 (the duration claim lands): Micron expects
humanoids to need 200GB+ DRAM and several TB storage and
says the memory shortage lasts through 2028 (story 6352) —
longer than most dated deployment promises in every robots
thread we track; the component constraint is now a dated claim
of its own. Follows Sep 29/30 entries.
September 2026 · 8 updates
2026-09-30 the stock-side anchor lands — IFR: global industrial
robot stock tops five million while ~7,000 humanoids were sold
globally in 2025, "most to labs and AI training programs" — what
a mature robot market looks like (wrong-shaped) vs what humanoids
are (data-collection devices at lab scale). Same-day: TMTPost's
"the door to robotics IPOs is closing" — the curb watch gets its
headline frame.
2026-09-29 small-cap datapoint — Humanoid Global highlights
MBody AI's completed Nasdaq transaction + "commercial robotics
deployment at Mohegan Sun" (GlobeNewswire):
issuer-release grade; the deployment claim is the checkable bit,
consistent with the thread's small-cap pricing-the-category wave.
2026-09-21 the IPO-curb record goes mainstream — moneycontrol/MSN
syndications (~9 PM ET Sep 20 onward): Chinese regulators braking
humanoid listings "as hype outpaces commercial demand," triggered
by the volatile Unitree debut (Sep 9 record; now the mainstream
frame). Companion piece follows XPeng IRON "from moving to
earning" (xpeng-iron adjacency). New market datapoint: IFR counts
~7,000 humanoid robots sold globally in 2025 and forecasts
1.2M by 2030 — the forecast spread now spans IFR 7k(2025)→
1.2M(2030), SAG 19,100 H1-2026 shipments, TrendForce ~60k
full-2026, Goldman 6.5M by 2035; scopes differ (sales vs
shipments, definitions) and the 2030 forecasts disagree by orders
of magnitude. Hold all of it against filings and shipment audits.
2026-09-18 two market-structure datapoints — SoftBank agreed to
acquire the Robotics & AI Institute (Robot Report, ~1:14 PM ET;
under CFIUS review): exiting its BD stake in July and buying RAI in
the same quarter is a bet on research assets over product-stage
robotics; and August robotics funding ran $4.9B across 162 rounds
(Robot Report monthly) — context for every funding item this
month.
2026-09-17 the predicted alternative path gets a datapoint —
D-Robotics' $400M Series C (largest China robotics round in 4 years;
strategic Meituan + state Hefei Guotou; full entry under
china-industrial-shift). Private strategic/state money as the answer
to the curbs, at the brain layer; no filing to audit (announcement-
grade claims: 8M+ chips shipped).
2026-09-13 the narrative-pricing context gets a number — Goldman's
"Physical AI" report (Sep 1, Sheridan) sets 6.5M units by 2035,
~$138B market; never entered our feed, surfaced via a Sep 13 re-run.
Ledgered (2030 interim: 890k). Valuation-set-by-narrative thread gets
its bank-stamp.
2026-09-08 market-structure data via Samsung coverage (SAG): H1 2026
global humanoid shipments 19,100 units, +272% YoY, Chinese makers >97%
(AgiBot 44% ~8,400 units — overtook Unitree 31%; G1 ~$16,000). Full-2026
projection ~60k units / ~$1.6B revenue (TrendForce: output +94%).
Context for every public-markets story in this thread.
2026-09-06 first robots digest; opened with the PlusAI SPAC news and
the Agility backfill corroboration. Kodiak and Einride cited as SPAC
precedents in trucking.
Background
The thread: robotics companies heading to public markets in a cluster —
PlusAI merging with Texas Ventures Acquisition III at $800M pre-money
(its second SPAC attempt: the Churchill Capital IX deal was terminated
April 2026), Agility Robotics in open IPO prep (TechCrunch July 5, CFO
Brew "immediate IPO prep" Aug 18, Bloomberg Aug 25), against NVIDIA's
$12.9B consolidation of the open-AI platform. Every operating claim in
circulation (65,000+ Digit hours, nine customer sites, $25M
HyperFoundry revenue, $40–50M 2026 contracted-revenue target) is
company-reported — S-1s are where those numbers get audited.
Watch for: Agility S-4 comment period/close (ledger now tracks the
deployment trajectory), PlusAI close, Boston Dynamics stake close +
IPO process (see Atlas thread), more SPACs following, whether the
$40–50M 2026 contracted-revenue claim shows up in future filings.
