Fourier Intelligence

1.1 Identity and business model

The assigned sources consistently place Fourier Intelligence's founding in 2015 and its operating center in Shanghai, while UpMarket confirms that it remained privately held as of March 2026. Its operating identity is broader than a single humanoid product: public profiles describe rehabilitation robots and exoskeletons sold into institutional care settings alongside the GR-series general-purpose humanoid platforms. That history supports a two-part business model in which rehabilitation equipment and related systems provide the nearer-term institutional channel, while humanoids serve as a newer platform and research proposition. KÜRE reports that the company formalized this split in 2024 as Fourier for general-purpose robotics and AI research and Fourier Rehab for rehabilitation technologies. This framing is credible as an operating description, not as proof of revenue mix, recurring economics, or a commercially scaled humanoid business.[CO001, CO002, CO003, CO004, CO005]

1.2 Leadership and governance

Founder identity is directionally clear but current role attribution is not. KÜRE names Alex Gu as the founder, while Tracxn separately lists Jie Gu or Gu Jie as founder and CEO and Zen Koh as co-founder and CEO. The records may reflect translated names, regional entities, co-leadership, or stale database entries; the assigned evidence does not resolve which explanation is correct. EquityZen adds Zhen-Hua Xu as CTO and Suyang Zhang as a director, but provides no effective dates or reporting lines. Tracxn's board section is also internally difficult to reconcile: it says there are two active members while displaying additional obscured or malformed names. Investors should therefore treat public leadership lists as a lead for verification rather than a complete governance map. Current board composition, decision rights, succession coverage, and founder key-person dependence require company-supplied evidence.[CO006, CO007, CO008]

1.3 Capitalization and operating scale

Public scale figures diverge too widely to combine. Tracxn reports $83 million raised across seven rounds, whereas RobotHub reports $229 million and includes a $114 million Series E in January 2025 plus a $43 million Series E+ in August 2025. RobotHub also attaches a $1.1 billion valuation to the August 2025 round. These are third-party database figures, not audited company disclosures, and the gap cannot be explained from the assigned evidence by currency conversion, primary-versus-secondary proceeds, or round labeling. The same caution applies to scale: independent profiles repeat a company-originated footprint of more than 2,000 institutions and hospitals across more than 40 countries, while UpMarket says more than 1,000 institutional healthcare customers. Tracxn reports 48 employees in late May 2026, but Humanoid Index reports 800-plus employees as of February 2026. The discrepancies may involve entity, business-unit, or methodology boundaries, but no source establishes a common denominator. Revenue, margins, cash, runway, customer concentration, and paid humanoid deployments remain undisclosed in the assigned pool.[CO009, CO010, CO011, CO012, CO013, CO014, CO015]

1.4 Milestones and current stage

Fourier's chronology shows a plausible progression from rehabilitation hardware toward humanoid platforms rather than an overnight category pivot. KÜRE dates a lower-limb exoskeleton prototype to 2016, the Fourier X1 launch to 2017, a bipedal humanoid-research transition to 2020, the Gen-1 prototype to 2022, GR-1 to 2023, and GR-2 to 2024. The same source dates the two-division brand structure to 2024, while RobotHub reports the later Series E+ financing in August 2025. The sequence supports technical continuity and a real rehabilitation heritage, but not commercial readiness of the humanoid line. ChoZan explicitly notes that CES 2026 exposure did not prove hospital-scale deployment. Accordingly, the defensible stage characterization is a private robotics company with an institutional rehabilitation footprint and a humanoid program still requiring deployment, service-economics, safety, and outcome validation.[CO001, CO005, CO010, CO011, CO016]

1.5 Exhibits

2.1 Market Boundary, Spend Taxonomy, and Status-Quo Substitutes

Evaluating Fourier Intelligence requires establishing clear boundaries between its established rehabilitation robotics business and its newer embodied AI humanoid initiative. In clinical rehabilitation, the addressable spend encompasses upper- and lower-limb robotic therapy devices, exoskeleton gait trainers, and connected clinical therapy management software such as RehabHub. This clinical boundary excludes non-robotic orthotics, general patient monitoring hardware, and robotic surgical systems. In this primary domain, Fourier competes against traditional physical therapy regimens, manual passive exercise equipment, and incumbent rehabilitation robotics manufacturers. Conversely, Fourier's humanoid program—anchored by the GR-1, GR-2, and recently unveiled GR-3 platforms—targets an aspirational market for care assistance, institutional companionship, and light industrial support. This secondary segment aims to address caregiver shortages in eldercare and hospital wards, yet it must contend with status-quo substitutes including specialized medical carts, fixed automation, and human staff. Conflating these two operational categories obscures commercial realities: Fourier Rehab operates as an established medical device provider with regulatory clearances and an installed base across 2,000 institutions in 40 countries, whereas the humanoid business remains an R&D venture seeking its first validated paid customer deployment.[CM001, CM002, CM003, CM009, CM010]

2.2 Addressable Market Sizing Across Rehabilitation and Embodied Care

Sizing Fourier's addressable opportunity requires distinct lenses rather than relying on a single broad total addressable market estimate. The global medical rehabilitation robotics market represents an established addressable market of $2.5 billion to $3.8 billion, expanding at approximately 18.5% annually driven by post-stroke care needs and expanding aging demographics across East Asia, Europe, and North America. Within this envelope, Fourier's serviceable addressable market centers on institutional neurorehabilitation and physical therapy clinics, where its proprietary compliant actuators and therapy devices address dedicated inpatient and outpatient budgets. In contrast, estimates for general professional service robotics span $28.0 billion to $35.0 billion, but this figure is overwhelmingly dominated by warehouse automated guided vehicles and commercial cleaning robots rather than humanoid embodiments. Promoters often cite China's national humanoid policy targets projecting an embodied AI market exceeding RMB 20 billion, yet these projections aggregate state subsidies and speculative manufacturing deployments rather than validated commercial revenues. Crucially, while third-party estimates suggest list prices for platforms like the GR-2 range from $100,000 to $150,000, public evidence confirms zero paying customers for Fourier's humanoid units. Underwriting Fourier therefore requires anchoring near-term valuation on verifiable rehabilitation equipment market share while treating humanoid market sizing as an unvalidated call option.[CM003, CM004, CM008, CM011, CM013, CM014]

2.3 Institutional Buyer, User, and Payer Dynamics

Purchasing dynamics for Fourier's products vary sharply across customer segments. In the core rehabilitation device segment, the buyer is typically a hospital procurement committee or rehabilitation department head, whereas the daily users are physical therapists, physiatrists, and recovering patients. Payer dynamics rely on national health insurance schemes, workers' compensation funds, or private medical insurers reimbursing defined physical therapy codes. The multi-million-yuan fleet installation at Malaysia's Sultan Nazrin Shah PERKESO Rehabilitation Center illustrates this institutional model, where a sovereign social security organization procured Fourier devices for a 55-acre facility designed to treat up to 700 patients. Here, adoption triggers center on verifiable patient throughput, therapist productivity gains, and objective recovery tracking. By contrast, in the emerging eldercare and care-companion humanoid market, prospective buyers are assisted living operators, municipal nursing homes, and research hospitals. These buyers face fundamentally different budget constraints, paying out of general operating margins or research innovation grants rather than established clinical insurance billing codes. Without standardized reimbursement or documented labor displacement metrics, humanoid adoption in care facilities remains confined to exploratory pilot programs and promotional demonstrations.[CM002, CM005, CM006, CM012]

2.4 Commercial Catalysts and Structural Deployment Bottlenecks

Adoption of Fourier's technology is driven by acute macroeconomic forces but throttled by stringent operational constraints. Growth is supported by accelerating demographic aging across East Asia and Europe, an acute global deficit of qualified physical therapists, and favorable industrial policies promoting embodied AI in China. Fourier's dual-market positioning attempts to leverage its clinical rehabilitation heritage, therapist workflow relationships, and human-robot physical interaction know-how to accelerate care humanoid acceptance. However, adoption constraints in patient-facing humanoid applications remain formidable. Clinical environments impose rigorous medical device safety standards, strict infection control requirements, and unresolved legal liability frameworks governing physical robot-patient contact. Furthermore, hardware limitations, including a three-hour battery duration per charge and unverified claims of continuous hot-swappable operation, impede autonomous 24/7 hospital ward staffing. While institutional partnerships with bodies such as Singapore's NHG Health and Japan's Nagoya University establish research dialogue, they do not bridge the gap to autonomous commercial deployment. Until Fourier demonstrates clinical outcome superiority and measurable economic returns in production settings, care humanoid adoption will encounter severe institutional resistance.[CM007, CM008, CM010, CM012, CM013]

2.5 Exhibits

3.1 Competitive Landscape & Category Segmentation

The competitive landscape surrounding Fourier Intelligence divides into two distinct categories: incumbent rehabilitation and exoskeleton manufacturers on one side, and emerging general-purpose humanoid robotics platforms on the other. Traditional medical peers such as Otto Bock, Cyberdyne, and Wandercraft focus strictly on orthotics, mobility assistance, and clinical physical therapy, benefiting from established hospital procurement channels and regulatory medical device clearances. In contrast, prominent humanoid developers such as Figure, Agility Robotics, 1X Technologies, and Apptronik are optimizing bipedal robots for industrial automation, automotive assembly, and logistics payload transfer. Fourier occupies an uncommon intermediate position, leveraging its established institutional healthcare presence while extending into bipedal humanoids through its 2024 corporate split into Fourier and Fourier Rehab divisions. This dual positioning allows Fourier to rank 19th among 158 active robotics competitors tracked by Tracxn, placing 3rd in total funding raised behind larger industrial conglomerates.[CP001, CP002, CP003, CP004, CP005]

3.2 Technical Specifications & Clinical Differentiation

Fourier's technical differentiation stems directly from its rehabilitation robotics heritage rather than generic humanoid locomotion. Clinical therapy robots demand high-fidelity force feedback, compliant physical interaction, and absolute safety around vulnerable patients, capabilities that Fourier developed across its ArmMotus, ExoMotus, and CycleMotus therapy lines. These capabilities translate into the GR humanoid series, most notably in the GR-2, which features 53 degrees of freedom, 380 Newton-meter maximum joint torque, and 12-DOF flexible hands, and the newly unveiled GR-3 Care Bot. The GR-3 incorporates 55 degrees of freedom, 31 distributed tactile pressure sensors, sound source localization, and automotive-grade foam padding. By prioritizing tactile sensitivity and human-robot interaction over raw payload capacity, Fourier differentiates its hardware from industrial peers whose rigid aluminum chassis and high-torque actuators are unsuitable for eldercare and patient contact.[CP004, CP006, CP007, CP011]

3.3 Commercial Pricing & Procurement Economics

Commercial pricing structures across Fourier's product catalog reflect a tiered hardware strategy spanning entry-level research units to high-end enterprise systems. According to Robot Price Index and RobotSourced catalog data, the lightweight Fourier N1 robot is positioned for sale between $16,000 and $40,000, creating an accessible entry platform for academic robotics labs and algorithm developers. The enterprise-grade GR-2 is quoted between $125,000 and $150,000 or higher, reflecting heavy-duty joint actuators and custom integration requirements for corporate research centers. The newer GR-3 Care Bot spans a wide published range from $27,500 to $120,000, varying significantly by regional distributor, sensor payload, and support package. However, verified commercial disclosures remain incomplete regarding recurring maintenance fees, software licensing, and specialized training packages, which typically represent a substantial portion of healthcare robotics total cost of ownership.[CP008, CP009, CP010, CP014]

3.4 Moat Durability, Switching Costs & Displacement Risks

Fourier's competitive durability is defended primarily by high institutional switching costs within its existing rehabilitation installed base of more than 2,000 hospitals across 40 countries. Hospitals that deploy Fourier's RehabHub interconnected therapy architecture invest heavily in therapist training, clinical protocol integration, and patient outcome tracking, creating friction against replacement by standalone devices. Furthermore, Fourier's commercial pivot toward the GR-3 Care Bot, highlighted at CES 2026, establishes early mindshare in eldercare and hospital assistance before industrial humanoid makers develop clinical competencies. Nevertheless, Fourier faces serious displacement risks over a 3- to 5-year horizon. Well-capitalized humanoid competitors backed by global technology giants could commoditize bipedal locomotion hardware and develop compliant end-effectors, eroding Fourier's hardware margins if the company cannot establish proprietary software and clinical therapy workflows.[CP002, CP003, CP004, CP011, CP012, CP013]

3.5 Exhibits

4.1 Revenue Streams and Commercial Monetization

Fourier Intelligence generates commercial revenue primarily through the sale and clinical deployment of medical rehabilitation robotics, supported by its RehabHub software and clinical integration platforms. The company's established rehabilitation portfolio includes upper-limb devices such as the ArmMotus M2 Pro and lower-limb exoskeleton systems like the ExoMotus M4, which are marketed to hospitals, specialized neurorehabilitation clinics, and public care facilities across more than 40 countries. In recent years, Fourier has broadened its commercial scope by introducing general-purpose bipedal humanoid robots, beginning with the mass-produced GR-1 platform and expanding to the GR-2 and GR3 series. However, while rehabilitation devices represent an active clinical revenue stream, the humanoid robotics segment remains largely in pilot delivery and academic research development stages rather than producing scaled commercial recurring revenue.[CI001, CI002, CI003, CI004]

4.2 Cost Structure, Unit Economics, and Margin Drivers

The cost structure of Fourier's rehabilitation hardware is driven by multi-axis actuator components, sensor arrays, specialized medical-grade materials, and rigorous clinical assembly standards. For advanced humanoid robots such as GR-1 and GR-2, bill-of-materials costs escalate due to the inclusion of 44 to 53 joint motors, high-torque FSA actuators, 12-DOF flexible hands, and embedded perception compute modules. Selling into institutional healthcare systems introduces extended procurement timelines, distributor margin splits, and ongoing clinical training requirements that elevate effective customer acquisition and service-delivery costs. Because Fourier maintains a private disclosure profile, precise component-level bills of materials, realized gross margins, and field maintenance costs remain confidential, necessitating conservative margin assumptions for underwriting.[CI005, CI006, CI007, CI008]

4.3 Commercial Traction Proxies and Metric Disclosures

Public indicators of commercial traction reflect Fourier's strong institutional presence in medical rehabilitation, where company statements report partnerships and service relationships with over 2,000 hospitals and rehabilitation centers worldwide. Tracking data across independent industry aggregators reveals significant divergence in operational scale: while specialized robotics indices estimate Fourier's workforce at over 800 employees globally as of early 2026, corporate directory services like Owler record estimated annual revenues below $1M and employee counts under 10, and Tracxn reports 48 employees in its Singapore entity. These sharp discrepancies highlight the opacity of private Chinese robotics ventures where entity-level filings in Singapore or offshore holding structures capture only partial operational footprints. Detailed revenue figures and financial statements remain accessible only via private secondary platforms.[CI002, CI009, CI010, CI011, CI012]

4.4 Capital Adequacy, Cash Runway, and Underwriting Verdict

Fourier Intelligence has demonstrated consistent fundraising capability, securing a reported $229M in cumulative equity capital across seven financing rounds since its 2015 founding. Recent capitalization includes an August 2025 Series E+ round of $43M led by Runyang Technology at an indicated $1.1B valuation, following a $114M Series E round in January 2025 backed by state-linked and institutional investors including Guoxin Investment and Prosperity7 Ventures. This capital influx provides substantial balance-sheet runway to sustain dual R&D tracks in embodied AI and medical robotics. Nevertheless, the commercial viability of humanoid robotics in unstructured healthcare and eldercare environments remains unproven at scale, with independent market analyses noting that repeatable procurement, maintenance operations, and service economics have yet to be validated. The absence of audited financials constitutes a key diligence blocker.[CI013, CI014, CI015, CI016]

4.5 Exhibits

5.1 Product Definition and Clinical Rehabilitation Robotics

Fourier defines its clinical product portfolio around patient physical rehabilitation and functional motor recovery. Originally founded in Shanghai in 2015 with a focus on robotic exoskeletons and orthoses, the company restructured in July 2024 into two dedicated operational divisions: Fourier, dedicated to general-purpose robotics and artificial intelligence research, and Fourier Rehab, housing its medical rehabilitation technologies. The clinical division delivers specialized robotic hardware targeting distinct anatomical zones, including upper-limb therapeutic manipulators such as the ArmMotus M2 Pro and WristMotus M1W, alongside lower-limb robotic exoskeletons such as the ExoMotus M4. These devices integrate into RehabHub, a unified clinical platform designed to connect multi-device training suites, automate biofeedback tracking, and standardize therapeutic exercise protocols across inpatient and outpatient care environments. Beyond conventional clinical stations, Fourier commercializes balance coordination systems and the MetaMotus Galileo virtual reality environment, which simulates activities of daily living to evaluate biomechanical stability. Company records indicate these rehabilitation platforms serve more than 2,000 hospitals and rehabilitation institutions across over 40 countries and regions.[CE001, CE002, CE006, CE011, CE012]

5.2 General-Purpose Humanoid Robotics: GRx Platform Evolution

Leveraging nearly a decade of clinical actuator and motor control engineering, Fourier expanded into general-purpose embodied artificial intelligence with the GRx humanoid robot series. Introduced in 2023 as Fourier's first mass-produced bipedal humanoid platform, the GR-1 stands 165 cm tall, weighs 55 kg, and utilizes 44 joint motors delivering a maximum joint torque of 230 Nm with an operational walking speed of 5 km/h. In 2024, Fourier announced its second-generation platform, the GR-2, scaling physical performance to a 175 cm, 63 kg form factor equipped with 53 joint motors, an improved internal cabling layout, and peak joint torque reaching 380 Nm. The GR-2 introduces 12-degree-of-freedom flexible dexterous hands and VR-based remote command integration to expand tactile manipulation. Official developer documentation has further referenced early development on the next-generation GR-3 platform. Across these iterations, Fourier positions the GRx series for academic research, medical assistance, and public service consultation, attempting to bridge clinical mechatronics into scalable embodied automation.[CE003, CE004, CE005, CE007, CE013]

5.3 Hardware Architecture, Aurora OS, and Developer Ecosystem

Fourier's underlying technology stack rests upon proprietary high-torque mechatronics paired with an open-source software and simulation infrastructure. At the hardware foundation, Fourier Smart Actuators (FSA) integrate high-power density motors, precision planetary reducers, and multi-axis force sensors capable of delivering up to 380 Nm of torque with closed-loop feedback algorithms. On the software layer, Fourier maintains the Aurora (Advanced Unified Robot Operation and Resource Architecture) operating system, providing a standardized messaging bus, hardware abstraction, and Docker-based containerized secondary development guidance. Developer tooling is distributed through public code repositories that host the DexHand SDK for programmatic communication with 12-DOF dexterous manipulators, URDF-to-MJCF model converters, and training environments integrated with NVIDIA Isaac Gym and Isaac Sim for reinforcement learning-based bipedal locomotion. This unified stack enables researchers and clinical partners to develop custom task policies while isolating underlying joint dynamics.[CE005, CE007, CE008, CE009, CE010]

5.4 Deployment Reliability, Technical Differentiation, and Diligence Risks

Fourier's primary technical differentiation stems from internalizing actuator design and possessing an established clinical deployment base, contrasting with pure-software humanoid entrants that lack hardware manufacturing know-how. The company's demonstrated ability to deliver 53-DOF mechatronics with 380 Nm torque reflects significant mechanical execution capabilities corroborated by growth equity partners like SoftBank Vision Fund and Prosperity7. However, substantial technical and diligence risks persist. Public documentation provides neither published mean time between failures (MTBF) figures nor continuous dynamic battery endurance data under loaded gait. Furthermore, while rehabilitation robots hold clinical footprints across Asian hospital networks, Fourier provides limited public disclosure regarding comprehensive Class II/III medical device clearance trajectories across Western regulatory jurisdictions (FDA 510(k), EU MDR), software cybersecurity penetration audits, or field failure logs, leaving enterprise reliability unverified.[CE004, CE005, CE011, CE013, CE014]

5.5 Exhibits

6.1 Institutional Healthcare Foundation vs. Emerging Embodied AI Piloting

Fourier Intelligence's commercial revenue and customer relationships divide sharply between its mature clinical rehabilitation business and its emerging humanoid robotics operations. Under the Fourier Rehab brand, the company serves more than 2,000 healthcare institutions, rehabilitation centers, and specialized therapy clinics spanning over 40 countries and regions. The economic buyers within this primary segment are hospital procurement directors, rehabilitation clinic administrators, and clinical research department heads seeking robotic automation to alleviate therapist shortages in neurological, orthopedic, and post-stroke physical recovery. In contrast, Fourier's general-purpose humanoid robotics division targets a speculative mix of university research laboratories, embodied AI developers, and institutional eldercare providers. These newer customer engagements remain pre-commercial or exploratory, relying on demonstration units and evaluation pilots rather than multi-year recurring SaaS or equipment service contracts.[CU001, CU002, CU003, CU004, CU005]

6.2 Clinical Installed Base Trajectory and Humanoid Delivery Status

Fourier's historical customer adoption trajectory began in 2015 with lower-limb rehabilitation exoskeletons, scaling through upper-limb therapy robots like the ArmMotus series and integrated clinic installations termed RehabHub. Independent industry registries document that Fourier's overall organizational stage sits at pilot delivery for its broader general-purpose humanoid systems, having progressed from the GR-1 release in 2023 to the GR-2 in 2024 and GR-3 demonstrations in 2025 and 2026. While the company reports over 2,000 clinical organizational deployments globally, tracking databases record zero disclosed recurring annual contract values, customer retention rates, or unit volume distributions across geographic territories. The adoption curve for the humanoid portfolio remains in developer-access and research-bench phases, where units are placed with university consortia and technical partners rather than driving high-volume enterprise rollouts.[CU002, CU006, CU007, CU008, CU009]

6.3 Institutional Collaborations, Clinical Validation, and Ecosystem Ties

Publicly verifiable customer deployments and strategic technical collaborations are concentrated in research and academic healthcare networks. In May 2026, Fourier formalized strategic agreements with Singapore's National Healthcare Group (NHG Health) and Japan's Nagoya University during the Global Summit on Rehabilitation and Assistive Technology. The partnership with NHG Health builds on joint work initiated in 2021, including a September 2025 clinical initiative with health startup injewelme evaluating GRx humanoids for contactless vital sign monitoring via remote photoplethysmography. At Nagoya University, Fourier's ArmMotus EMU upper-limb rehabilitation robot is deployed for clinical muscle synergy analysis. Beyond clinical trials, Fourier's 2026 'Technology + Ecosystem' initiative engages hardware and integration partners such as Singularity Intelligent Control, Shenzhen Yita Power Technology, Yihang Lingguang, and Shanghai Runke Juneng to test embodied AI models and industrial payloads. However, these technology showcases and joint research centers have not demonstrated enterprise-wide software subscription revenues or commercial production SLAs.[CU010, CU011, CU012, CU013, CU014]

6.4 Procurement Friction, Payer Dependency, and Commercial Opacity

Evaluating the durability of Fourier's customer relationships reveals material institutional risks. Hospital and clinical procurement cycles for capital-intensive rehabilitation hardware typically require 9 to 18 months, subject to regional medical device certifications, hospital tender laws, and variable healthcare reimbursement schedules across municipal markets in China, Southeast Asia, and Europe. Third-party industry analyses explicitly highlight that Fourier has not demonstrated repeatable procurement economics, customer retention rates, net revenue retention (NRR), or maintenance contract renewal margins for humanoid deployments. Furthermore, because humanoid implementations in eldercare and hospital assistance interface directly with vulnerable populations, safety liability protocols, staff training burdens, and regulatory clearances introduce steep friction that prevents pilot projects from converting into permanent, multi-unit enterprise contracts. Lacking disclosed top-customer revenue concentration figures or distributor terms, Fourier's commercial base remains exposed to public healthcare capital expenditure budget volatility.[CU004, CU007, CU015, CU016]

6.5 Exhibits

7.1 Regulatory, legal, privacy, and safety evidence is incomplete

Fourier's public site presents both rehabilitation systems and robots intended for human collaboration, while independent profiles describe a broader move from medical rehabilitation into general-purpose robotics. That scope creates plausible device, workplace-safety, privacy, and cross-border compliance obligations, but the assigned evidence does not provide model-level registrations, certifications, incident records, recall history, or security-control detail. The official landing pages link ethics and personal-information statements, which is evidence of a policy surface rather than evidence that operating controls have been tested. No assigned adverse result documents litigation or enforcement, so likelihood cannot be inferred from silence. Investment approval should require primary regulatory records and a jurisdiction-by-jurisdiction product classification map.[CR001, CR002, CR009, CR010, CR015]

7.2 Operational risk is driven by disclosure gaps and external dependencies

The operational risk is not a proven field-failure event; it is the inability to quantify reliability and supplier exposure from the available evidence. One independent analysis characterizes Fourier as stronger in body control than foundation models, describes semiconductors as primarily externally procured, and says specialized cybersecurity differentiation is not highlighted publicly. Those observations point to diligence priorities, not verified single-source dependencies. Public profiles also conflict on whether the humanoid business is at pilot delivery or commercial status. Without unit shipments, acceptance criteria, mean time between failures, warranty cost, uptime, or critical-component sourcing, neither production readiness nor residual safety exposure can be scored with confidence.[CR003, CR004, CR010, CR011, CR014, CR015]

7.3 Counterparty and financing risk remain unquantified

Two secondary profiles repeat Fourier's claim of serving more than 2,000 institutions and hospitals across more than 40 countries, but that footprint does not reveal payer mix, distributor dependence, renewal, procurement duration, or top-customer concentration. Financing evidence is similarly inconsistent: public profiles report cumulative funding of $83 million, $229 million, and $234.88 million using undisclosed aggregation rules. These are funding-raised figures, not revenue, cash availability, or runway. Reported valuation references also differ in currency, date, and methodology and cannot be merged into one mark. The key financial risk is therefore continued capital intensity without audited segment economics, not a quantified burn estimate.[CR005, CR006, CR007, CR008, CR015, CR016]

7.4 Monitoring priorities and thesis-break triggers

The highest residual exposures are the evidence gaps themselves. Before investment, Fourier should provide model-level regulatory approvals, safety and cybersecurity assessments, paid humanoid deployment records, field-reliability cohorts, audited segment financials, a current cash bridge, and customer and supplier concentration schedules. Thesis-break triggers should be measurable: failure to validate required certifications before target-market deployment; failure to reconcile commercial-stage claims with paid acceptance records; field reliability or safety performance outside contractual thresholds; a runway that cannot fund the next independently verified commercialization milestone; or concentration above the investment committee's approved limit without a credible second source or replacement customer. Until those tests are completed, risk severity remains conditional rather than numerically estimated.[CR006, CR007, CR014, CR015, CR016]

7.5 Exhibits

8.1 Investment Thesis, Commercial Roots, and Financing History

Fourier Intelligence was founded in 2015 in Shanghai with the objective of developing advanced medical rehabilitation robotics and exoskeletons. Over the subsequent decade, the company successfully commercialized a portfolio of upper- and lower-limb rehabilitation systems, culminating in a global clinical footprint serving more than 2,000 hospitals and healthcare facilities across 40 countries. This clinical hardware foundation differentiates Fourier from pure-play humanoid robotics peers that possess neither commercial revenue nor regulated deployment history. Between 2018 and 2025, Fourier raised more than $200 million in cumulative equity financing across multiple institutional rounds. Crucial milestones included a $62 million Series D financing in January 2022 led by SoftBank Vision Fund 2 with participation from Prosperity7 Ventures and Visionplus Capital, followed by a 2025 Series E round that established a reported post-money valuation anchor of approximately $1.1 billion. While this funding trajectory demonstrates sustained investor demand and strategic support from both international venture funds and municipal investment vehicles, the company has operated without disclosing audited annual revenues, cash burn rates, or gross margin figures, making entry valuation underwriting unusually reliant on private market signaling rather than audited fundamentals.[CV001, CV002, CV004, CV005]

8.2 Recommendation, Risk Assessment, and Valuation Stance

We conclude that Fourier Intelligence warrants a research-more recommendation with a high risk rating and a stretched valuation stance. The investment thesis hinges on Fourier's dual-market sequence: using its established rehabilitation distribution channels, clinical credibility, and high-performance actuator technologies to commercialize general-purpose care humanoids. In 2024, the company formalized this strategy by restructuring into two distinct operating divisions: Fourier Rehab, dedicated to clinical therapy robotics, and Fourier, focused on embodied artificial intelligence and humanoid platforms including the GR-1, GR-2, and GR-3 care bots. However, the anti-thesis carries substantial weight. Humanoid care robotics remains in early pilot testing with zero public evidence of recurring annual contracts, volume customer purchase orders, or positive unit gross margins. Furthermore, media releases highlighting defense and military robotics applications present severe geopolitical and regulatory friction, creating potential barriers to international clinical procurement across North American and European healthcare systems. At an effective $1.1 billion private valuation anchor, capital allocation cannot be justified until management provides auditable commercial traction metrics and clears dual-use perception risks.[CV003, CV005, CV006, CV007, CV008]

8.3 Peer Comps, Multiple Compression, and Scenario Modeling

Evaluating Fourier against global peers reveals a stark divergence between private humanoid multiples and commercial hardware realities. In the humanoid sector, bipedal robotics developers such as 1X Technologies have secured over $125 million in funding, whereas Agility Robotics and Apptronik have raised substantially larger capital pools to target high-throughput industrial and logistics automation. Fourier's total equity raised of over $200 million places it among well-funded global challengers, yet its stated focus on healthcare care-interaction rather than industrial material handling demands different operating economics. Traditional medical device and rehabilitation hardware comps trade at modest multiples of 2.5x to 5.0x next-twelve-months revenue due to procurement cycle friction, whereas AI-driven robotics platforms frequently seek venture valuations exceeding 10x revenue. In our base-case scenario, multiple compression toward 4x to 5x revenue yields an intrinsic valuation of $700 million to $900 million, indicating a 20% to 35% downside from the Series E headline anchor. In a bear-case scenario characterized by domestic medical pricing pressure and export-control restrictions, valuation could compress to $350 million to $450 million. Conversely, our bull case of $1.8 billion to $2.2 billion requires verifiable commercial adoption of the GR3 care bot alongside sustained 30%+ rehabilitation revenue expansion.[CV006, CV009, CV010, CV011]

8.4 Exit Readiness, Pre-IPO Liquidity, and Diligence Asks

Fourier Intelligence's exit readiness remains constrained by financial opacity, cap table complexity, and cross-border regulatory considerations. While pre-IPO shares and secondary market liquidity avenues have surfaced on platforms such as EquityZen, lack of standardized financial disclosures prevents institutional secondaries from clearing at transparent prices. The company's expansion to 800+ employees and establishment of 17 international joint research laboratories in cities such as Singapore, Chicago, Phoenix, Zurich, Melbourne, and Madrid demonstrate operational scale, but also imply substantial ongoing operational expenditures and monthly cash burn. Potential public listing paths on the Shanghai STAR Market or Hong Kong Stock Exchange will require rigorous scrutiny of revenue recognition policies, customer concentration, and state subsidy dependence. To transition from a research-more posture to a buy recommendation, prospective investors must execute four non-negotiable diligence asks: obtaining multi-year audited financial statements, verifying customer-level contract values and gross margins for the rehabilitation division, inspecting committed purchase agreements for GR series humanoids, and receiving legal confirmation regarding the strict civilian segregation of its robotics IP.[CV012, CV013, CV014]

8.5 Exhibits