Startup Funding News Today, September 11, 2026: Kinetix AI, AIDIN Robotics, Enigmata, Furo & More

Physical AI, enterprise control layers, and software tied directly to real-world operations are shaping startup funding this morning. The largest freshly announced financing in the latest 16-hour window comes from Shenzhen-based Kinetix AI, which disclosed more than RMB500 million across a series of Angel+ financings for its embodied-AI stack. In South Korea, AIDIN Robotics closed a KRW16 billion strategic round involving HD Hyundai Robotics and Samsung Venture Investment, putting fresh capital behind the sensors and robotic hands needed to move humanoids from demonstrations into factories and shipyards.

The money is also moving into less visible layers of AI. Nashville-based Enigmata raised $6.5 million to let AI systems operate on encrypted data without exposing the underlying plaintext. Japan’s nocall.ai raised about ¥800 million for generative-AI phone agents, while Beijing’s Yuedian Technology raised tens of millions of yuan to expand an enterprise-AI platform built around private company data, decision-making and execution.

Across today’s 10 selected companies, the disclosed capital includes more than RMB500 million plus another tens-of-millions-of-yuan round, KRW16 billion, ¥1.27 billion, and $24.5 million in U.S.-dollar-denominated financings, before counting EdTechX’s undisclosed Series Seed. Six of the 10 are directly tied to AI or robotics. Eight are headquartered in Asia.

That cutoff changes the story substantially. Positron AI’s eye-catching $875 million Series C was already public before noon ET on September 10, while Motive’s $1.3 billion growth financing was announced at 9:00 a.m. ET. Both remain prominent in today’s feeds, but neither qualifies for this report.

The Macro Environment: Capital Moves From AI Models to Systems That Touch Reality

Today’s largest fresh bets are not on another chatbot or frontier model. They are on technologies that help AI perceive, remember, decide, communicate, or act inside environments where errors carry economic consequences.

Kinetix AI is building the entire embodied stack around humanoid robots. AIDIN Robotics sits deeper in the supply chain, making force-torque and tactile sensors that allow robots to understand contact with physical objects. That pairing captures two approaches to physical AI: own the robot and its intelligence, or own a component that many robot makers may eventually need.

The same movement is visible in software. Enigmata addresses the privacy barrier preventing companies from applying AI to sensitive data. nocall.ai attaches generative models to a measurable business workflow: phone calls that lead to bookings, payments and customer retention. Yuedian is trying to make AI useful inside private enterprise systems rather than competing at the foundation-model layer. EdTechX is combining AI tutoring with human-led physical classrooms.

Outside AI, investors are still financing businesses where software, operational control, and physical supply chains intersect. Furo optimizes industrial batteries. Blue Farm is turning contracted tea acreage into production infrastructure. 9lione is trying to redistribute unused pharmaceutical inventory. Provilac is using institutional capital to scale a vertically controlled high-protein dairy network.

The pattern is increasingly clear: access to intelligence is abundant; control over the environment in which that intelligence produces an economic outcome is scarce.

Kinetix AI raises over $74.5 million across Angel+ financings to scale embodied AI

Shenzhen-based Kinetix AI, also known as 超维动力, has completed more than RMB500 million ($74.5 million)  in a series of Angel+ financings, with backing from Vertex Ventures, Fangguang Capital, Wanshi Capital and others, according to Chinese robotics media. The company plans to use the capital to accelerate its combined data, model and robot-development stack and move embodied-AI products toward commercial deployment.

One naming clarification matters: this is the Chinese robotics company Kinetix AI, not the Paris-based motion-AI company whose team joined Runway earlier this week. The Shenzhen company says it was founded in July 2025 and is building humanoid robots, dexterous hands, embodied-AI models, first-person training data, and supporting AI infrastructure.

The financing is notable both for its size and for what investors are funding so early. Instead of separating the robot body from the intelligence layer, Kinetix is trying to own the feedback loop between real-world data, model development, and hardware.

That is expensive, but it can also produce stronger defenses than software alone. Real robot data is harder to acquire than internet text, and improvements in manipulation depend on hardware, sensors, training data, and control systems evolving together. The open question is whether Kinetix can translate that technical breadth into repeatable deployments before capital requirements grow substantially.

Funding Details

Startup: Kinetix AI (超维动力)
Investors: Vertex Ventures, Fangguang Capital, Wanshi Capital and others
Amount Raised: More than RMB500 million across Angel+ financings
Total Raised: Not disclosed
Funding Stage: Angel+ series
Funding Date: September 11, 2026 announcement; individual closing dates not disclosed
Headquarters: Shenzhen, China
Sector: Embodied AI / humanoid robotics / physical AI
Valuation: Not disclosed

AIDIN Robotics raises KRW16B to give industrial robots a sense of touch

South Korea’s AIDIN Robotics closed a KRW16 billion strategic financing involving HD Hyundai Robotics and Samsung Venture Investment, according to current funding data. HD Hyundai separately confirmed that it invested KRW13 billion for an equity stake in the robotic-sensing company.

AIDIN develops force-torque and tactile sensors that let robots detect how much force they are applying when touching an object. It also builds robot-control and surface-processing systems. The Anyang-based company grew out of research at Sungkyunkwan University’s mechanical-engineering robotics laboratory.

The investor relationship is unusually concrete. HD Hyundai Robotics plans to combine AIDIN’s sensing technology with its own robot platforms, including a five-finger robotic hand designed for heavy-industry humanoids. The companies also plan to automate polishing, grinding, and other surface-treatment jobs in shipbuilding before targeting broader manufacturing and overseas shipyards.

That makes AIDIN a useful example of the picks-and-shovels thesis in physical AI. Humanoid manufacturers may compete fiercely over complete robots, but nearly all of them need better tactile perception and force control if machines are going to manipulate objects reliably. Component specialists can therefore benefit even if another company ultimately wins with a robot platform.

Funding Details

Startup: AIDIN Robotics
Investors: HD Hyundai Robotics, Samsung Venture Investment
Amount Raised: KRW16 billion
Total Raised: Not disclosed for all rounds
Funding Stage: Strategic financing; formal round label not disclosed
Funding Date: September 11, 2026
Headquarters: Anyang, Gyeonggi-do, South Korea
Sector: Robotics / tactile sensing / physical AI
Valuation: Not disclosed

Provilac raises $14M to scale India’s high-protein dairy category

Pune-based Provilac raised $14 million from Panthera Growth Partners in its first institutional funding round, after building the business for more than a decade without institutional capital. The company will use the funding to expand manufacturing, develop additional high-protein dairy products and enter new markets, including Bengaluru.

Provilac is not a conventional software startup. Its investment case rests on vertical control of fresh dairy production, processing, distribution and direct customer relationships. The company says it now delivers to more than 50,000 households per day across markets including Mumbai, Pune, Hyderabad, Delhi NCR and Jaipur.

That operating model creates both an advantage and a risk. Owning more of the supply chain gives Provilac tighter control over quality and product consistency, particularly in a category where consumers pay a premium for nutrition claims. But physical production, refrigeration, last-mile delivery, and inventory make growth far more capital-intensive than digital subscription businesses.

The round suggests investors are willing to finance consumer-health categories when the company has already demonstrated distribution and repeat demand rather than asking venture capital to discover whether a market exists.

Funding Details

Startup: Provilac
Investors: Panthera Growth Partners
Amount Raised: $14 million
Total Raised: At least $14 million of disclosed institutional capital
Funding Stage: First institutional round; formal stage not disclosed
Funding Date: September 11, 2026
Headquarters: Pune, India
Sector: Consumer health/food technology/dairy manufacturing
Valuation: Not disclosed

nocall.ai raises ¥800M to turn generative AI into a phone operator

Tokyo-based nocall.ai raised approximately ¥800 million in Series A funding, taking cumulative funding above ¥1 billion. JAFCO and Archetype Ventures co-led the financing, with existing investor Coreline Ventures also participating.

nocall.ai automates business phone conversations using generative AI. The service, launched in late 2023, is being deployed across automotive, financial services, recruiting and IT workflows, where calls often have a direct connection to bookings, collections or customer conversion.

The company reports measurable results from customers. Automotive marketplace MOTA reduced same-day appointment cancellations from roughly 15% to 5.7% after automating confirmation calls, while Racoon Financial reportedly improved the rate at which collection calls secured payment commitments by three to four times. Those figures come from nocall and its customers rather than independent audits, but they show how voice agents are moving from novelty demos toward ROI-based purchasing decisions.

That distinction is important. Voice AI becomes more defensible when it owns telephony integration, workflow logic, customer context, and outcome data. The model speaking on the phone may become commoditized; the system that knows what should happen before and after the call is harder to replace.

Funding Details

Startup: nocall.ai
Investors: JAFCO, Archetype Ventures, Coreline Ventures
Amount Raised: Approximately ¥800 million
Total Raised: More than ¥1 billion
Funding Stage: Series A
Funding Date: September 11, 2026
Headquarters: Shinagawa, Tokyo, Japan
Sector: Enterprise AI/voice agents/business communications
Valuation: Not disclosed

Enigmata emerges from stealth with $6.5 million in funding to let AI work on data it cannot see

Nashville-based Enigmata emerged from stealth with $6.5 million in seed funding led by Blockchange Ventures. Founded in 2024, the company is commercializing Cipher, a cryptographic system that lets AI models train, search, and analyze information while the underlying data remains encrypted.

The problem is becoming more economically important as AI spreads into banks, healthcare systems, insurers and other data-heavy industries. Those organizations may possess exactly the proprietary information that could make AI more useful, but exposing raw customer records or intellectual property to an external model creates security, compliance and governance risks.

Enigmata is effectively betting that privacy becomes an infrastructure layer in the AI stack. If companies can use sensitive data without converting it to plaintext during processing, datasets currently isolated could become usable for model training, search, and analytics.

The technical burden is high. Enigmata says Cipher can operate at production speeds and preserve model performance, but those are company claims and will need broader third-party validation. The company is currently working with selected enterprise design partners rather than presenting a mature mass-market deployment.

Funding Details

Startup: Enigmata
Investors: Blockchange Ventures
Amount Raised: $6.5 million
Total Raised: $6.5 million disclosed
Funding Stage: Seed
Funding Date: September 10, 2026
Headquarters: Nashville, Tennessee
Sector: Privacy-preserving AI/cryptography / AI infrastructure
Valuation: Not disclosed

Yuedian Technology raises tens of millions of yuan to build enterprise AI around private data

Beijing-based Yuedian Technology raised tens of millions of yuan in Pre-A financing, led by Infotech Ventures with participation from ShuiMu Venture Capital. The new capital will support model training, product development, and replication of deployments already tested with enterprise customers.

Yuedian grew out of Minglue’s knowledge-graph product line and has long used Palantir as a reference point for its product philosophy. Its software connects private enterprise data, knowledge, and AI agents so customers can move from finding information to making decisions and executing tasks inside existing systems.

The company’s reported commercial numbers are unusually strong for a Pre-A business. Yuedian says it has nearly 100 paying customers, has reached break-even, recorded annual compound growth above 300% in orders and revenue between 2023 and 2025, and generated more revenue in the first half of 2026 than in all of 2025. Those figures come from company interviews and have not been independently audited.

Even with that caveat, the underlying thesis is interesting. Enterprise AI becomes more defensible when the product knows how a company actually operates: its internal ontology, permissions, processes, expert knowledge and decision rules. That context does not arrive automatically with a stronger foundation model.

Funding Details

Startup: Yuedian Technology
Investors: Infotech Ventures, ShuiMu Venture Capital
Amount Raised: Tens of millions of RMB; exact figure not disclosed
Total Raised: Not disclosed
Funding Stage: Pre-A
Funding Date: September 11, 2026
Headquarters: Beijing, China
Sector: Enterprise AI/decision intelligence / agentic software
Valuation: Not disclosed

Furo raises $4M to make industrial batteries work harder

Munich-based Furo raised $4 million in fresh funding led by TQ Ventures, with participation from Neo, Sandberg Bernthal Venture Partners and Munich’s Center for Digital Technology and Management. The company develops software for industrial battery-storage systems and has already signed customers including Deutsche Bahn.

Furo’s software helps industrial companies use batteries to reduce electricity costs. That can involve deciding when to charge, discharge or reserve battery capacity against changing power prices and operating needs.

The interesting part of the model is that Furo does not need to manufacture or own the battery itself. It can capture value from the increasing deployment of storage by owning the optimization layer sitting between the physical asset and the electricity market.

The startup also illustrates a geographic shift in European company building. Its founders spent time in Silicon Valley but moved back to Germany to be closer to industrial customers, technical universities and energy-market expertise while continuing to raise capital from U.S. investors. Furo remains incorporated as a Delaware C corporation.

Funding Details

Startup: Furo
Investors: TQ Ventures, Neo, Sandberg Bernthal Venture Partners, CDTM
Amount Raised: $4 million
Total Raised: Not disclosed
Funding Stage: Not disclosed
Funding Date: September 10, 2026
Headquarters: Munich, Germany
Sector: Energy software / industrial battery storage
Valuation: Not disclosed

Blue Farm raises ¥350M to turn tea fields into matcha supply infrastructure

Hamamatsu-based Blue Farm raised ¥350 million in Series A funding, bringing cumulative capital raised to ¥460 million. Investors include AKaFUJI’s DeepBlue1 fund, Suzuyo Shoji, Kawata Construction-related LtG Capital Partners, Hamamatsu Iwata Shinkin Bank-linked capital, TOKAI Holdings and NOBUNAGA Capital Village.

Blue Farm is responding to a distinctly physical supply problem. Rather than buying matcha on the market when customers need it, its MatchaaS model secures tea fields based on annual customer demand and coordinates cultivation, production and processing. The company is now using the round to expand that model internationally.

Calling that model “as a service” should not obscure what Blue Farm is actually doing: committing to agricultural production capacity.

That can create a valuable supply advantage when overseas matcha demand rises faster than farms can respond. But it also pulls the startup deeper into weather, crop yields, labor, processing capacity, and other operational risks that software companies do not carry.

Blue Farm is therefore a useful example of venture capital backing supply-chain control, not simply technology. If reliable access to a constrained input becomes valuable enough, the production network itself can become the moat.

Funding Details

Startup: Blue Farm
Investors: AKaFUJI/DeepBlue1, Suzuyo Shoji, LtG Capital Partners, Hamamatsu Iwata Shinkin-related fund, TOKAI Holdings, NOBUNAGA Capital Village
Amount Raised: ¥350 million
Total Raised: ¥460 million
Funding Stage: Series A
Funding Date: September 11, 2026
Headquarters: Hamamatsu, Shizuoka, Japan
Sector: Agtech / food supply infrastructure
Valuation: Not disclosed

9lione raises ¥120M to redistribute surplus medicines instead of wasting them

Tokyo-based 9lione, pronounced Clione, raised ¥120 million in Series A funding led by Yokohama Capital, taking cumulative financing to ¥440 million, including previous equity investors and debt financing.

The company’s platform addresses a costly inefficiency inside healthcare: medicines that remain unused at one medical institution while another provider still needs the same inventory.

9lione is building a distribution network that moves surplus pharmaceutical stock to where it can still be used. New capital will support expansion of that network, development of products for medical institutions, and hiring.

This is the type of vertical software problem where regulation can eventually work in the startup’s favor. Pharmaceutical distribution requires traceability, institutional trust, and operational compliance. Those requirements slow growth, but once a network has earned the confidence of medical institutions, a generic marketplace is not necessarily a substitute.

The underlying economics are also attractive if 9lione can scale responsibly: value is created by reducing waste in inventory already purchased rather than manufacturing another product.

Funding Details

Startup: 9lione
Investors: Yokohama Capital led the Series A; prior disclosed backing includes Incubate Fund, MS&AD Ventures-related capital, DG Ventures, and others
Amount Raised: ¥120 million
Total Raised: ¥440 million including debt financing
Funding Stage: Series A
Funding Date: September 11, 2026
Headquarters: Nerima, Tokyo, Japan
Sector: Healthtech / pharmaceutical distribution / supply-chain software
Valuation: Not disclosed

EdTechX raises undisclosed Series Seed funding to pair AI tutoring with physical classrooms

Tokyo-based EdTechX raised Series Seed funding from PKSHA Algorithm Fund, though the company did not disclose the amount. The startup operates X Junior, an English-learning business for young children that combines AI-personalized instruction, home tablet learning and physical classrooms.

The product’s design is more interesting than simply adding a chatbot to education. AI handles repetition, progress tracking, assessment, and recommendations, while human staff provides encouragement and the social structure that helps younger students keep showing up.

EdTechX also uses its own physical classrooms as product-development environments, collecting learning and operational feedback before distributing the system through home teachers, educational businesses and other partners.

That gives investors a different type of data flywheel. The company is not simply measuring clicks inside an online course; it can observe how AI recommendations perform inside repeated real-world teaching sessions. The challenge will be proving that those learning gains remain consistent as the classroom network expands beyond locations EdTechX controls directly.

Funding Details

Startup: EdTechX
Investors: PKSHA Algorithm Fund
Amount Raised: Not disclosed
Total Raised: Not disclosed
Funding Stage: Series Seed
Funding Date: September 11, 2026
Headquarters: Tokyo, Japan
Sector: Edtech / AI tutoring/education software
Valuation: Not disclosed

What Today’s Funding Activity Reveals

The first pattern is that physical AI is moving money into both full-stack companies and component suppliers. Kinetix is attempting to control the robot, data, model, and hardware stack. AIDIN is specializing in the tactile and force-sensing layer. Both approaches can work, but they produce very different capital profiles. Kinetix needs substantial funding to build an integrated system; AIDIN could sell a common component across several robot ecosystems.

Second, AI investment is shifting toward control and context. Enigmata controls what models are permitted to see. Yuedian embeds company-specific data and decision logic. nocall.ai controls a communications workflow. EdTechX combines learning data with physical teaching environments. None is trying to win by owning the largest model.

Third, strategic investors are showing up where commercial deployment matters. HD Hyundai is not merely betting on robotics valuations; it wants AIDIN’s sensors inside shipyard automation. PKSHA can provide EdTechX with AI expertise and ecosystem access. Regional industrial investors around Blue Farm have ties to the supply infrastructure the company must build.

Fourth, several of today’s startups gain defensibility from physical networks. Provilac has production and last-mile dairy operations. Blue Farm secures agricultural capacity. 9lione needs relationships with medical institutions. Furo integrates with expensive battery systems.

Those networks slow scaling compared with pure SaaS, but they can also make competition less trivial. A rival cannot reproduce a contracted agricultural base, trusted pharmaceutical network or installed industrial-energy system by shipping an identical user interface.

Fifth, interpret today’s heavy Asian representation carefully. Eight of the 10 selected companies are Asian, but the report closes at 6:15 a.m. Eastern Time. That timing naturally captures Japanese, Chinese, Korean, and Indian announcements that arrived during their Friday business day while excluding U.S. rounds that will be announced later Friday.

Finally, the funding mix is overwhelmingly early-stage. Today’s report is therefore less about who already dominates a market and more about where investors believe new bottlenecks are forming.

Tech Funding Table

Strategic Takeaways for Founders and Investors

Own a bottleneck, not merely an AI feature. Kinetix owns scarce robot data and hardware integration. AIDIN owns sensing technology. Enigmata attacks encrypted-data access. The stronger investment cases today are built around something customers cannot obtain simply by switching foundation models.

Physical AI may create more winners below the robot brand. AIDIN is especially instructive. Sensors, hands, actuators, data infrastructure, and control software can serve multiple robot platforms. Founders do not necessarily need to build the next humanoid manufacturer to benefit from humanoid adoption.

Trust is becoming an investable AI layer. Enigmata’s encryption architecture and Yuedian’s emphasis on private enterprise deployments both reflect the same problem: organizations want AI capability without surrendering control of sensitive information. Security, permissions, and governance are becoming product features customers may pay for directly.

Attach AI to an outcome customers already measure. nocall.ai can point to bookings, cancellations and collections. Furo can point to electricity costs. Yuedian can point to manufacturing and decision cycles. AI products whose value maps directly onto an existing P&L line have an easier path to budget than tools selling generalized productivity.

Strategic capital is most valuable when it shortens deployment. HD Hyundai can provide AIDIN with factories and shipyards in which to prove its sensors. PKSHA can help EdTechX with applied AI expertise. Regional industrial investors can help Blue Farm build supply. The logo on the cap table matters less than what it removes from the commercialization path.

Vertical integration can create a moat and a balance-sheet burden at the same time. Provilac and Blue Farm gain control by moving deeper into physical production. That control can improve quality and supply reliability, but it also means inventory, equipment, labor and production risk consume capital. Investors need to distinguish defensibility from operational complexity.

Regulated networks can compound slowly but powerfully. 9lione’s relationships with healthcare institutions and Enigmata’s work with sensitive enterprise data illustrate why difficult customer onboarding is not always a weakness. Once trust and compliance are established, they can make customer relationships harder to displace.

Conclusion

September 11’s funding cycle is not another morning dominated by a billion-dollar American AI financing. In fact, enforcing the 16-hour cutoff deliberately removes the largest deals still circulating in today’s news feeds.

What remains may be more revealing.

Kinetix AI is raising heavily to connect robot bodies, models, and training data. AIDIN Robotics is financing the tactile layer robots need to work safely around physical objects. Enigmata is making sensitive information usable without exposing it. nocall.ai and Yuedian Technology are embedding AI inside specific enterprise workflows. Furo, Blue Farm, Provilac and 9lione are tying technology to batteries, agriculture, food production and pharmaceutical inventory.

The common thread is no longer intelligence by itself.

Capital is moving toward the systems that make intelligence usable in the real economy: sensors that touch the world, data you can trust, workflows tied to measurable outcomes, and physical networks that software alone cannot reproduce.