Noetive officially came out of stealth on September 16, 2026, with a $41 million seed round led by Eclipse Ventures. Its ambition: bring a self-improving AI model to the physical operations of factories, logistics networks, energy sites, and data centers — a $30 trillion economy segment still largely underserved by AI vendors.
🔑 Key takeaways
- Noetive raised $41M in seed funding led by Eclipse Ventures.
- Target: industrial AI for factories, logistics, energy, and data centers.
- Product centers on a self-improving AI model the company calls a « brain ».
- Steuben Foods is among the first design partners.
- Team drawn from Meta, Alphabet, Amazon, and foundational model researchers.
An outsized seed round for industrial AI
The round brought together Eclipse Ventures at the helm, alongside Craft Ventures, The Westly Group, Swish Ventures, Factory, Incite Ventures, Gigascale Capital, Operator Partners, and Liquid 2 Ventures. Individual backers include Andrew Bosworth, CTO of Meta, Ahmad Al-Dahle, CTO of Airbnb, and the three co-founders of Decart.AI.
Incite was founded by Matt Rogers, co-founder of Nest, and Gigascale by Mike Schroepfer, former CTO of Meta. The $41 million figure is an unusually large seed for the industrial AI segment, where capital deployment still trails digital applications.
A « reference intelligence » for physical operations
Noetive was built inside Eclipse Ventures after the firm recruited Amir Frenkel, formerly a VP at Meta for nearly a decade and an executive at Alphabet and Amazon, as its first head of AI. The rest of the team comes from those same groups and from researchers who worked on foundational models.
The core product is described as « reference intelligence », built on a self-improving AI model the company calls a « brain ». Agents built on this model sit on top of tools already deployed at the customer and tackle the most complex end-to-end operational problems. Each deployment is expected to strengthen the model’s capabilities. On-site physical conditions are captured through a multimodal sensing capsule designed by Noetive.
« Most AI products built to date were designed for information work that lives on the internet. Running a factory or a distribution network means managing materials, machines, labor, and suppliers that won’t stop shifting. »
Amir Frenkel, Head of AI at Noetive
According to Noetive, the decisions that link all these elements still rely on data from siloed systems never designed to coordinate an entire enterprise. The self-improving model is meant to bridge the divide between digital data and the physical reality of production sites.
Early field feedback from Steuben Foods
The young company already counts design partners in manufacturing, logistics, energy, and data centers. Food and beverage manufacturer Steuben Foods is among the first users. Its CEO, Menachem Katz, delivered a first quantified testimonial on how the system is being used.
« Production planning has always been one of the most time-consuming tasks in our business, because even small changes ripple through the whole operation. Noetive now reads both the data from our systems and the conditions on the factory floor, and the work that used to be done monthly and took a week of planning is now done daily and takes minutes. »
Menachem Katz, CEO of Steuben Foods
The testimonial points to a significant shortening of planning cycles, moving from a monthly cadence taking one week to a daily cycle measured in minutes. No other public data has yet been published on gains observed at other partners.
Open operational challenges to monitor
Lior Susan, founder and CEO of Eclipse Ventures, summed up the macro stakes: « The $30 trillion physical economy has been largely left behind by AI. We saw an enormous opportunity to change that, and we built Noetive to seize it. »
Early market analysis points out that published evidence of customer gains is still missing; reported results come from a narrow circle of design partners. Several open questions will condition large-scale adoption.
| Open question | Description | Indicator to watch |
|---|---|---|
| Sensing capsule | Installation and maintenance of multimodal hardware | Total cost of ownership, payback period |
| Software integration | Compatibility with existing ERP and MES systems | Number of native connectors shipped |
| Decision governance | Conflicts between AI recommendation and planner judgment | Effective substitution rate of human decisions |
| Operational liability | Who bears the cost of an erroneous decision on a live line | Contractual framework and insurance coverage |
The main indicator to watch will be the composition of design partners in the coming months. If manufacturers, logistics operators, and data center owners outside the initial group start using the software in production and report planning cycles shortened from weeks to days, the self-improving AI model for physical operations will move from a funding story to an operational reality. Until then, the $30 trillion market remains a stated opportunity rather than a measured one.
Conclusion
With $41 million in seed, Noetive joins a small group of startups trying to industrialize AI for physical operations rather than for information flows. The bet is technically ambitious — a self-improving model fed by a sensing capsule deployed at every customer — and is being measured against a still poorly equipped $30 trillion market.
By the end of 2026, two scenarios will compete: a rapid widening of design partners validating the approach, or a stagnation confined to a pioneer circle lacking public evidence of effectiveness. The next funding round, the sectoral diversity of deployments, and the publication of independent benchmarks will be the real credibility markers of Noetive’s promise.
Sources
- SiliconANGLE – Noetive launches with $41M
- aixvz.cc – Noetive launches coverage
- TechStartups – Daily funding recap September 16, 2026
This article is published for informational and educational purposes only. It does not constitute investment advice. Do your own research (DYOR) before making any decision.
