The best science and technology news from France
Provided by AGP
By AI, Created 4:26 PM UTC, May 18, 2026, /AGP/ – Understand Tech launched an on-premise AI platform on May 19, 2026, built for technical and regulated industries that need data to stay inside their own networks. The package includes ready-made apps, a no-code builder, and an NVIDIA GB10-powered appliance that customers install and control themselves.
Why it matters: - Understand Tech is targeting enterprises that want AI apps without sending sensitive data to public cloud services. - The platform is designed for technical industries and regulated industries where data control, domain precision, and compliance are core requirements. - The launch reflects a shift from chatbots to production applications that actually support day-to-day work.
What happened: - Understand Tech launched an AI Application Factory on May 19, 2026, in Montpellier, France. - The offer combines ready-made AI applications, a no-code App Builder, and a secure on-premise appliance called AI-In-a-Box. - The appliance is powered by NVIDIA GB10 and is installed inside the customer’s own network. - The customer owns the appliance and the apps, and the data stays inside the building.
The details: - The platform is already in production at large and medium enterprises. - The on-premise stack includes applications, the no-code builder, RAG, model orchestration, agents, and workflows. - No outbound calls are required, and the system has no cloud dependency in the on-premise setup. - Customers who do not need the appliance can use SaaS or customer-managed VPC deployments. - The platform, apps, and builder are the same across deployment options. Only the form factor changes. - The marketplace includes apps for technical industries and regulated industries. - Technical-industry apps include Test Case Generator, which turns 400-page specifications into structured, traceable, automation-ready test assets in minutes instead of weeks. - Standards Navigator lets engineering teams query a standards library and get cited answers, while surfacing contradictions and compliance gaps. - Sales Enablement Portal is coming soon and will turn a semiconductor company’s product catalog into an instant-answer layer for field application engineers. - Workbench and Compliance apps support engineering, certification, and quality workflows. - Regulated-industry apps include Multi-LLM Portals Platform, Assistants, and Chat with CSV. - Those tools are built for accounting firms, HR teams, legal practices, and advisory networks working with payroll records, client files, contracts, and regulatory submissions. - App Builder turns a plain-language description into a working AI application in minutes. - Users describe what the app does, what knowledge it uses, how it behaves, and who it serves. - App Builder supports hot-reloading previews and one-click version restore. - App Builder runs on the same RAG, agents, security, and deployment stack as the marketplace apps. - Understand Tech also offers REST APIs with Python and TypeScript SDKs for developers embedding the platform as an AI backend. - The company provided a product page at More information and a launch video at Watch the launch video.
Between the lines: - Understand Tech is betting that enterprise AI buyers want control first and flexibility second. - The appliance model gives the company a way to serve organizations that cannot use public AI services, while SaaS and VPC options broaden the market. - Packaging apps, a builder, and infrastructure together lowers the barrier for companies that want AI outputs without standing up an ML team.
What’s next: - Understand Tech said App Builder is one of three paths to production, alongside the apps marketplace and developer APIs. - The company is positioning the marketplace to expand as new use cases are packaged into customer-owned applications. - The launch points to more enterprise deployments where AI is embedded into workflows instead of used as a standalone chat interface.
The bottom line: - Understand Tech is selling enterprise AI as a controlled, installable system, not a cloud chatbot. - The pitch is simple: keep data inside the customer network, and still ship usable AI apps fast.
Disclaimer: This article was produced by AGP Wire with the assistance of artificial intelligence based on original source content and has been refined to improve clarity, structure, and readability. This content is provided on an “as is” basis. While care has been taken in its preparation, it may contain inaccuracies or omissions, and readers should consult the original source and independently verify key information where appropriate. This content is for informational purposes only and does not constitute legal, financial, investment, or other professional advice.
Sign up for:
The daily local news briefing you can trust. Every day. Subscribe now.
We sent a one-time activation link to: .
Confirm it's you by clicking the email link.
If the email is not in your inbox, check spam or try again.
is already signed up. Check your inbox for updates.