Technical
FAQ

Household generic AI tools can answer general questions, but fall short in industrial use cases where accuracy, structure, and safety matter. The key difference is data scale: generic AI solutions may handle single documents well, but struggle when you need to process multiple documents spanning thousand, or even hundreds of thousands of pages. Our Assistants are purpose-built to manage this vast amount of specialized data, ensuring reliable, context-aware guidance. You can find detailed comparisons on our Tech Assistant, Smart Assistant and Technology pages. 

Even with an in-house team, building this from scratch takes serious time, money, and trial-and-error. We’ve already figured out the main concerns, edge cases, and user experience cases for machinery use.  You might be interested on our blogs about the 2025 MIT study about failed generative AI pilots and what it takes to build a conversational AI assistant for machinery. 

Yes. Both Assistants support all major languages, and can be localized to virtually any language you require. 

The system runs online by default, with all processing done in the cloud. Fully offline use is possible if the necessary hardware is available on the machine. Tech Assistant requires less processing power than Smart Assistant for offline use.

All data is handled securely in compliance with GDPR, UK Data Protection Act, CalOPPA, and other relevant legislation. Customer and usage data is stored on EU or US servers depending on individual customer requirements. You can view our full data storage and privacy policy here.

We don’t use your data to train third-party models. Data pipelines are isolated and auditable. You can view our full data storage and privacy policy here.

We support a wide range of formats and sources: PDFs, Word, XML, JSON, spreadsheets, databases, and many more. A short integration project may be needed to handle unique formats or integrations to your systems.

No. We handle the preprocessing for you.

Yes. The system can connect with your existing tools. A short integration project may be needed to handle unique formats or integrations to your systems.

Absolutely. Embedding is supported for seamless integration into your own UI or platform. We provide an API for flexible integration.

We will get a pilot up and running in 4-8 weeks, depending on the exact scope of the project. Moving from the pilot to a full deployment is quick by leveraging the insights gained during the pilot phase.

Setting the Tech Assistant up and testing takes a few working days, usually about a week.

Virtually any kinds of digital formats your technical manuals, spec sheets and documentation are in: PDFs, Word, XML, JSON, spreadsheets, databases, and many more.

Yes. These are parsed and presented clearly with titles and legends.

The answers are grounded in your approved documentation. Unlike generic AI solutions prone to hallucinations, spogen.ai clearly flags when information is missing or uncertain.

The system maintains machine-specific, and task-specific context to deliver precise, actionable answers, grounding them in approved technical material. Additionally, Smart Assistant can also leverage telemetry data, and inputs from integrated systems to provide richer, real-time guidance. 

Yes. Both of our Assistants support both voice and touch/typed interaction.

Yes. It is possible to create step-by-step instructions for specific queries, standardizing key procedures and ensuring guidance is accurate and free from AI hallucination.