Do I need to know about AI to use Innguma?

What does Innguma’s AI do without you having to ask?

• It filters and groups duplicates, so that the same news item published in several outlets does not appear repeatedly.
• It classifies each signal by category, technology, entity, type of event, place and tone.
• It generates the newsletter summaries.
• It allows the content to be browsed by the dimensions it identifies, with AI Feed.
• It keeps each signal linked to the publication it comes from.

None of this requires writing instructions (prompts), choosing models or adjusting parameters.

What is worth knowing about AI, even if you are not technical?

Since 2 February 2025, the European AI Regulation has required organisations that use AI systems to address the AI literacy of their staff (Article 4)^[1]. Regulation (EU) 2026/1744, known as the Digital Omnibus on AI and in force since 27 July 2026, reworded that obligation: it now means adopting measures that support such literacy, without requiring a specific level^[2]. The Commission gives an everyday example: a company whose staff use ChatGPT to write or translate must inform them of risks such as hallucinations^[1].

For working with a watch supported by AI, that minimum comes down to four ideas:

1. AI can state something false with confidence. OWASP recommends training users on this limitation and on the need to verify^[3].
2. The source comes first. Before quoting a figure or a date, open the original publication.
3. A tag is a proposal. If it does not fit, it is corrected.
4. Not everything goes to any assistant. Sensitive information is only handled in the tools your organisation authorises.

Which part of the watch is not automated?

The part that adds the most value: judgement. Knowing which questions the watch must answer, which sources come in, what is discarded, and who each piece of information must reach. For example, ‘I want to know which competitors are filing patents on mould coatings, and which regulatory changes affect that market’ is a business question that only your team can formulate. AI processes; people decide what is watched, what is published and who receives it.

Why not set up the watch directly with an assistant?

You can ask a general-purpose assistant for news, but a reliable watch requires deciding and maintaining sources, repeating the search continuously, verifying each result and keeping a history. In addition, general-purpose assistants often fail at attributing sources: in the study coordinated by the EBU and led by the BBC, 31% of the responses on news had serious attribution problems^[4]. At Innguma that work is handled by the platform, with the source always linked. It is the sequence of search, watch, history and context.

And if you already use ChatGPT, Claude, Microsoft Copilot or Gemini?

Innguma offers connectivity through the Model Context Protocol (MCP), an open standard for connecting AI assistants to external systems^[5], currently compatible with ChatGPT, Claude, Microsoft Copilot and Gemini. From your assistant you can ask, in natural language, for a summary, a comparison or a briefing on your watch and, if your organisation enables it, manage sources, filters and publications. Your AI provides analytical capacity. Innguma provides context.

And if you have no time or team to run it?

Innguma Service runs the watch for you, on the same platform: do it with your team or with ours. You define what you need to know; Innguma’s team takes care of keeping the watch running.