The questions we are asked first.

An AI visibility and narrative intelligence platform. It tracks how AI assistants describe your organisation, reads the public Telegram channels and Reddit communities that shape those answers, groups what it finds into narratives with evidence, and routes recommended actions to a named person for approval.

Eidoscore monitors answers from ChatGPT, Claude, Gemini and Mistral today, with engine and model version recorded on every run. Further engines, including Perplexity, DeepSeek and Kimi, are added as each integration is verified.

Public Telegram channels, Reddit communities and the AI answer engines. Web and news pages appear as the sources the engines cite. We never monitor private or closed groups, and more channels are added only where platform terms allow.

An executive read in plain language, a decision pulse of what to watch and what to fix, the insights collected, active high-risk items and new narratives, AI perception as a median with its range across scheduled runs, and open citation gaps each with a recommended action. A weekly brief every Monday at 06:00 and a monthly brief, both fixed once generated and exported as PDF or DOCX.

Ask a language model the same question twice and you get two answers. Eidoscore asks each question once per engine on every scheduled run and reports the low, high and median across runs, with every raw answer stored, so you can see when the engines disagree instead of being handed an average.

A language model classifies each item for topic, language and stance (hostile, questioning, neutral or supportive) and marks a risk type where one applies. Every classification carries a confidence score and a manual correction field. Misinformation flags are early warnings for a reviewer, never automated accusations.

An Eidoscore analyst reviews flags before the brief is generated. Approval of any action stays with a named person on your side.

Yes. You define a comparison set and see share of AI answers and position against it for the same questions, with the evidence behind each answer.

On infrastructure inside the European Union. Collected content, derived insights and reports stay there. The language models used for classification are listed as subprocessors in the security note.

Public sources only, EU hosting, and a default of monitoring topics and narratives rather than individuals. A Data Protection Impact Assessment is completed before any scaled monitoring of political content for a client.

No. Eidoscore drafts, recommends and routes. A named person on your side approves anything before it goes anywhere, and every decision is written to an audit log that cannot be edited or deleted.

Single sign-on, two-step verification, and admin and reviewer roles set per workspace.

A short call to confirm fit and agree the topics and channels that matter to you, then a working session in the Eidoscore workspace against your own questions, with real answers and sources.

Terms are agreed in the proposal. Founding cohort places run 90 days with no fee, in exchange for structured feedback and reference rights agreed per partner.

Yes. Each client gets its own workspace with its own topics, sources, reviewer roles and briefs.