Blog
Everything starts with the source.
RAG, documents, credibility, sovereignty: our analysis, sourced like our answers.
Knowledge
AI in document management: what it adds to your DMS, and what it does not replace
Search by meaning, cited answers, scans and emails read, fact sheets extracted: what AI really brings to a document management system, and what to ask before buying.
4 min read
Credibility
AI hallucinations: the problem starts with the sources
AI assistant errors often come down to sources: what the EBU-BBC study, the Stanford study and Damien Charlotin’s database show, and what to check.
3 min read
Credibility
The Deloitte report in Australia: what a reference check would have caught
In 2025, a Deloitte report for the Australian government cited sources that did not exist. What the case reveals, and the check that catches it before delivery.
3 min read
Sovereignty
Eight questions to ask any AI supplier
Data law, choice of model, pricing, exit: eight questions to check that an AI supplier leaves your organisation free to change its mind.
4 min read
Credibility
Audit rigour for AI: the finding as the unit of proof
Sufficient, appropriate, traceable, reproducible: what the audit standard of evidence asks of a finding produced with AI, and what regulators now expect.
4 min read
Sovereignty
Sovereignty: a model’s nationality matters less than you think
Sovereign AI means keeping the ability to choose and switch models: where processing runs, the law that applies and RAG count for more than the flag.
3 min read
Knowledge
Five triggers that make a document speak
Deadlines, external shocks, forgotten commitments, volume, memory: the five situations in which a document should reach you without you looking for it.
3 min read
Knowledge
What dormant knowledge costs
Studies redone, clauses forgotten, know-how gone with a colleague: what the knowledge sleeping in documents costs, and which part can be measured.
3 min read
Knowledge
Document management and AI: two different questions
A DMS tells you where the document is; AI tells you what you know. Why one builds on the other, and three questions to ask of the AI layer.
3 min read
RAG
Five requirements for a serious RAG system
Ingestion, metadata, hybrid search, passage-level anchoring, structured data: the five places where the quality of a RAG system is built.
3 min read
RAG
RAG is not dead: what giant context windows really show
Lost in the Middle, NoLiMa, Context Rot: why a million tokens of context did not make search pointless, and what is bringing RAG back.
3 min read
The pilot
Start a pilot.
A scoped pilot, on your own documents and data, to measure what Oxebo changes in your deliverables.
- ScopeA scoping workshop: the case, the uses, the sources to connect.
- DeployOn the hosting of your choice, with our team.
- DeliverA first sourced deliverable, and the measure of time saved.
Supported deployments, open to a limited number of organisations.
Your request has arrived.
We will get back to you within two working days to set up the scoping workshop.