Precise. accurate. compliant.
Make content your competitive advantage. And your AI’s source of truth.
Discover how Author-it helps your team reduce errors, accelerate workflows, and deliver accurate, compliant content at scale, and feed every AI you build with a source it can trust.
Author-it AION FAQ
AION output is optimised for Retrieval-Augmented Generation pipelines. Each content component includes rich metadata, clear hierarchy, and resolved text - making it easy for retrieval systems to find the right content and for language models to generate accurate, grounded responses. This reduces hallucination and improves answer quality.
AION publishes content from Author-it's Library as structured JSON that preserves the full hierarchy, metadata, and relationships of your content. Any Book in Author-it can be published to AION format - no special configuration required. The output includes resolved text in Markdown, object IDs, timestamps, authorship, and taxonomy data.
No. Any existing Book in Author-it can be published to AION format without special configuration. Because Author-it content is already structured with taxonomy, metadata, and component-level organisation, it's inherently ready for AI consumption. AION simply unlocks that value as a publishing output.
AION is Author-it's AI-ready publishing output. It converts structured, governed content into clean, metadata-rich JSON purpose-built for consumption by large language models, RAG pipelines, chatbots, AI agents, and enterprise knowledge systems.
AION output includes object IDs, descriptions, template types, timestamps, authorship information, folder paths, resolved variable values, and image alt text. This metadata gives AI systems the context they need to understand, categorise, and accurately use your content.
AION output is platform-agnostic and compatible with OpenAI GPT, Microsoft Copilot, Google Gemini, Anthropic Claude, AWS Bedrock, and any system that consumes structured JSON. It's designed to integrate into any AI pipeline - from internal knowledge bases to customer-facing chatbots.
AI systems produce more accurate, reliable results when they work with structured, well-organised content. Unstructured documents force AI to guess at meaning, context, and relationships - leading to hallucinations and errors. Structured content gives AI explicit signals about hierarchy, terminology, and intent, dramatically improving output quality.
AION preserves the full governance chain in its output - version history, approval status, and authorship are all included. This means AI systems consuming AION content can trace every piece of information back to its approved source, supporting accountability and compliance requirements.
Author-it's architecture - structured components, rich metadata, taxonomy, and single-source publishing - is exactly what AI systems need to work accurately. This wasn't designed for AI specifically, but the principles of structured authoring that Author-it has championed for 25+ years are now the foundation of effective AI content strategies.
Yes. AION is the first shipped AI capability in Author-it, and the company is actively investing in AI across the platform. The focus is on practical, governed AI features that help teams work more efficiently while maintaining the accuracy and compliance standards that regulated industries require.
A PDF export strips structure - the AI receives a wall of text with no content type information, no hierarchy signals, and no metadata. A Markdown export may retain some heading structure but loses provenance, authorship, content type, and the organisational hierarchy the content sits within. AION preserves all of it: content type and template, position in the content hierarchy, modification history, authorship, library folder path, and resolved variable values. The AI receives context alongside the text, not just the text.
AION produces standard structured JSON, which any system that ingests JSON can consume. This includes RAG pipelines, LLM fine-tuning workflows, AI chatbots and agents, internal knowledge bases, copilots, and content delivery platforms such as Fluid Topics and Zoomin. Author-it does not require a specific AI platform or vendor. If your system ingests structured JSON, it can ingest AION output.
RAG (retrieval-augmented generation) works by retrieving the most relevant chunks of content from a knowledge base before an LLM generates an answer. When content is chunked from a PDF, the splits are arbitrary - a paragraph might span two unrelated topics, or a key sentence might be cut across a chunk boundary. AION chunks at the topic level: each chunk is a meaningful, typed unit with metadata attached. This produces more precise embeddings, better retrieval accuracy, and better-contextualised answers from the LLM.
No. AION is a publish target, not a workflow change. Authors write in Author-it exactly as they always have. The structured authoring they already do - component reuse, topic types, variables, conditions - is precisely what produces AI-ready output. Publishing to AION requires a Library Administrator to set up a publishing profile once. After that, it appears alongside all other publish targets and requires no additional effort from authors.