A research hub for building AI-native companies—curated know-how and an advanced research agent.
Knowledge repository
Global research. Local context.
Access a proprietary and extensive repository of AI research across strategy, engineering, governance, and more.
Inform local decisions with research, policy and regulatory resources from Australia, New Zealand and other country-specific sources.
Our AI agents continually discover and enrich research, giving your agent a curated knowledge base it cannot access through web search alone.
Specialist research agent
Research to guide your next move.
Work with a specialist research agent built to help design AI-native organisations—from operating models and workflows to agents and tooling.
Combine the library’s knowledge with broader specialist research to compare approaches, uncover insights and answer your business question.
Give your team sourced recommendations and trade-offs to shape investment decisions, designs and delivery plans.
From your question to a better-informed plan.
01
Frame your research brief
Start with what you want to build or decide. Define the question, your business context and the constraints the research needs to address.
02
Investigate what matters
The research agent draws on specialist knowledge, explores gaps and weighs different approaches—explaining what could work, why it matters and the trade-offs for your AI-native company.
03
Equip your agent to deliver
Findings, sources and recommendations are packaged for your agent to combine with your own data and context—building tailored briefs, designs and plans for your team to review.
Live View
See what’s in the library, explore research highlights and browse the sources.
AIKL / COLLECTION EXPLORERLIVE SNAPSHOT
Inside AIKLThe collection, in view.
LIVE SNAPSHOT
CURATED SOURCE RECORDS3,913
A research base for building and using AI.
3,913
sources in the library
The daily AI brief
Latest available edition: 2026-11-12. Today's update is delayed.
Edition
RESEARCH SPOTLIGHT 01 / 03
Publication date unknown · research
TRIAD finds multimodal retrieval struggles with complete, evolving requests
WHY IT’S WORTH YOUR TIME
It separates similarity ranking from satisfying hierarchy, exclusions, preserved conditions, and revised dialogue state.
TRIAD evaluates 17 systems across six multimodal retrieval subsets, each containing 1,000 queries and 1,000 candidates, using multi-positive mAP. Performance generally declines with hierarchical distance and negative constraints; multi-turn retrieval scores lower in 39 of 40 comparisons with paired final intent, though candidate pools highly overlap without being identical. Evaluations are zero-shot and descriptive across public checkpoints; downstream agentic task completion was not tested.
Get our daily brief automatically via RSS or your own agent.
Use with your agent
Add Scaled Memory to your brief.
Copy and paste this into your agent:
Include research from https://www.scaledmemory.com/accelerators/catalogue/ai-knowledge-library in my daily brief. Read the daily brief and source feed linked on that page, using WebMCP if available. Include why the featured stories matter, the original source links and edition date. Flag stale or incomplete content.