AI for Information Architects
For twenty years this profession explained why structure matters while the room checked its phones. AI just made the room look up.
The role today
Information architects design how an organisation’s knowledge is organised: taxonomies, metadata, site structures, findability. It has been chronically undervalued work — invisible when done well, blamed when done badly — and AI has changed its status almost overnight, because every AI initiative now depends on exactly what IAs build.
How AI changes this role
AI helps you with
- First-draft taxonomies and term sets from content samples
- Content audits at a scale humans can’t sustain
- Metadata suggestions across large libraries
- Structure documentation and diagrams
You provide
- Designing for how this organisation actually thinks
- The trade-offs: depth versus simplicity, control versus adoption
- Winning agreement across departments that disagree
- Knowing when the model is good enough to ship
This is The Knowledge Worker AI Framework™ applied to this role — AI takes the production, you keep the bookends.
This one is personal — it’s my profession. I’ve spent two decades saying metadata matters, governance matters, ownership matters, mostly to polite nods. The AI era is the first time executives have asked me about taxonomy unprompted. Nothing about the work changed. The stakes did.
Where AI genuinely helps
Audits that used to take months
AI can classify, cluster and flag duplication across thousands of documents — the discovery phase of an IA engagement compresses dramatically, leaving more time for the design phase.
Taxonomy drafting
Generate candidate term sets from real content, then do the human work: testing them against how people search, argue and label in practice.
The AI-readiness assessment
A new, urgent service line: evaluating whether an organisation’s information can support AI — structure, currency, permissions, ownership. IAs are the natural owners of it.
Living documentation
Architecture decisions, models and standards maintained as structured, queryable content rather than a slide deck nobody reopens.
A day with AI
An organisation wants Copilot; you’re asked whether their content can support it. AI-assisted audit across the main libraries by lunch: duplication clusters, orphaned sites, a decade of ‘keep everything’. The afternoon is the real work — presenting what the mess will do to their AI answers, in language a steering committee funds. There is no AI without IA, and for once, nobody in the room needs convincing.
The risks
The profession’s risk has inverted: it’s no longer being ignored, it’s being asked to bless AI rollouts under time pressure. Don’t certify readiness you haven’t verified. Be careful with AI-generated taxonomies — they optimise for linguistic similarity, not organisational meaning, and the difference is the whole job. And guard scope: ‘AI readiness’ can quietly become ‘fix everything by Thursday’.
Where the role is heading
This is the profession AI vindicated. Expect the titles to shift — AI Readiness Architect, Knowledge Systems Designer — while the substance stays what it has always been. IAs who can run the executive conversation, not just the modelling, will set their own terms for the next decade.
Your learning path
Start with the cards below, in order — each one is a five-minute read built for a task this role repeats. The full K-Series (ten foundations, ten workflow cards) lives on the Knowledge Worker Playbooks hub.
- K-01 · What Is a Knowledge Worker? — the production/judgement split your profession explains to everyone else
- K-05 · How to Spot AI Opportunities — the method behind the readiness assessments you’ll be asked for
- K-10 · Choosing the Right AI Tool — reach versus reasoning — the architecture question in tool form
- K-15 · AI for Workplace Research — AI-assisted audits and comparisons, verified
- K-07 · Verifying AI Output — the habit that keeps assessments defensible
Is Information Architecture still relevant in the AI era?
It has never been more relevant. AI systems answer from whatever structure exists — organised information makes them useful, chaos makes them confidently wrong. Every AI initiative is now downstream of IA.
Can AI design a taxonomy?
It can draft one from content patterns. It can’t know how your organisation thinks, argues and labels — and a taxonomy that ignores that gets abandoned. Generation is fast; agreement is the craft.
What’s the emerging role for IAs?
AI readiness. Someone has to assess whether an organisation’s information can safely support AI — structure, currency, permissions, ownership. That assessment is Information Architecture wearing its new importance.